Amhara Agricultural Research Institute

Livestock Research Directorate

Proceedings of the 12th Annual Regional Conference on

Completed Livestock Research Activities

May 13 - 16, 2019, Bahir Dar, Ethiopia

 Amhara Agricultural Research Institute
Livestock Research Directorate

Proceedings of the 12th Annual Regional Conference on

Completed Livestock Research Activities

May 13 - 16, 2019, Bahir Dar, Ethiopia

Editors:

Mesfin Lakew
Likawent Yeheyis
Assemu Tesfa
Amhara Agricultural Research Institute
Livestock Research Directorate

P.O. Box 527

Tel: +251 - 582 - 20 64 00
Fax: +251 -582 - 26 60 77
Website: www.ARARI.ORG
E-mail: [email protected]

Published June, 2020

Copyright © 2020 by Amhara Agricultural Research Institute (ARARI), Livestock Research

Directorate, Bahir Dar, Ethiopia

The Views and interpretations expressed in this proceeding are entirely those of the authors and
should not be attributed in any manner to Amhara Agricultural Research Institute, Livestock
Research Directorate.

ARARI encourages fair use of this material. Proper citation is requested.

Correct citation: Mesfin Lakew, Likawent Yeheyis and Assemu Tesfa (Eds.). 2020.
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research
Activities May 13 - 16, 2019, Amhara Agricultural Research Institute, Bahir Dar, Ethiopia.
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table of Contents

Comparative Advantages of Cervical Insemination over Natural Mating on Production of

Crossbred Lambs for Export Market .............................................................................................. 1

Performance of Central Highland Goat under Community Based Breed Improvement

Program in Gumara-Maksegnit Watershed .................................................................................. 11

Participatory Identification of Breeding Objective Traits and Optimizing Community Based

Breeding Program for Simien Sheep in the Highlands of North Gondar ..................................... 25

Growth and Milk Production Performance of Abergelle Goat under the Community Based
Breeding Program at Ziqula District, Northern Ethiopia .............................................................. 36

Comparative Evaluation of Lowland Sheep Breeds for their Growth, Carcass Yield and
Quality under Graded Levels of Concentrate Supplementation ................................................... 50

Effect of Pigeon Pea Dried Leaves and Concentrate Mix Supplementation on the Growth,
Feed Utilization, Carcass Yield and Market Weight of Central Highland Goats ......................... 67

Early Finishing of Grazing Crossbred Sheep under Graded Level of Concentrate

Supplementation ........................................................................................................................... 80

Milk Yield and Composition of Fogera Cows Fed with Napier Grass and Concentrate Feed at
Andassa Livestock Research Center ............................................................................................. 89

Examining Preservation and Feeding Effects of Brewery Spent Grain on Dairy Cattle at
Gondar Town, Amhara Region, Ethiopia ................................................................................... 102

Effect of Substitution of Noug Seed Cake with Pigeon Pea Dried Leaves on Feed Intake and
Weight Gain of Old Oxen Fed a Basal Diet of Teff Straw ......................................................... 120

Feed Resources, Nutritive Value and Alternative Balanced Formulated Diets for Lactating
Dairy Cows at Andassa Livestock Research Center, Ethiopia ................................................... 130

Yield and Yield Related Performances of Cowpea (Vigna Unguiculata L.) Varieties Tested
with and without Fertilizer under Irrigation in Central Gondar, Ethiopia .................................. 151

Evaluation of Cowpea (Vigna Unguiculata L.) Varieties for Yield and Yield Related
Attributes with and without Fertilizer under Rainfed Conditions at Central Gondar Zone,
Ethiopia ....................................................................................................................................... 164

Effect of Variety and Cutting Interval on Biomass Yield and Quality of Alfalfa (Medicago
Sativa L.) under Irrigation in Central Gondar Zone, Ethiopia .................................................... 174

Evaluation of Biomass Yield and Nutritional Composition of Soybean (Glycine Max (L.)
Merrill) Varieties Grown in Lowland Areas of Eastern Amhara, Ethiopia ................................ 183

I
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Participatory On-Farm Evaluation of Different Forage Species in Bure and Guagusa

Shikudad Districts of Western Amhara Region, Ethiopia .......................................................... 198

Evaluation of Anthelmintic Drug Efficacy and Risk Factors for Anthelmintic Drug
Resistance in Ruminants in Selected Districts of Western Amhara Region ............................... 213

Seroprevalence of Border Disease in Recently Introduced Dorper Sheep Flocks at Debre

Birhan Agricultural Research Center, Ethiopia .......................................................................... 224

Seroprevalence and Associated Risk Factors of Maedi-Visna Virus in Sheep Population of

Selected Areas of Eastern Amhara ............................................................................................. 231

The Effect of Strategic Deworming Programs on the Prevalence of Gastrointestinal

Strongyles and Fasciola in Sheep at Selected Districts of North Shewa Zone, Ethiopia ........... 248

Retrospective Analysis of Clinical Cases of Caprine and Ovine in Sirinka Agricultural

Research Centre, Amhara Region ............................................................................................... 260

A Case Control Study of Infectious and Non-Infectious Causes of Abortion in sheep and
Goats of Abergelle and Ziquala Districts of Amhara Region, Northern Ethiopia ...................... 276

Evaluation of the Propagation Techniques of a Native Becium Grandiflorum in Wag-Himra

Zone, Ethiopia ............................................................................................................................. 290

Evaluation and Characterization of Physico-Chemical Quality Parameters of Stingless Bee

(Apidae Meliponini) Honey in Amhara Region ......................................................................... 299

Assessment of the Current Trophic Status of Lake Tana, Ethiopia ............................................ 319

Recent Trends in the Physico-Chemical Characteristics of Lake Tana, Ethiopia ...................... 332

Frame and Catch Assessment Survey of Lake Tana Fisheries, Ethiopia .................................... 348

Controlling Overpopulation of Nile Tilapia (Oreocromis Niloticus) through Polyculture

System ......................................................................................................................................... 358

Diversity, Abundance of Fishes and some Physico-Chemical Characteristics in Lake Tirba,

Awi Zone, Ethiopia ..................................................................................................................... 365

Participatory Evaluation of some Fish Drier Methods in the Surrounding of Tekeze

Reservoir, Wag-Himra Zone, Ethiopia ....................................................................................... 377

II
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Preface

The livestock research directorate in Amhara Agricultural Research Institute (ARARI) is one
of the six research directorates in the institute. The directorate is conducting livestock
research activities to solve the major livestock production constraints in the region. The
research activities have both basic and applied nature with the later having the lion share of
all activities. Similar to other institutions in the country the directorate has its own GTP II
target and the research activities are geared towards achieving that goal. However, preparing
a research proposal, conducting research, and finalizing an experiment is not an end in any
research process. The collected data and the analysis result has to be written in the form of a
scientific paper and has to be published. It is through publishing that the findings of any
research activity can be communicated to the end-user. Unless a result is published the
energy and money invested to conduct the experiment will be lost. The contents of this
proceeding are papers presented in our annual regional completed research activities review
forum. In this proceeding, several research results that are very relevant to end-users are
compiled in the area of animal production, animal breeding, feeds and nutrition, animal
health, apiculture, poultry and fishery. I hope the research results published in this proceeding
will contribute to the development of the livestock sector in the region in particular and in the
country at large. The papers in this proceeding will also have an impact on modernizing the
traditional livestock production system in our region. In addition, the contents of this
proceeding will also be an important reference material for researchers, university students
and instructors, investors and development workers. Finally, I would like to thank those
researchers and their assistants who contributed and shared their research results to be
published in this proceeding. My special thanks also go to the reviewers for their unreserved
assistance to review the articles and the editors who edited the proceeding. In addition, I
would like to thank ARARI for covering the cost associated with the reviewing of the papers,
editing of the proceeding and printing. On this occasion, I would like to remind our
researchers to publish their research findings in time.

Likawent Yeheyis (PhD)

Director, Livestock Research Directorate

Amhara Agricultural Research Institute
Bahir Dar, Ethiopia

III
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Comparative Advantages of Cervical Insemination over Natural Mating on Production

of Crossbred Lambs for Export Market

Shanbel Besufekad, Asfaw Bisrat, Chekol Demis and Aschalew Abebe

Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia

ABSTRACT

The study was conducted at Debre Birhan Agricultural Research Center to evaluate the
success of fresh semen cervical insemination following estrus synchronization in Washera
ewes. A total of 32 Washera ewes were selected from this center. All experimental animals
were synchronized using a single injection of 1 ml Enzaprost® intramuscularly at an
unknown stage of the estrous cycle. Then ewes were divided into Dorper and Awassi fresh
semen cervical insemination. Fixed time cervical insemination was performed in estrous
ewe’s 48-51 hrs following Enzaprost® injection with 0.25 ml fresh diluted semen. Out of 32
ewes synchronized 22 (68.75%) of ewes exhibited overt signs of oestrous within 48 hrs. of
hormone administration. Ewe’s BSC wasn’t found to be a significant factor determining
oestrous response to hormone treatment. The overall pregnancy, lambing and fecundity rates
for cervical insemination and natural mating were 63.64, 77.27, 121.43% and 57.46, 61.57,
107.14, respectively. There was no significant difference in pregnancy, lambing and
fecundity rates among ewes mated with fresh semen cervical insemination and controlled ram
mating (p<0.05). The pregnancy, lambing, and fecundity rates for Awassi and Dorper semen
were 58.3, 75.0, 128.57% and 70.0, 80.0, 114.29%, respectively. No significant differences
were observed in pregnancy, lambing and fecundity rates among the genotypes (p˂0.05). On
the other hand under controlled natural mating, ewes mated during the short rain season had
significantly higher (p<0.05) pregnancy and lambing rates as compared to ewes mated during
the main rain season. The current work strongly indicated that AI following oestrous
synchronization could be thought possible as a strategy in the production of crossbred lambs
for the export market. Fresh semen AI has a tremendous impact on the terminal crossing,
allows us to import fewer exotic genotypes and produces a large number of crossbred lambs
from an imported breed within a few years. The further success rate of AI with chilled ram
semen in Washera sheep needs to be investigated.

Keywords: Artificial insemination, Commercial lamb production, Enzaprost, Washera

1
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Ethiopia is a home for a diverse small ruminant population, numbering 30.7 million sheep
and 30.2 million goat heads (CSA, 2017). They account for 40% of cash income earned by
farm households, 19% of the total value of subsistence food derived from all livestock
production, and 25% of total domestic meat consumption (Adane and Girma, 2008). Despite
the large numbers and the multiple roles, small ruminant productivity in Ethiopia is
considered low as compared to productivity levels in developed livestock industries of the
world (Solomon et al., 2010). This low level of productivity could be attributed to several
factors; chief among them is the low genetic potential of the animals (Girma et al., 2010).

Ethiopia meat exporters are not competent in the world market due to lack of supply from the
traditional sheep husbandry that meets the export market standard. Since 2012, Debre Birhan
agricultural research center (DBARC) has conducted research to evaluate the combining
ability of local ewes (Washera and Bonga sheep) with Awassi rams in growth and carcass
traits. The experiment was designed to utilize the fast growth and larger carcass of the exotic
breed to satisfy the required export market standard. The study revealed that crossbred lambs
that were finished at an early age (at seven months of age) could satisfy the required export
market standard (DBARC unpublished data).

However, reproduction data recorded in DBARC from 2013 to 2017, under terminal crossing
of Washera ewes using Awassi showed that lambing occurred in a scattered manner (within
140 days of the commencement of lambing). This has a great impact on the effectiveness of
terminal crossing and management of lambs to finish at an early age. However, artificial
insemination (AI) following oestrous synchronization could be a solution for achieving
planned and concentrated lambing without significantly affecting pregnancy. In addition, AI
could be a solution to import fewer exotic genotypes and allow us to produce a large number
of crossbred lambs from imported exotic sheep breeds. Therefore, the objective of this study
was to evaluate the success of fresh semen cervical insemination following estrus
synchronization in Washera sheep.

2
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Location of Study Area

DBARC is located 120 km North-east of Addis Ababa at an altitude of 2,765 meter above sea
level and at a latitude of 09°36′23′′N and longitude of 39°39′10′′E. Debre Birhan is
characterized by a bimodal rainfall pattern, where the main rainy season is from June –
September accounting for 75% of the total rainfall, an erratic unreliable short rainy season is
February/March - April/May and a dry season October - January. Based on the
meteorological data obtained from DBARC, the average annual rainfall is 923 mm. The mean
annual maximum temperature is 19.87°C occurs between March and June, while the mean
annual minimum temperature is 6.59°C occurs between October and January.

Experimental Animals and their Management

A total of 32 Washera ewes were used for the AI experiment. The ewe’s age ranges from 4 to
6 years and that gave at least one birth, which their body condition score (BCS) of 2.5 and
above and ewes not suckling were used in this trial. Whereas, in a natural controlled mating
system all ewes were sired by pure Awassi ram in a group of 26 - 30 ewes per ram and
mating lasted on average between 51 and 90 days. All experimental animals were kept
indoors at night and allowed to graze on natural pasture for 6 hrs. a day. In addition, the
experimental animals supplemented with 200g head/day mixed commercial concentrate
consists of 33% noug (Guizotia abyssinica) cake, 65.5% wheat bran, 1% limestone and 0.5%
salt. The animals had free access to fresh water twice a day. As a routine flock health
management practice of the research center, the experimental animals were drenched against
internal parasites (Rafoxanide at 1ml/4kg body weight, Chanelle pharmaceuticals
manufacturing Ltd., Ireland and Tetraclozash-900® at 1 bolus/30kg body weight, Ashish life
Science Pvt. Ltd., India) and were vaccinated against Ovine Pasteurellosis, Peste des Petitis
Ruminants (PPR), Sheep and Goat Pox, Blackleg and Anthrax (National veterinary institute,
Debre Zeyit, Ethiopia) in the different seasons of the year before the beginning of the
experiment. Moreover, the experimental animals were sprayed against ectoparasites (Diazinol
60% E.C at 1ml/1Lit. of clean water, Kafr EL Zayat pesticides and chemicals CO., Egypt) in
July 2017.

3
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Oestrous Synchronization

All experimental ewes received a single injection of 5 mg of the PGF2α analogue dinoprost
(1ml Enzaprost®; CEVA laboratories, Libourne, France) intramuscularly at an unknown
stage of the estrous cycle. Then ewes were subdivided into two groups randomly by blocking
the animals for BCS and allocated to Awassi and Dorper fresh semen cervical insemination.
Ewes were observed for oestrous at 4 hrs intervals for 48 hrs. Ewes were kept indoor the
whole day during oestrous identification. Ewes at heat were identified using proven ram fitted
with an apron. Standing to be mounted was the key sign used to determine estrus response.

Semen Collection and Artificial Insemination

Semen was collected from Awassi and Dorper rams trained to serve an AV, after allowing the
rams to mount in the estrus ewes. The collected semen was immediately evaluated for
volume, appearance (color and contamination), sperm cell concentration and mass motility.
The volume of semen was measured with a calibrated collection glass and sperm cell
concentration was evaluated using Accu-Read IMV Technologies SA, 232
Spectrophotometer. Sperm mass motility was estimated subjectively by using a phase
contrast microscope (Scope Technology Scope Photo 3.0.12). The qualified semen to be used
for AI was diluted with OviXcell extender (preservation medium for ovine semen, IMV
Technologies, France). Fixed time cervical insemination had performed in the estrus ewes
within 48-51 hrs following Enzaprost® ejection with 0.25 ml of diluted fresh semen
(approximately 150x106 sperm/straw and its mass motility score 3 and above) using a
speculum equipped with a light source and an insemination gun, slowly releasing the semen
into the first fold of the cervix.

Statistical Analysis

The data were analyzed using Statistical Package for the Social Sciences (IBM SPSS version
20). Effects were considered to be significant when the level of probability was 5% or less.

Oestrous response, pregnancy, lambing, and fecundity rates for artificial insemination were
calculated by using the following formula:

ewes show oestrous

Oestrous response = x 100
ewes administered hormone

4
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

number of ewes pregnant

Pregnancy rate = x 100
number of ewes inseminated

number of lambs born alive

Lambing rate = x 100
number of ewes inseminated

number of lambs born alive

Fecundity rate = x 100
number of pregnant ewes

Pregnancy, lambing, and fecundity rates for controlled natural mating were calculated by
using the following formula:

number of ewes pregnant

Pregnancy rate = x 100
number of ewes sired by rams
number of lambs born alive
Lambing rate = x 100
number of ewes sired by rams

number of lambs born alive

Fecundity rate = x 100
number of pregnant ewes

RESULTS AND DISCUSSION

Oestrous Response

As indicated in Table 1, out of 32 ewes treated with a single injection of PGF2α analogue
dinoprost (1ml Enzaprost®), 22 (68.75%) ewes exhibited overt signs of oestrous within 48 hrs. of
hormone administration. The current result revealed that oestrous could be effectively
synchronized using a single injection of PGF2α analogue dinoprost (1ml Enzaprost®) for
achieving concentrated lambing and production of a large number of crossbred lambs using AI
within a few years. Oestrous response attained in this trial was comparable to values (65%)
reported by Zeleke et al. (2015) in Menz sheep synchronized with single injection of PGF2α (2
ml Lutalyse® and 1 ml Synchromate®). Findings of the current study were, however, higher than
those obtained by Solomom et al. (2016) who reported 57.5% oestrous response in Tigray
highland sheep synchronized with a single injection of PGF2α (2 ml lutalyse®) within 96 hrs post
hormone administration. However, it was lower than the values reported by Solomon et al. (2016)
who recorded an oestrous response of 80% in Washera sheep treated with single injections of
PGF2α (2 ml Lutylase®). The differences in oestrous responses reported in different studies
might be due to differences in breed, season and overall management conditions of the animals.
In the current study, ewe’s BSC was not found to be a significant factor determining oestrous
response to hormone treatment.

5
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Estrus response of Washera ewes to estrus synchronization protocol

Parameter n Oestrous response (%) X2 P value
Body condition 1.663 0.265
Between 2.5 and 3.0 17 58.8
3.5 and above 15 80.0
Overall 32 68.75

This may be due to ewes included in this experiment had a BCS of 2.5 and above. Ewes BCS
of 2.5 and above are recommended BCS for achieving good reproduction in sheep. The
Santoralia et al. (2011) review of factors affecting the efficiency of synchronization indicated
that high BCS has been associated with an increase of ovulation, with recommended BCS of
2.5 - 3.0 and a score of <2 resulting in the lowest pregnancy rates in sheep.

Pregnancy, Lambing and Fecundity Rates

To evaluate the comparative advantages of cervical insemination following oestrous

synchronization over natural mating on the production of crossbred lambs, five years
reproduction data (n = 268) recorded under controlled natural mating at DBARC were used
as a comparison. The overall pregnancy, lambing and fecundity rates recorded in the natural
mating were 57.46, 61.57 and 107.14% respectively, while the overall pregnancy, lambing
and fecundity rates recorded following oestrous synchronization and AI were 63.64, 77.27
and 121.43%, respectively. There was no significant difference in pregnancy, lambing and
fecundity rates among ewes mated with fresh semen cervical insemination and controlled ram
mating (p<0.05). Moreover, pregnancy, lambing and fecundity rates for Awassi x Washera
and Dorper x Washera were 58.3, 75.0, 128.57% and 70.0, 80.0, 114.29% respectively. No
significant differences were observed in pregnancy, lambing and fecundity rates among the
genotypes (p˂0.05). On the other hand under controlled natural mating, ewes mated during
the short rainy season had significantly higher (p<0.05) pregnancy and lambing rates as
compared to ewes mated during the main rainy season. This may be due to the difference in
feeding and overall management conditions of the animals. During the main rainy season,
because of a high level of grazing pasture infestation with fasciola and nematodes, all
experimental animals were kept indoors all day and night. Experimental Animals were
allowed to feed hay prepared from natural pasture and supplemented with 200 g/head mixed
commercial concentrate. On the other hand, during the short rainy season, all experimental
animals were allowed to graze on green pasture for 6 hrs a day and provide hay prepared
from natural pasture at night.

6
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In addition, the experimental animals were supplemented with 200 g/head mixed commercial
concentrate. According to Liu et al. (2014), vitamin E is predominantly found in green feed,
while dried pasture and most grains had below the recommended requirements to attain good
reproduction in sheep. Deficiency of vitamin E had a significant effect on sheep reproductive
efficiency in both males and females and the survival of lambs (El-Shahat et al., 2011).

According to Allaoui et al. (2014) in sheep, fertility rates ranges from 60 and 100% qualified
as an acceptable performance, in this regard the present study revealed that the
implementation of fresh semen cervical insemination in Awassi x Washera terminal crossing
could be a solution for achieving concentrated lambing and also allows us to produce a large
number of crossbred lambs from the imported breed. The overall pregnancy and lambing
rates achieved in the current study under fresh semen cervical insemination were comparable
to the values reported by Allaoui et al. (2014); Kumar et al. (2015); Pervag et al. (2010);
Najafi et al. (2014). However, it was lower than the values obtained by Zeleke et al. (2005)
who reported pregnancy rates of 70.6 and 70.4% in Dorper sheep synchronized with MAP
and FGA sponges, respectively. Similarly, Fornazari et al. (2018) reported pregnancy rates of
76.5% in Assaf sheep synchronized with progestagen sponges.

However, the current result was higher than the values obtained by Olivera-muzante et al.
(2011) who reported a pregnancy rate of 51% synchronized with two doses of PGF2α 7 days
apart and inseminated cervically with fresh semen. The pregnancy rate under the fresh semen
cervical insemination experiment was not significantly influenced by the BCS of ewes.
However, ewes BCS was found to be a significant factor in determining lambing and
fecundity rates (Table 2). A study in Suffolk ewes showed that BCS of ewes did not
significantly affect pregnancy rate (Fukui et al., 2010). However, it is contrary to the findings
of Solomon et al. (2016) who reported the significant implication of BCS on fertility rates of
ewes. However, in natural mating, the effect of ewe’s BCS on pregnancy, lambing, and
fecundity rates was not considered because in the case of natural mating ewes were allocated
to rams based on their live body weight without scoring their BCS. However, as a routine
flock management of the research center ewes that had poor BCS were not included in the
mating group.

7
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Pregnancy, lambing and fecundity rates (%) in Washera sheep

Factors N Pregnancy rates Lambing rates Fecundity rates
NS NS NS
Natural mating 268 57.46 61.57 107.14
Artificial insemination 22 63.64 77.27 121.43
Genotype NS NS NS
Awassi x Washera 12 58.3 75.0 128.57
Dorper x Washera 10 70.0 80.0 114.29
Mating season for NM * * NS
Short rain 111 72.07b 80.18b 111.25
Long rain 157 45.22a 46.50a 102.85
Body condition NS * *
Between 2.5 and 3.0 10 70.0 110.0b 157.14b
3.5 and above 12 58.3 50.0a 85.71a
a, b
on the same column, numbers bearing the same superscript are not statistically different
at p ˂ .05, NS: not significant. N: number, NM: natural mating

CONCLUSIONS AND RECOMMENDATION

AI could be thought possible as a strategy in the production of crossbred lambs for the export
market. Fresh semen AI has a tremendous impact on the terminal crossing, allows us to
import fewer exotic genotypes and produces a large number of crossbred lambs from
imported breeds within a few years. The further success rate of AI with chilled ram semen in
Washera sheep needs to be investigated.

ACKNOWLEDGEMENTS

I am indebted to express my deepest respect and heartfelt thanks to the staff of the livestock
department of DBARC for their ultimate support in facilitating research animals and
laboratory equipment. I am very grateful to my partner Mekibeb Worku and Deribew Bekele
for their unreserved support during research works.

REFERENCES

Adane Hirpa and Girma Abebe. 2008. Economic Significance of Sheep and Goats. In Yami
A., Merkel R.C. (eds), Sheep and goat production handbook for Ethiopia. Ethiopian
Sheep and Goat Productivity Improvement Program (ESGPIP). Pp. 1–4.
Allaoui A., Tlidjane M., Safsaf B. and Laghrour W. 2014. Comparative Study between Ovine
Artificial Insemination and Free Mating in Ouled Djellal Breed. Apcbee Procedia, 8
December 2014, 254–259.

8
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CSA (Central Statistical Agency). 2017. Agricultural sample survey 2016/17. Report on
livestock and livestock characteristics (private peasant holdings). Vol. II, Addis Ababa,
Ethiopia. Pp. 13.
El-Shahat K. H. and Abdel Monem U. M. 2011. Effects of dietary supplementation with
vitamin E and/or selenium on metabolic and reproductive performance of Egyptian
Baladi ewes under subtropical conditions. World Applied Sciences Journal 12(9): 1492-
1499.
Fornazari R. R., Mateus Ó., Correia T. M., Quintas H., Maurício R., Conradi A. and Valentim
R. 2018. Estrus synchronization and artificial insemination with fresh and chilled semen
in Assaf ewes. Agricultural Sciences 9 (1): 8-22.
Fukui Y., Kohno H., Okabe K., Katsuki S., Yoshizawa M., Togari T. and Watanabe H. 2010.
Factors affecting the fertility of ewes after intrauterine insemination with frozen-thawed
semen during the non-breeding season. The Journal of Reproduction and Development
56 (4): 460–466.
Girma Abebe, Kannan G. and Goetsch A. L. 2010. Effects of small ruminant species and
origin (highland and lowland) and length of rest and feeding period on harvest
measurements in Ethiopia. African Journal of Agricultural Research 5 (9): 834-847.
Kumar K. P. 2015. Evaluation of estrus synchronization with different doses of Prostaglandin
f2α in Nellore Jodipi ewes. Doctoral dissertation, Sri Venkateswara Veterinary
University, TirupaTI–517 502, AP.
Liu S., Masters D., Ferguson M. and Thompson A. 2014. Vitamin E status and reproduction
in sheep: potential implications for Australian sheep production. Animal Production
Science 54(6): 694-714.
Najafi G., Cedden F., Kohram H.and Akbari Sharif A. 2014. The effects of using artificial
insemination techniques on reproductive performance in Ghezel sheep. International
Journal of Advanced Biological and Biomedical Research 2(12): 2898-2904.
Olivera-Muzante J., Fierro S., López V. and Gil J. 2011. Comparison of prostaglandin-and
progesterone-based protocols for timed artificial insemination in sheep. Theriogenology
75 (7):1232-1238.
Pervage S., Hassan M., Ershaduzzaman M. and Khandoker M. 2010. Preservation of liquid
semen and Artificial Insemination in native sheep. Journal of the Bangladesh
Agricultural University, 7(2). https://doi.org/10.3329/jbau.v7i2.4739.

9
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Santolaria P., Palacin I. and Yániz J. 2011. Management Factors Affecting Fertility in Sheep
In: Artificial Insemination in farm animals. (http://www. intechopen.com/books/
artificial-insemination-in-farm-animals/management-factors-affecting fertility-in-sheep).
Solomon Gizaw, Azage Tegegne, Birhanu Gebremedhin and Hoekstra D. 2010. Sheep and
goat production and marketing systems in Ethiopia: Characteristics and strategies for
improvement. IPMS (Improving Productivity and Market Success) of Ethiopian Farmers
Project Working Paper 23. ILRI (International Livestock Research Institute), Nairobi,
Kenya. Pp. 58.
Solomon Gizaw, Yayneshet Tesfay, Yoseph Mekasha, Zeleke Mekuriaw, Tadesse Gugsa,
Ebro A., Berhanu Gebremedhin, Hoekstra D. and Azage Tegegne. 2016. Hormonal
oestrus synchronization in four sheep breeds in Ethiopia: impacts on genetic
improvement and flock productivity. LIVES Working Paper 25. Nairobi, Kenya:
International Livestock Research Institute (ILRI).
Zeleke Mekuriaw, Greyling J. P. C., Schwalbach L. M. J., Muller T. and Erasmus J. A. 2005.
Effect of progestagen and PMSG on oestrous synchronization and fertility in Dorper
ewes during the transition period. Small Ruminant Research 56(1-3): 47-53.
Zeleke Mekuriaw, Habtemariam Assefa, Azage Tegegne and Dagne Muluneh. 2015. Estrus
response and fertility of Menz and crossbred ewes to single prostaglandin injection
protocol. Tropical animal health and production 48(1): 53-57.

10
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Performance of Central Highland Goat under Community Based Breed Improvement

Program in Gumara-Maksegnit Watershed

Solomon Abegaz1*, Alayu Kidane2*, Samrawit Tsehay2, Aynalem Haile3, Tsegaye Asredie2
and Yosef Amsalu2

1
Woldia University, P.O. Box 400, Woldia, Ethiopia
2
Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia
3
International Center for Agricultural Research in the Dry Areas, P.O. Box 5689, Addis
Ababa, Ethiopia
* The authors contributed equally to this work

ABSTRACT

The study was conducted in Gumara-Maksegnit watershed to evaluate the performances of

Central Highland goats under the community-based goats improvement program from 2012 -
2017. A total of 1362 kids from the base year and the following years were used to evaluate
the performances of goats under the improvement program. General linear model procedures
were employed to analyze the performance of goats in different traits. The overall least
square means birth weight, weaning weight, six months weight, nine months weight and
yearling weight of goat taken was 2.66±0.03kg, 9.54±0.19kg, 13.25±0.22kg, 17.76±1.01kg
and 20.36±0.46kg, respectively. Sex had a significant (P<0.0001) effect only at birth weight.
Birth year significantly affected the birth weight, weaning weight and six months weights.
The weights were increased in consecutive years; birth weight improved from 2.37 to 2.84 kg
(0.47 kg), weaning weight improved from 8.28 to 10.22 kg (1.94 kg) and six months weight
improved from 11.87 to 14.51 kg (2.24 kg). Parity, birth season and birth type had also a
significant effect on birth weight. The kids from first parity, kids born at the dry season and
kids from triple birth had significantly lower birth weight than their counterparts. The overall
litter size of the Central Highland goat was 1.72±0.02. The weight increments from the base
year to the consecutive years found in this study were the good indicators of the genetic
improvement of weight through CBBP. Appropriate management systems should be
emplaced to manage the influence of non-genetic factors.

Keywords: central highland goat, community based breeding program, performance

11
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Goats are mostly kept by smallholders and the rural poor, including women-headed
households. They contribute substantially to the livelihoods of Ethiopian smallholder
households as a source of income, food (meat and milk), and non-food products such as
manure, skins and wool. They also serve as a means of risk mitigation during crop failures,
property security, monetary saving and investment in addition to many other socioeconomic
and cultural functions (Herpa and Adane, 2008; Abegaz et al., 2013).

Demand and prices for sheep and goat meat show an increasing trend owing to urbanization
and increased income in the cities and increased demand from the Gulf countries. However,
the annual meat production from small ruminants is relatively small compared with the
number of heads. The average annual off-take rate and carcass weight per slaughtered animal
for the years 2000 to 2007 were estimated at 32.5% and 10.1 kg, respectively (FAO, 2009)
which is the lowest among sub-Saharan African countries.

Selective pure breeding of the adapted indigenous breeds is the best possible option of
genetic improvement in tropical countries. Indigenous breeds in harsh tropical environmental
conditions have special adaptive features such as tolerance of a wide range of disease and
water scarcity and ability to better utilize the limited and poor quality feed (Baker and Gray,
2004; Kosgey and Okeyo, 2007). The recent approach of establishing community-based
breeding programs is advocated for low input traditional smallholder farming systems
(Sölkner et al., 1998; Kahi et al., 2005; Haile et al., 2009; Wurzinger et al., 2011).

This is because community-based breeding programs take into account the producers needs,
views, decisions, and active participation from inception through to implementation, and their
success is based upon proper consideration of farmers’ breeding objectives, infrastructure,
participation, and ownership (Mueller 1991; Sölkneret al. 1998; Wurzingeret al. 2011). There
are several practical examples of the successful implementation of community-level livestock
breed improvement programs in different parts of the world. Those are fiber improvement for
Llamas in Bolivia (Wurzinger et al., 2008), dairy improvement for goat in Mexico
(Wurzinger, 2013), meat improvement for sheep in Ethiopia (Haile et al., 2012) and meat
improvement for pig in Vietnam (Valle Zairate and Markeman, 2010).

12
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Hence, since 2012 community-based breeding program of goat has been implemented in two
villages of Gumara-maksegnet watershed to increase the productive performance of the local
goat type to increase the economic benefit of the indigenous goat. However, the progress of
the breeding program had not yet been evaluated. Therefore, the objective of this study was
to evaluate the performance of local goats under the community-based breed improvement
program.

MATERIAL AND METHODS

Study Area

Gumara-Maksegnit watershed lies in the Lake Tana basin of the North West Amhara region
in Ethiopia. This catchment drains into the Gumara River, which ultimately reaches Lake
Tana. The Gumara-Maksegnit watershed is found in Gondar Zuria district of the Central
Gondar Administrative zone. It is located between 347500 E and 1383500 N at the upper part
of the watershed and 344000 E and 1371000 N at the outlet. The watershed is located about
45 km southwest of Gondar town. Altitude within the watershed ranges from 1933 to 2852
meters above sea level. The topography of the area ranges from a gentle slope to a sharp steep
slope. The total area of the Gumara-Maksegnit watershed is about 60 km2. The watershed is
inhabited by 1148 households and 4246 individuals with an average family size of four
persons. Settlement in the watershed is scattered and the landholding is characterized as small
and fragmented. About 55% of the total land is cultivable, 23% of the area is covered by
forest and grazing land, 7% is wasteland and 15% of the land is used for settlement. The
livelihood of households in the watershed is dependent on forest, livestock and crop
production.

Community Mobilization

Before the commencement of the actual fieldwork, the community was sensitized about the
objectives, intentions and possible outcomes of the project for their genuine participation
during data collection. Accordingly, community meetings were organized in the village. The
meetings were generally helpful in establishing a mutual understanding with the local people
while avoiding unrealistic expectations. During the meetings, the research center and
participant farmers agreed to participate in the selection of breeding sires/bucks and to cull
the unselected ones to avoid uncontrolled mating in communal grazing land or watering
points.

13
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Agreement on when & how to cull undesirable males and on the effective use of the selected
young males for breeding before they are sold off were reached. After this agreement, the
village level goat improvement program with 56 participants at Dinzaz village starting from
2012 and 51 participants at Chinchaye village starting from 2015 was established. The
participants have been grouped in 12 and 14 buck user groups at Dinzaz and Chinchaye,
respectively based on their proximity and number of breeding does to facilitate easy
management of selected bucks. Five representatives (committee members) from each village
were selected by the participants. The committees' members have been working on ram
selection, ram distribution, and ram management and overall coordination of the community-
based breed improvement program.

Data Collection and Follow-Up

To restrain goats and facilitate the selection of superior sires by farmers, a temporary wooden
crash (holding yard) was constructed at each village. An ID number on plastic ear tags was
given for all goats of the participants. Two trained enumerators were employed at each
village for a day to day follow-up of activities undertaken by the participant farmers.
Moreover, farmers’ perception and other relevant data (productive and reproductive
performance data) have been collected continuously throughout the experimental period.

Bucks Selection and Management

Selections of young bucks were undertaken at the age of six months based on their
performance breeding value. The candidate bucks were collected at the central place and the
final selection was done by the representative farmers. The selected bucks have been
distributed to the organized bucks’ user groups based on 1:15 to 1:20 male to female ratio.
The bucks were rotated between the buck user groups to avoid inbreeding. Undesirable males
were culled before they reached puberty (i.e. before sexual maturity) through castrating,
selling or slaughtering.

Data Analysis for Performance Recording Data

Before analysis, the data was checked for the normality test by using UNIVARIATE
procedure of SAS. The largest and smallest outlier values were filtered out from the data.
Data was analyzed using Statistical Analysis System (SAS) version 9.1.3.

14
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Productive data (birth, three months, six months, nine months and yearling weight) and
reproductive performance (litter size) were analyzed using the GLM procedure of SAS. Sex,
parity, type of birth, year of birth and season of birth of the doe were considered as the fixed
effects in the model. When analysis of variance declared a significant difference, least square
means were separated by using Tukey’s HSD test.

The model for growth performance was:

Yijklmn = µ + Si + YBj + BSk + BTl + Pm + eijklmn

Where,

Yijklmn = the observed growth performance

µ = overall means
Si = the effect of ith sex (male and female)
YBj = the effect of jth year of birth (2012, 2013, 2014, 2015, 2016 and 2017)
BSk = the effect of kth birth season (wet and dry)
BTl = the effect of lth birth type (single and twin)
Pm = the effect of mth parity (1, 2, 3, 4, 5, 6 and ≥7)
eijklmn = random residual errors

The model for litter size was:

Yijklm = µ + Vi + BYj + BSk + Pl + eijklm

Where,

Yijklm = the observed litter size

µ = overall means
Vi = the effect of ith village (Chinchaye and Dinzaz)
BYj = the effect of jth birth year (201, 2013, 2014, 2015, 2016 and 2017)
BSk = the effect of kth birth season (Dry and wet)
Pl = the effect of lth parity (1, 2, 3, 4, 5, 6 and ≥7)
eijklm = random residual errors

15
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Buck Selection

Two rounds of buck selection based on simple sire selection and four rounds of buck
selection based on performance recording (EBV) have been undertaken at Dinzaz village
while one round of buck selection based on simple sire selection and two rounds of buck
selection based on performance recording (EBV) have been undertaken at Chinchaye village.
The number of selected bucks at each round of buck selection is described in Table 1. At
Dinzaz, a total of 27 breeding bucks were selected based on simple sire selection methods
(without recording) while 40 bucks have been selected based on their breeding value (growth
performance) and farmers selection criteria (color and conformation). At Chinchaye village, a
total of 12 breeding bucks were selected based on a simple sire selection method while 19
breeding bucks have been selected based on their breeding value (growth performance) and
farmers’ selection criteria.

Table 1: Number of bucks selected at each round and village

Dinzaz Selection Selected Chinchaye Selection Selected
village Round Bucks village Round Bucks
SSS 1 15 - -
SSS 2 12 - -
EBV 3 12 - -
EBV 4 6 SSS 1 12
EBV 5 10 EBV 2 5
EBV 6 12 EBV 3 14

Cooperative Establishment

At Dinzaz village, “Yegenetua Mariam” goat improvement and fattening cooperative had
legally established and registered with 46 legal members in January 2016. Currently, they
have 24 castrated, 12 intact bucks and around 30,000 Ethiopian Birr at the bank. On the other
hand, participants at Chinchaye village are on their way to legally registering as a cooperative
by taking the experience of Dinzaz cooperative. Currently, they have 10 castrated, 14 intact
bucks and around 5,000 Ethiopian Birr.

16
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Growth Performances

Birth Weight

The weight at different ages (at birth, 3 months, 6 months, 9 months and yearling) of the
Central Highland goats in the study watershed is presented in Table 2. The overall least
square mean of the birth weight of Central Highland kids was 2.66±0.03 kg. The observed
mean birth weight of Central Highland goats in this study is higher than 2.01 kg and 2.32 kg
reported by Belay and Mengstie (2013) and by Tesfaye et al. (2008) for the same breed in
Sekota and Bati, respectively. The higher birth weight observed in this study could be related
to the better availability of feed and the selection program implemented in the area. Zergaw et
al. (2016) reported a very close value of birth weight (2.68 kg) for the same breed around
Ambo.

Male animals had significantly (P<0.0001) higher birth weight than female counterparts. The
result was in line with the finding of previous authors who reported that male kids had
heavier than their female counterparts (Getaw et al., 2019; Belay and Mengstie, 2013). The
heavier weight at birth is attributed to the fact that with the presence of a Y- chromosome and
the products of sry gene activation has sex-specific effects on fetal growth. As a result, males
appear to grow faster than the respective females in the utero (Gardner et al., 2007).

There was an increasing trend in birth weight from 2012 to 2017; kids born in 2016 and 2017
showed significantly higher birth weight than the preceding year of birth (2012 to 2015). The
weight increments from the base year (2012) to the consecutive years were the good
indicators of the genetic improvement of weight through CBBB (Figure 1). Birth weight
improved from 2.37 to 2.84 kg (0.47 kg), weaning weight improved from 8.28 to 10.22 kg
(1.94 kg), and six months of weight improved from 11.87 to 14.51 kg (2.24 kg).

Single born kids were heavier (P<0.001) at birth than those born as twins and triplets. This
difference may be due to the effect of maternal influence. The reduction in birth weight of
lambs for large litter size is related to the fact that as the number of fetuses increases in the
utero, the number of caruncles attached to each fetus decreases, as a result of the feed supply
to the fetuses thus reduces (Gardner et al., 2007). Kids born during the wet season had
heavier (p<0.001) weight than their dry season born contemporaries which is associated with
better feed availability in quantity and quality in the wet season.

17
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Parity of the does affect the birth weight of kids significantly (p<0.0001). Does with higher
parities produced heavier kids. The effect of parity is in agreement with the other authors
(Getaw et al. 2019; Belay and Mengstie, 2013). However, the result is in contrast with the
report of Zergaw et al. (2016) who reported that parity had no significant effect on the birth
weight of kids. The heavier birth weight at late parities can be explained by the heavier does
weight and larger size at later parities (London, 1993; Kassahun, 2000) and physiological
imprint in the uterus during the first pregnancy which will facilitate relatively greater fetal
growth in the subsequent pregnancies (Gardner et al., 2007). In addition, the reproductive
organs of first parity does are also less developed to bear large fetuses in which case the
physiology adjusts the fetal size (Tibbo, 2006). Indeed, younger does are still growing and
then there would have been a competition for nutrients between the fetus and the dam
(Gemeda et al., 2002).

18
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Birth weight and weight at different ages for Central Highland Goat
Birth Weight (Kg) 3 Month Weight (Kg) 6 Month Weight (Kg) 9 Month Weight (Kg) 12 Month Weight (Kg)
Source of Variation N LSM±SE N LSM±SE N LSM±SE N LSM±SE N LSM±SE
Overall 1362 2.66±0.03 1157 9.54±0.19 927 13.25±0.22 604 17.76±1.01 503 20.36±0.46
CV (%) 14.43 20.57 19.21 24.79 16.26
Sex **** NS NS NS NS
Male 670 2.72±0.03 575 9.61±0.20 449 13.38±0.22 306 17.61±1.12 245 20.48±0.48
Female 692 2.60±0.03 582 9.47±0.20 478 13.36±0.22 298 17.92±1.16 258 20.26±0.48
Birth year **** **** **** NS ***
2012 109 2.37±0.04c 35 8.28±0.34d 78 11.87±0.34c 30 15.49±1.73 114 18.71±0.58c
2013 332 2.61±0.03b 288 8.96±0.18cd 241 12.88± 0.28b 159 16.65±1.26 84 19.50±0.63bc
2014 189 2.52±0.03b 171 9.21±0.19cd 134 13.36±0.27ab 108 16.63±1.20 85 20.88±0.55ab
2015 367 2.63±0.02b 285 9.28±0.16c 238 13.72±0.25a 247 18.74±1.13 178 21.36±0.53a
2016 242 2.76±0.03a 288 9.81±0.17b 187 14.02±0.29a 51 18.72±1.55 42 21.40±0.73a
2017 113 2.84±0.04a 86 10.22±0.24a 49 14.51±0.42a 9 20.36±2.93 - -
Birth season *** NS ** NS NS
Dry 769 2.63±0.03 729 9.53±0.20 576 13.05±0.22 373 19.55±1.21 273 20.30±0.50
Wet 593 2.69±0.03 428 9.55±0.20 351 13.69± 0.23 231 15.98±1.14 230 20.44±0.48
Birth type **** * ** NS NS
Single 581 2.86±0.02a 493 9.88±0.18a 403 14.20±0.16a 268 18.68±0.77 236 21.49±0.27
Twin 738 2.71±0.02b 636 9.67±0.13ab 504 13.91±0.14a 328 18.56±0.69 260 20.87±0.23
Triple 42 2.41±0.06c 28 9.06±0.40b 20 11.99±0.57b 8 16.05±2.94 7 18.75±1.33
Parity **** NS NS NS NS
1 238 2.51±0.04b 186 9.36±0.24 154 13.54±0.29 105 17.26±1.36 89 19.77±0.59
2 251 2.66±0.04a 214 9.71±0.22 163 13.60±0.28 104 17.22±1.32 78 20.80± 0.59
3 252 2.68±0.03a 201 9.78±0.22 168 13.63±0.26 116 16.87±1.25 106 20.56±0.55
4 199 2.65±0.04a 175 9.70±0.23 146 13.69±0.27 74 18.16±1.39 59 20.75±0.61
5 162 2.64±0.04a 144 9.38±0.23 111 12.89±0.31 83 17.01±1.35 68 20.23± 0.58
6 121 2.72±0.04a 113 9.26±0.26 94 12.01±0.34 66 17.15±1.49 55 19.81±0.66
≥7 139 2.76±0.04a 123 9.57±0.26 91 13.22±0.33 56 20.67±1.53 48 20.67±0.68
Means with different superscripts within the same column and class are statistically different (at least p < 0.05); NS = non-significant (p>0.05);
* p<0.05; ** p<0.01; *** p<0.001; **** p<0.0001; N= number of observations

19
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Weight at base and fifth year (Kg)

21.40
21.00 18.72

18.00
14.02
18.71
15.00
15.49
12.00 9.81
11.87
9.00
8.28
6.00
2.76
3.00
2.37
0.00
Birth Weight 3 Month Weight 6 Month Weight 9 Month Weight Yearling Weight

Year One Year Five

Figure 1: Weight of kids at the base year and year five (kg)

Weight at Specific Ages

Factors affecting weights of Central Highland goats at different ages (weights at three months,
six months, nine months and at yearling age) are presented in Table 2. Most of the factors
considered are not consistent in weight at specific ages.

The least-squares mean weight (in kg) at three months, six months, nine months and at yearling
age were 9.54±0.19 Kg, 13.25±0.22 Kg, 17.76±1.01 Kg and 20.36±0.46 Kg, respectively. The
analysis of variance (Table 2) for the weights at specific ages showed that the effect of fixed
effects was different at different ages. The three-month weight (9.54±0.19) of the present study
was comparable to the result of Arsi Bale goat (8.4 Kg) (Tatek et al., 2004) central highland goat
(9.42 kg) (Zergaw et al., 2016) and Highland kids (9.02 Kg) (Tucho et al., 2000). However, it
was higher than the value of Abergelle kids (7.4 kg) (Brhanie et al., 2018).

The birth year was a significant source of variation for three months, six months and yearling
weight. There was an increasing trend in 3, 6 and 12 months weight from 2012 to 2017; kids
born in 2016 and 2017 were heavier than the other preceding years. This might be a good
indicator of the positive response of the selection program implemented in the area.

20
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The effect of season of birth was only significant at six months of age and was not consistent.
Kids born in the wet season were heavier at six months (13.69±0.23 vs. 13.05±0.23, p<0.01) than
those born in the dry season which is associated with better feed availability in quantity and
quality in the wet season.

The effect of type of birth on weights of kids was significant at three and six months of weight.
Single and twin born kids weighed heavier (p<0.05, at three and six months) than their triple
born contemporaries. Many scholars reported similar effects of birth type on the growth of kids
and part of this difference can be explained by the carry-over of the heavier weight of singles at
birth and less computation for milk (Gemeda et al., 2002; Tibbo, 2006).

Reproductive Performance

Litter Size

Reproductive efficiency is always considered to be the most important factor in ensuring the
increase in productivity for certain environmental conditions (Hossain et al., 2004). Increased
production efficiency can be obtained from goats since they have high reproductive efficiency
with the potential for increased litter size and shorter generation interval in comparison to other
farm animals (Safari et al., 2007). On the other hand, the goats’ reproductive performance is an
indicator of their adaptation to adverse conditions. The litter sizes of the local goat with different
non-genetic factors are described in Table 4. The overall litter size found in this study was
1.72±0.02. The result obtained in this study was higher than the value (1.16) for Abergelle goat
(Belay et al., 2008) and comparable with Arsi- Bale goat (Tesfaye et al., 2012). Except for
season of birth, all factors considered showed a significant influence on litter size. Does kidded
in Chinchaye village were more prolific than does kidded in Dinzaz village (P<0.001). The
higher litter size at Chinchaya could be associated with better availability of grazing land in the
area. Does kidded during the year 2016 were less prolific than does kidded in other years.

Parity of does had a significant (p<0.001) effect on the litter size. The does at higher parity (3, 4,
5, 6 and ≥7 parity) had significantly higher litter size than the lower parity (parity 1 and 2). The
present result in agreement with the finding of Belay et al. (2013 ) who reported that higher
parity does give higher litter than lower parity ones.

21
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 4: Litter size of Central Highland goat

Litter Size (Number)
N LSM±SE
Overall 1369 1.72±0.02
CV (%) 31.27
Village ***
Chincaye 513 1.77±0.03
Dinzaz 856 1.66±0.02
Birth year **
2012 109 1.68±0.04ab
2013 336 1.62±0.03ab
2014 189 1.75±0.04a
2015 367 1.64±0.03ab
2016 242 1.58±0.03b
2017 126 1.72±0.07ab
Birth Season NS
Dry 774 1.73±0.03
Wet 595 1.71±0.03
Parity ****
1 239 1.28±0.03b
2 251 1.55±0.03b
3 256 1.72±0.03a
4 199 1.93±0.04a
5 164 1.85±0.04a
6 121 1.79±0.05a
≥7 139 1.91±0.04a
Means with different superscripts within the same column and class are statistically different (at least p <
0.05); NS = non-significant (p>0.05); *** p<0.001; **** p<0.0001; N= number of observations

CONCLUSION

• Successful implementation of CBBP in the two villages confirmed the possibilities of

establishing organized selection schemes in the small scale livestock producer
• CBBP improve the weight of goats at early ages (birth, weaning and six months)

o Birth weight improved from 2.37 to 2.84 kg (0.47 kg), weaning weight improved
from 8.28 to 10.22kg (1.94 kg) and six months weight improved from 11.87 to
14.51kg (2.24 kg)
• Environmental factors such as birth season, parity and type of births had significant influence
on the performances of goats

22
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RECOMMENDATIONS

• For wider genetic improvements of goat population in the region, the experiences from pilot
villages should scale-out in other areas
• The existing and new CBBP need further optimization to attain the maximum genetic gain
from the scheme
• Appropriate management systems should be emplaced to manage the influence of non-
genetic factors

REFERENCES

Abegaz, S., Sölkner, J., Gizaw, S., Dessie, T., Haile, A. and Wurzinger, M. 2013. Description of
production systems and morphological characteristics of Abergelle and Western low-land goat
breeds in Ethiopia: implications for community-based breeding programmes. Animal Genetic
Resources Information, 53.69-78.
Belay Derbie and Mengstie Taye. 2013. Evaluation of growth Performance of Abergele Goat under
traditional Management systems In Sekota District, Ethiopia. Pakistan Journal of Biological
Sciences 16(14):692-696
Mengistie Taye, Belay Derbie and Mussie H. Melekot. 2013. Reproductive performance of central
highland goat Under Traditional Management in Sekota District, Ethiopia Asian Journal of
Biological Sciences 6(5).
Birhanie M, Alemayehu K and Mekuriaw G. 2018. Performance evaluation of Abergelle goat under
community based breeding program in selected districts, Northern Ethiopia. Livestock Research
for Rural Development. Volume 30, Article #64.
Duguma, G., Mirkena, T., Haile, A., Iñiguez, L., Okeyo, A. M., Tibbo M., Rischkowsky, B., Sölkner,
J. and Wurzinger, M. 2010. Participatory approaches to investigate breeding objectives of
livestock keepers. Livestock Research for Rural Development, 22(64)
FAO. 2009. Contributions of smallholder farmers and pastoralists to the development, use and
conservation of animal genetic resources; proceedings of the intergovernmental technical
working group on animal genetic resources for food and agriculture, 5th session. 28-30 January
2009, Rome.
Haile A., Wurzinger M., Mueller J.P., Mirkena T., Duguma G., Mwai O., Solkner J. and
Rischkowsky B. 2011. Guidelines for Setting Up Community-Based Sheep Breeding Programs
in Ethiopia. ICARDA Tools and Guidelines No. 1, Aleppo, Syria.
Hirpa, A. and Abebe, G. 2008. Economic significance of sheep and goats. In Yami, A., Merkel, R.C.
(eds), Sheep and goat production handbook for Ethiopia. Ethiopian Sheep and Goat Productivity
Improvement Program (ESGPIP), pp 1–4.

23
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Hulunim Gatew, Halima Hassen, Kefelegn Kebede, Aynalem Haile, Raimundo Nonato Braga Lobo
and Barbara Rischkowsky. 2018. Early growth trend and performance of three Ethiopian goat
ecotypes under smallholder management systems. Agric & Food Secur (2019) 8:4
Hossain S, Sultana N, Alam MR and Rashid M. 2004. Reproductive performance of Black Bengal
goat under semi-intensive management. J Biol Sci 4: 537-541.
Kosgey I.S., Baker R.L., Udo H.M.J. and van Arendonk J.A.M. 2006. Successes and failures of small
ruminant breeding programmes in the tropics: a review. Small Rum.Res., 61, 13– 28.
Kosgey I.S. and Okeyo A.M. 2007. Genetic improvement of small ruminants in low-input,
smallholder production systems: technical and infrastructural issues. Small Rum Res., 70,76–88.
Kosgey, I.S. and Okeyo, AM. 2007. Genetic improvement of small ruminants in low-input,
smallholder production systems: Technical and infrastructural issues. Small Rum Res, 70:76-88.
Legesse, G., Abebe, G., Siegmud-Schultze, M. and Valle-Zárate, A. 2008. Small ruminant
production in two mixed-farming systems of southern Ethiopia: Status and prospects for
improvement. Experimental Agriculture, 44:399-412.
Safari E, Fogarty NM. and Gilmour AR. 2007. A review of genetic parameter estimates for wool,
growth, meat and reproduction traits in sheep. Livest Prod Sci 92: 271-289.
Sölkner, J., Nakimbugwe, H. and Valle-Zárate, A. 1998. Analyses of determinants for success and
failure of village breeding programmes. In: Proceedings of sixth world congress on Genetics
Applied to Livestock Production, Vol. 25 Armidale, NSW, 11-16 January 1998, Australia, 273-
280.
Tesfaye Kebede., Haile, A. and Dadi, H. 2012. Smallholder goat breeding and flock management
practices in the central rift valley of Ethiopia. Trop Anim Health Prod., 44(5): 999-1006.
Tibbo, M. 2006. Productivity and health of indigenous sheep breeds and crossbreds in the Central
Ethiopian highlands, PhD Thesis, Swedish University of Agricultural Sciences, Uppsala,
Sweden.
Valle Zarate A. and Markemann A. 2010. Community-based breeding programmes incorporating
local breeds: Concept, research results and implementation strategy on pigs in Northern
Vietnam. In: Proceedings of the 9th World Congress on Genetics Applied to Livestock
Production. Leipzig (Germany), 1–7 August 2010, p. 845.
Wurzinger M., Escareño L., Pastor F., Salinas H., Iñiguez L. and Sölkner J. 2013. Design and
implementation of a community- based breeding program for dairy goats in Northern Mexico.
Trop. Subtrop. Agroecosyst., 16, 289–296.
Wurzinger M., Sölkner J. and Iñiguez L. 2011. Important aspects and limitations in considering
community-based breeding programs for low-input smallholder livestock systems. Small
Rumin. Res., 98, 170–175.
Wurzinger M., Willam A., Delgado J., Neurnberg M., Valle Zarate A., Stemmer A., Ugarte G. and
Söolkner J. 2008. Design of a village breeding programme for a llama population in the High
Andes of Bolivia. J. Anim. Breed. Gen., 125, 311–319.
Zergaw N, Dessie T and Kebede K. 2016. Growth performance of Woyto-Guji and Central Highland
goat breeds under traditional management system in Ethiopia. Livestock Research for Rural
Development. Volume 28, Article #8.

24
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Participatory Identification of Breeding Objective Traits and Optimizing Community

Based Breeding Program for Simien Sheep in the Highlands of North Gondar

Solomon Abegaz1*, Alayu Kidane2*, Samrawit Tsehay2, Tsegaye Asredie2, Yosef Amsalu2 and
Belete Shimelash2

1
Woldia University, P.O. Box 400, Woldia, Ethiopia
2
Gondar Agricultural Research Centre, P.O. Box 1337, Gondar, Ethiopia
*The authors contributed equally to this work

ABSTRACT

The study was conducted in the highland of North Gondar Administrative Zone of Amhara
region to identify the breeding objectives of sheep owners and to simulate different selection
schemes for implementation of a community-based breeding program. The participatory own
animal phenotypic ranking method was used to identify the breeding objectives. A total of 55
households were visited and asked to choose first best, second best, third best, and inferior ewes
in their flock. Community-based breeding schemes with four different alternatives were
simulated using the SelAction computer program. The alternatives were within the village
selection at 5% and 10% selection proportion and across village selection at 5% and 10%
selection proportion. Various breeding objectives traits were mentioned by sheep owners.
Growth and growth-related traits were identified as the most important traits. The genetic gain in
breeding objective traits and the rate of inbreeding (ΔF) per generation varied depending on the
scenarios. The scenario across village selection at a 5% selection proportion had a higher
response than the other scenarios. Thus community-based breeding programs with few traits
across village selection at 5% of proportion would be feasible in the study areas.

Keywords: breeding objectives, selection scheme, Simien sheep

25
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Ethiopia is a home for an estimated 30.7 million sheep that are distributed throughout the country
(CSA, 2017). The indigenous sheep breeds are very well adapted to the rigorous environmental
conditions and low input systems in these areas utilizing low-quality shrubs and remaining
productive under an escalating disease pressure (Feki, 2013; Kosgey and Okeyo, 2017).

However, sheep production is constrained by many technical and non-technical factors. Among
them, lack of appropriate breeding programs is the one. Therefore there is a critical need to
improve productivity to sustain the supply of food and income to communities that are coming to
rely more and more on the drought tolerant commodity, sheep. Centralized pure-breeding and
crossbreeding programs in the tropics have generally failed. They overlooked the significance of
livestock in the tropics, which transcends economic considerations and enters the social, cultural
and ritual realm (FAO, 2009). The community based breeding program has now emerged as a
promising approach in the tropics where-in farmers actively participate from the designs to the
implementation of the breeding program (Haile et al., 2009; Wurzinger et al., 2011).

As any breeding program, a community-based breeding program involves a precise definition of

breeding objectives and optimizing alternative schemes (Abrham et al., 2018). A breeding
objective defines the direction in which the farmer would like to go towards satisfying his
demand for specific products and services from the animal (Soalkner et al., 2008). Optimization
of the community based breeding programs at different alternative selection schemes to predict
the genetic gain and the economic return is very helpful during implementation. It gives the
chance to adjust the technical, infrastructural and socio-economic issues ahead of the
implementation.

Objectives

• To identify breeding objectives and evaluate different alternative selection schemes of a

community-based breeding program for Simien sheep in the highlands of North Gondar.

26
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIAL AND METHODS

Study Area Description

The study was conducted in two districts (Dabat and Debark) of the North Gondar Zone of
Amhara Region, Ethiopia. The districts were purposely selected based on the availability of
Simien sheep breed and accessibility for monitoring. Dabat and Debark districts are found in the
North-East direction of Gondar town about 800 and 840 km from Addis Ababa, respectively.

The study area is generally situated in the Amhara plateau in the Western Simien Mountains
between 13°11'N and 38°04'E with an altitude range from 2000 - 3900 meter above sea level and
rugged terrain and thus known as "The Roof of Africa" near Ras Dejen Mountain. The mean
annual rainfall of the study area is 1550mm falling into two wet seasons, from February to
March, and July to September (Magin, 2001). Mean annual temperature ranges from a minimum
of -2.5 to 4°C and that of maximum from 11 to 18°C. There is often dry wind during the day;
frosts may occur at night, and snow sometimes settles on the summit of Ras Dejen.

Data Collection

Breeding Objectives Identification

Breeding objectives were identified through participatory phenotypic own animal ranking
methods. Fifty five households were visited. The household members were asked to select their
first, second, third best, and worst breeding ewes among their sheep. The reasons for selection
were inquired and recorded. The life history such as number of birth, number of lambs born, and
the number of lambs weaned was asked and the linear measurements were taken from the ranked
animals.

Design Alternative Selection Scheme

Population Structure

Trait preferences of farmers were used from own-flock animal ranking experiments. As it was
implemented under smallholder farmers, only two traits with high preference by farmers and
easy to measure were considered. The two objective traits were: six months weight and pre-
weaning survival rate.

27
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The design and evaluation of selection schemes relevant to this study were illustrated using
parameters for Simien sheep populations in the mixed farming system in the two districts. A
population with discrete generation was simulated based on the existing population structure in
the districts (500 breeding ewes at Debark and 300 breeding ewes at Dabat). In this study, it was
considered males are used for breeding once at one year of age and culled as yearlings; and
generations in the male population were considered as discrete. In practice, breeding ewes are
maintained for more than one year, thus selection of female animals was ignored. It was assumed
that only one year old does were used and no selection was practiced on ewes.

Selection Scheme

Four different schemes of different proportions of rams were considered. Mating was considered
as random mating because grazing is on communal grazing land. It was assumed that all
breeding rams would be replaced in each generation. The selection of replacement rams was
according to the intensity of selection and within the village and across village selection (Table
1). The selection of replacement females was ignored in this study.

Table 1: Alternative selection schemes used

Scheme 1 Within village selection at 10% selection proportion
Scheme 2 Within village selection at 5% selection proportion
Scheme 3 Across village selection at 10% selection proportion
Scheme 4 Across village selection at 5% selection proportion

Genetic and Phenotypic Parameters

The value of weighted heritability, phenotypic and genotypic standard deviation, economic
values and genetic and phenotypic correlation (Table 2) of breeding objectives was used for
simulation. Since the genetic parameters are lacking for the study population, the parameters
estimated from other tropical sheep breeds were used.

Table 2: Phenotypic variance, relative economic values, heritability along diagonal, genetic
(above diagonal) and phenotypic (below diagonal) correlation
Traits Relative economic weight Variance SMWT Survival
SMWT 75 4.04 0.30 0.01
Survival 25 0.01 0.01 0.09

28
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Statistical Data Analysis

Identification of Breeding Objectives

The qualitative attributes from own ranking experiments were described by descriptive statistics
using Statistical Package for Social Sciences (SPSS version 23, 2007). The quantitative
characteristics collected from the life history and linear measurement were analyzed by the
general linear model procedure of SAS (SAS version 9.4, 2013) fitting the rank as fixed effects
in the model.

Yij = µ + Ri + eij

Where: Yij is the body measurements and other production traits, µ is the over all means, Ri is the
effect of ith rank (i = 1st, 2nd, 3rd and inferior) and eij is the residual error.

Designing of Optimizing Selection Scheme

The response, genetic gain per year, inbreeding level per generation, and economic benefits of
different alternatives were predicted by the deterministic simulation model. The computer
programming SelAction was used for simulating alternative breeding scheme (Rutten et al.,
2002).

Schemes were compared under mass selection (own performance) for pre-weaning survival rate
and six-month weight whereas own performance and BLUP selection (information on relatives)
for litter size to calculate the breeding value of selection candidates. Response to selection was
calculated for individual objective traits. The response was expressed in three ways: in trait units,
in economic units, and as a percentage of the total selection response.

29
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULT AND DISCUSSION

Breeding Objectives Identification

The list of breeding goal traits is indicated in Table 3. Diverse traits were mentioned by the
farmers as preferred traits of breeding ewes. Lamb growth, body conformation, mothering
ability, lambing interval, and color were mentioned as important attributes of ewes in Debark
districts, whereas lamb weight at birth, lambing interval, body conformation, and mothering
ability were considered as important traits of ewes in Dabat. The diverse traits as the selection
criteria for breeding female animals for tropical countries are also well documented in many
research reports (Duguma et al., 2011; Mirkena, 2011; Berhanu et al., 2012; Solomon et al., 2013
and Abraham et al., 2018).

The preference of big body size/body conformation, high lamb weight at birth and fast lamb
growth as the preferred attributes in both study areas are expected when the main purpose of
keeping sheep is for cash source. The animals with big sizes are highly demanded in the market
and fetch good local market prices. Higher preference values of body size for breeding animals
were reported by similar studies in Ethiopia and elsewhere in the tropics (Mbuku et al., 2006;
Kebede et al., 2012; Duguma et al., 2011; Berhanu et al., 2012; Solomon et al., 2013, 2014).

Table 3: List of traits of ewes from own flock ranking method

Dabat Debark Overall
Trait Freq. % Freq. % Freq. %
Twin birth 9 9.28 7 5.04 16 6.78
Body conformation 17 17.53 26 18.71 43 18.22
Color 8 8.25 15 10.79 23 9.75
Disease Resistance 1 1.03 9 6.47 10 4.24
Lamb growth 6 6.19 29 20.86 35 14.83
Lamb weight at birth 21 21.65 8 5.76 29 12.29
Lambing interval 19 19.59 21 15.11 40 16.95
Mothering ability 16 16.49 24 17.27 40 16.95

30
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Next to growth and growth associated traits, reproductive performance such as lambing interval,
19.59% for Debark, and 15.11% for Dabat, was also mentioned as the important traits of
selection breeding ewes. Similar preference to lambing interval also reported for the selection of
breeding ewes in Ethiopia (Mirkena, 2010). Frequent lambing will improve the productivity of
the flock by increasing the number of animals for market and replacement. However, it would be
difficult to improve the lambing interval through a community-based selection program because
of the low heritability of the trait.

The mean of reproductive performance and linear body measurement of the ranked ewes are
depicted in Table 4. Chest girth, body weight, number of laming, number of lambs born and
number of lambs weaned significantly (p<0.001) influenced the owners’ decision. Height at
wither also significantly (p<0.01) influenced owners’ preferences. There was a clear trend for the
different traits of ewes ranked from best to inferior. The best ranked ewes had significantly
higher values than the other ranked groups. When comparing the mean value of the first best
animals with the inferior animals, one can see a clear difference in traits considered. For
example, the difference in chest girth was 6.35 cm; the difference in weight was 6.17 kg and the
difference in the number of lambs weaned was 3.08. This result is comparable with the finding of
Mirkena (2010) who reported that 5.3, 9.0, 7.1 and 3.8 kg of live weight and 2.82, 3.97, 3.75 and
2.96 of lamb survive to weaning for Afar, Bonga, Horro and Menz sheep.

Table 4: Means ± SE value of the traits for different rank group of ewes

Rank
Traits Sig 1st 2nd 3rd Inferior
a ab b
Chest girth (cm) *** 78.07±0.68 75.98±0.68 74.62±0.70 71.72±0.72c
Body weight (Kg) *** 32.11±0.63a 30.91±0.63ab 29.40±0.64b 25.94±0.66c
Number of lambing *** 4.88±0.25a 4.16±0.25ab 3.92±0.26b 2.93±0.26c
a b b
Number of lambs born *** 5.87±0.3 4.48±0.31 4.24±0.31 2.80±0.35c
Number of lambs weaned *** 5.12±0.27 a 4.34±0.28 ab 3.78±0.28 b 2.04±0.29 c
a a ab
Wither height (cm) ** 68.07±0.49 67.48±0.49 66.79±0.50 65.58±0.51b
The means in the same row with different superscripts are significantly different from each other

31
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Attribute Levels Used by Farmers

Sheep breeders in the study areas used qualitative descriptions big, medium and small for growth
related traits like a weight at birth and body size of the adult animal; good, bad and medium for a
trait like mothering ability. There was a general tendency of giving a favorable description
(Figures 1, 2 and 3) like big and good for animals ranked best and vice versa. This finding is
comparable with previous studies (Mirkena, 2010) who reported similar attribute levels used by
farmers for Menz, Horro, and Afar sheep.

Lamb Growth Lamb weight at Birth

18
40 16
14

Frequency
Frequency

30 12
10 Big
20 Bad 8
6 Medium
10 Good 4
2 Small
0 Medium 0
1 2 3 W 1 2 3 W
Rank
Rank

Mothering Ability
30
Frequency

20
Bad
10
Good
0
1 2 3 W Medium

Rank

Figure 1, 2 and 3: Selected attributes and their levels from the own-flock ranking

Design of Alternative Breeding Schemes

The predicted genetic gain (ΔG) in breeding objective traits, accuracy of selection and rate of
inbreeding per generation of different schemes are described in Table 5. There were slight
differences between those parameters among different alternatives. A relatively higher total
genetic gain was predicted at scheme four (across village selection at 5% of selection
proportion).

32
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The higher genetic gain at this scheme was because of the large selection differential due to the
availability of a large number of candidates for selection. Similar genetic gain 0.625 Kg, 0.623
Kg and 0.626 Kg per year of six month weights were predicted for alternative within village
selection at 5% selection proportion at Debark, Dabat and across village selection at 5%
selection proportion. The genetic gain of six month weight predicted from this study was in the
range of predicted genetic gain of six months weight for Menz and Horro sheep breed (Mirkena
et al., 2012) and relatively higher than the predicted six month weight for Gumuz sheep (Dagnew
et al., 2018). The predicted rate of inbreeding was higher than the acceptable range (0.01-0.005).
A high rate of inbreeding per generation (1.04%) was predicted as an alternative within village
selection at 5% of selection proportion at Dabat district. The high inbreeding rate for this scheme
is associated with low breeding ewes population and high selection intensity.

Table 5: Genetic gain, accuracy and rate inbreeding in different alternatives

Within village Across villages
10% proportion 5% proportion 10% 5%
Parameters Dabark Dabat Debark Dabat proportion proportion
Breeding goal 40.06 39.98 46.99 46.819 40.10 47.04
SMWT (kg) 0.533 0.532 0.625 0.623 0.533 0.626
Survival 0.004 0.004 0.004 0.004 0.004 0.004
Accuracy 0.527 0.527 0.526 0.526 0.527 0.526
Rate of inbreeding (%) 0.313 0.518 0.633 1.04 0.196 0.773

CONCLUSION AND RECOMMENDATIONS

• Growth and growth related traits were identified as the most important traits of sheep owners
in the study areas.
• The farmers’ decisions for the ranking of breeding ewes were highly correlated with the
performance of the given animals.
• The ranking experiments approach is useful to elicit objective traits for designing breeding
plans especially under traditional production systems where recording practices have not
been in place. However, given the usually large number of attributes identified in such
studies, only a few priority traits should be included in designing breeding plans in order to
keep them as simple as possible for easy implementation under farmers’ circumstances.

33
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

• Among simulated different scenarios across village selection at 5% proportion had a

relatively higher total response for breeding objectives traits than the other scenarios. Hence,
across village selection at 5 % of proportion would be feasible for the study breed and sites.

REFERENCES

Abegaz S. 2014. Design of community based breeding programs for two indigenous goat breeds
of Ethiopia, PhD thesis, BOKU- University, Vienna, Austria. 115pp.
Abegaz S., Sölkner J., Gizaw S., Dessie T., Haile A. and Wurzinger M. 2013. Description of
production systems and morphological characteristics of Abergelle and Western low-land
goat breeds in Ethiopia: implications for community-based breeding programmes. Animal
Genetic Resources Information, 53, 69-78; ISSN 1014-2339
Bett R.C., Okeyo A.M., Kosgey I.S. Kahi A.K. and Peters K.J. 2012. Evaluation of alternative
selection objectives and schemes for optimisation of village goat improvement programs. .
Livestock Research for Rural Development 24 (1) 2012.
CSA. 2017. Federal Democratic Republic of Ethiopia, Agricultural sample survey, Reports on
livestock and livestock holding characteristics. Statistical Bulletin, Addis Ababa Ethiopia.
Dagnew Y., M. Urge Y. Tadesse and S. Gizaw. 2018. Conservation-based breeding program
design for genetic improvement in Gumz sheep in the western lowlands of Ethiopia. Agric
and Food Secur (2018) 7:34
FAO. 2009. Contributions of smallholder farmers and pastoralists to the development, use and
conservation of animal genetic resources; proceedings of the intergovernmental technical
working group on animal genetic resources for food and agriculture, 5 th session. 28-30
January 2009, Rome. (http://www.fao.org/ag/againfo/programmes/en/genetics/angrvent
2009)
Feki Misbah. 2013. Community Based Characterization of Afar Goat Breed around Aysaita
district of Afar Region. MSc thesis, Jimma University, Ethiopia.
Hagos Abraham, Solomon Gizaw and Mengistu Urge. 2018. Identification of breeding objectives
for Begait goat in western Tigray, North Ethiopia. Tropical Animal Health and Production
https://doi.org/10.1007/s11250-018-1640-5

34
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Haile A., Tibbo M., Duguma G., Mirkena T. Okeyo M., Iniguez L., Wurzinger M. and Sölkner J.
2009. Commercialization of Livestock Agriculture in Ethiopia. In: Tamrat Degefa and
Fekede Feyissa (Eds). Proceedings of the 16th Annual conference of the Ethiopian Society
of Animal Production (ESAP) held in Addis Ababa, Ethiopia, October 8 to10, 2008. Part II
Technical Session.ESAP, Addis Ababa 329 pp.
Kosgey I.S. and Okeyo A.M. 2007. Genetic improvement of small ruminants in low-input,
smallholder production systems: Technical and infrastructural issues. Small Ruminant
Research, 70: 76-88.
Mirkena T., Duguma G., Willam A., Wurzinger M., Haile A., Rischkowsky B., Okeyo A.M.,
Tobbo M. and Sölkner J. 2012. Community-based alternative breeding plans for indigenous
sheep breeds in four agro-ecological zones of Ethiopia. J. Anim.Breed.Genet., 129:244-253.
Mirkena T. 2010. Identifying breeding objectives of smallholders/pastoralists and optimizing
community-based breeding programs for adapted sheep breeds in Ethiopia, PhD thesis,
BOKU- University of Natural Resource and life Sciences, Vienna Austria. 125pp.
SAS (Statistical Analysis System) institute inc. 2009. SAS procedure guide, Version 9.2. Cory,
NC, USA.
Sölkner J., Nakimbugwe H. and Valle-Zárate A. 1998. Analyses of determinants for success
and failure of village breeding programmes. In: Proceedings of sixth world congress on
Genetics Applied to Livestock Production, Vol. 25 Armidale, NSW, 11-16 January 1998,
Australia, 273-280.
Wurzinger M., Sölkner J. and Iniguez L. 2011. Important aspects and limitations in considering
community-based breeding programs for low-input smallholder livestock systems. Small
Ruminant Research, 98:170–175

35
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Growth and Milk Production Performance of Abergelle Goat under the Community Based
Breeding Program at Ziqula District, Northern Ethiopia

Yeshiwas Wale, Bekahagn Wondim, Mulatu Gobeze, Wubneh Akilog and Alemu Demilie

Sekota Dry land Agriculture Research Center, P.O. Box 62, Sekota, Ethiopia

ABSTRACT

Community based breeding program (CBBP) is becoming a new approach which is implemented
nationwide in Ethiopia. The program was also being implemented in Abergelle goat breed since
2014 at Bilaque village in Ziquala district, Wag himra zone. The study aimed to evaluate the
growth and milk yield performance of Abergelle goat under community based breeding program.
Monitoring was done in 33 clustered goat keepers of the selected village. A total of 2410
progenies (1607 for body weight and 803 for milk yield) were used to evaluate the genetic
improvements of the selective breeding. Best bucks were selected at yearly basis based on their
indexed estimated breeding value (EBV). Generalized linear model (GLM) procedure of SAS
(9.0) was used for data analysis. Birth type, year of birth and parity had significant (P<0.05)
effect on pre-weaning growth performances of Abergelle goat. Mean yearling weight of kids had
slightly increased from 12.8±0.11kg in 2014 to 13.7±0.12kg in 2018. Yearling weight had better
correlation (p<0.05) with six month weight (r = 0.3) and nine month weight (r = 0.31). Daily
milk yield was significantly (p<0.05) affected by season of lactation and selection years. Daily
milk yield was increased from 300.31±7.41ml to 389.34±3.66ml during the selection years. It
was concluded that, the program was found to be a compatible strategy which can suit with the
existing breeding practices of farmers as it can allow for the use of elite bucks and removal of
inferiors from the population.

Keywords: Abergelle goat, body weight, community based, milk yield, selection

36
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Goats comprise 5.32% of the total tropical livestock units of Ethiopia, contribute an estimated 12 to
14% of annual meat products, 10.5% of milk production, 47% of the agricultural GDP, 30% of all
domestic meat consumption (Behnke, 2010) and 6% of all animals exported (Solomon et al., 2014).
Together with sheep, goats contribute about 90% of the live animal/meat and 92% of the total skin
export trade value (Solomon et al., 2014). Abergelle goat is one from the Ethiopian goat breeds
which is under rift valley and has an estimated population of over 300,000 (FARM AFRICA, 1996).
It is found along the Tekeze River in Southern Tigray (Tembien and Inderta), Wag-himra, Raya
Azebo, and North Gondar (Simien) and kept by the Agew and Tigray ethnic groups (FARM
AFRICA, 1996).

Goat together with other livestock species contributes about 75% of the cash income sources of
smallholder farmers in the mid and lowland areas of Wag-himra (Bekahegn et al., 2019). According
to Alubel (2015), the average flock size of Abergelle goat was 27 heads per household in the
established Community Based Breeding Programs. Under the traditional management condition, the
breed is characterized by its lower body weight, longer kidding interval, and lower litter size with
better meat quality and temperament in comparison to other indigenous goat breeds of Ethiopia
(Alubel, 2015). As a result of this, a community based selective breeding program has been
developed and implemented for the improvement of growth and production performance of the
breed. The approach is truly participatory, which considers the active involvement of communities in
designing, implementation and ownership of the scheme/system. The study aimed, therefore, to
evaluate the growth and milk yield performance of Abergelle goat under community based breeding
program.

MATERIALS AND METHODS

Description of the Study Area

The program was implemented at Bilaqu village of Ziquala district, located at 12º48´41.39˝N and
38º43´22.02˝E; 775km north east of Addis Ababa. The district has rugged topography characterized
by mountains, steep escarpments and deeply incised valleys. Mixed crop-livestock system with high
priority of livestock production is the major farming practice in the area. The mean annual rainfall of
the area was 250 - 650mm with very short and an erratic distribution from late early July to late
August; mean annual temperature of the area was 25 - 39oC with an altitude of 1308m.a.s.l.

37
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Village Selection

Once the mega project entitled “Harnessing genetic diversity of indigenous goat in Africa'' has
developed by ILRI/Beca since 2013, the program was implemented at Bilaqu village. The village
was selected based on goat production potential, separate herding practice, accessibility of road
and feed resources, and willingness of the farmers to participate in the program. Reports of the
district office of agriculture and rural development were also considered as secondary
information during site selection. About 33 goat keepers were selected and continuous discussion
on the principle and implementation of community-based goat improvement program was done.
Baseline survey study on general household characteristics, purpose of keeping goats, livestock
ownership, flock structure, trait selection criteria’s, performance of goats, management and
breeding practices, and goat production constraints was collected and documented. The breeding
objective identification components for the breed was taken form the report of Solomon et al.
(2014) done at Saziba village of Abergelle district. Based on the report, growth and milk
production traits were taken as priority traits of interest for improvement according to their
respective order (Table 1).

Table 1: List of preferred traits by farmers (own flock ranking method for females and group
animal ranking for males), adopted from (Solomon, 2014)
Sex Traits Percentage (%)
Female Milk Yield 20.47
Drought resistance 14.96
Body size 14.17
Kid growth 11.81
Twining 10.24
Kidding Interval 9.45
Other cumulative traits 18.9
Male Body size 21.03
Color 23.86
Body conformation 10.12
Height 9.09
Fast growth 7.95
Other cumulative traits 27.95

38
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Animal Management and Recording

All animals in the villages were identified using plastic ear tags. In addition, local names were
provided for each individual goat which was helpful during the loss of an ear tag. Three
enumerators were recruited and trained for data recording (pedigree and performance) and follow
up of the breeding program. The baseline information collected includes parity of the doe's using
farmer recall method, age of the doe's based on their dentition and farmers information, total
flock number of the households with each age class category, date of birth, milk yield of the
dam, birth weight and subsequent weights of kids. The enumerators made rounds of visits to the
villages every morning to record and weigh the newborn kids and identify them by ear tags, and
milk data measurement. The kids were also weighed at three, six, nine and twelve months of age.
Milk yield data were recorded up to twelve weeks at a week interval.

Estimation of Breeding Value (EBV) and Selection

During program implementation, the overall procedure was considered the guideline protocol
designed for setting up Community-based Sheep Breeding Programs in Ethiopia by a one-tier
breeding structure. All first birth kids of the population born from unselected previous village
sires were evaluated and selected as first round breeding population, after all, the growth
performance of individuals and milk production performance of their respective dams were
recorded.

Index selection methods were implemented for the first two round selections (on yearling weight
and dam ADMY traits) whereas for the recent three round selections, breeding values of
candidate bucks were estimated by computing correction factors for the non-genetic/fixed factors
like season, birth type, and parity. Yearling bucks from all flocks in the project villages were
evaluated together as cohorts and to be subjected further approval by farmers’ from phenotypic
criterions like color, and body conformation and the selection process were conducted based on
the selection criteria which have been defined earlier by (Solomon et al., 2013).

39
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Buck Usage Modality and Mating

The selected bucks were assigned to buck groups with 1:15 sex ratio following an own flock
mating plan due to presence of larger flock sizes at each household by considering the previous
mating history of bucks to avoid inbreeding (Mueller et al., 2015). Unselected bucks were culled
out through castration and sold before the mating season was approaching. In cases of small
flock numbers faced during the selection process, a group mating system was used. Bucks were
rotated among the individual members based on mapped rotation modality by considering the
grazing corridor of flocks, settlement and previous mating history. At the end of service period
the bucks fattened and sold; and the income used as a revolving fund for the established
cooperative.

Data Analysis

Descriptive statistics and Microsoft Excel program were used to perform simple population
characteristics and graphical presentation of the data. Productive performance (quantitative) data
were analyzed using the GLM procedures of SAS (version 9.0). Birth type, sex, season of birth,
parity and year were fitted as fixed effects for body weight variables whereas season of birth,
parity and year were fitted as fixed effects for milk production traits. Least square mean with
respective standard error was separated using Tukey-Kramer test. Pearson correlation was
employed to investigate associations among weight at different ages. The following models were
used.

For growth parameter:

Yijklmn = µ + Bti + Sj + Sbk +Pl + Ym + eijklmn

Where: Yijklmn is the observed growth performance of the goat, µ is the overall mean, Bti is the
effect of ith birth type (Single, twin and triplet), Sj is the effect of jth sex, Sbk is the effect of
kth birth season (wet and dry), Pl is the effect of lth parity (one to ≥7), Ym is the effect of the mth
year (2013 - 2018) and eijklmn is the random residual error

40
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

For milk parameters:

Yijkl = µ + Sbi + Pj + Yk + eijkl

Where: Yijkl is the observed milk yield, µ is the overall mean, Sbi is the effect of ith birth season
(wet and dry), Pj is the effect of jth parity (one to ≥6), Yk is the effect of kth year (2013 - 2018)
and eijkl is the random residual error

RESULTS AND DISCUSSION

Population Structure and Off-take Rates

From the monitored goat population during the baseline, about 51.7% of the population was
covered by does, 40.2% weaned kids, 2.1% kids, 5.2% bucks and about 0.6% castrated bucks
(Figure 1). The current result on the total proportion of doe’s was slightly lower than the reports
of Belay and Mengistie (2013) and Solomon et al. (2014) who reported 56.6% and 51.8%,
respectively for the same breed. Lower numbers in this study might be due to the smaller number
of households used and study period difference. Death, sale, share and slaughtering with
respective percentage of 22%, 44%, 5.5% and 8% were the major off-take reasons recorded
(Figure 2). The result also revealed that sale was recorded as the highest proportion for goat
disposal. Male kids and doe have got the higher proportion in the off-take, which might be
directly linked with the reason that goat keepers mostly took either male kids or old does to the
market.

Figure 1: Population structure by age category of the baseline flock

41
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 2: Goat off-take reasons by age category at the study village

Body Weight Changes across Selection Years

The body weight of kids at all ages (birth, weaning, six month, nine month and yearling weight)
was significantly (p<0.05) affected by selection years (Table 2). Nevertheless, the body weight
changes across selection years were very slight which might be due to the area being
characterized as very fragile, feed scarce and prolonged drought prone (Figure 3). The overall
mean of weight at birth, three, six, nine and twelve months were 2.09±0.01kg, 6.99±0.03kg,
8.61±0.04kg, 11.4±0.12kg and 14.1±0.1kg, respectively. The body weight changes revealed in
the present study was lower than some of Ethiopian indigenous goats, 2.71±0.04kg, 2.36±0.05kg
for birth weight of Bati and Borana goats, respectively (Hulunim et al., 2019) and 1.91, 6.84,
9.13 and 14.25kg for birth weight, weaning weight, six month weight and yearling weight,
respectively for same breed (Belay and Mengistie, 2013). Zeleke (2018) reported higher value of
birth weight (3.19kg) for Somali goats.

Sex of kids had not produced any pronounced effect on the body weight changes across different
ages. The result of the present study differed with the result of Hulunim et al. (2019) who
reported a significant (p<0.05) difference observed between males and females for their daily
weight gain for both Borana and Short-eared Somali goats. However, the author also reported
sex of kids affected significantly (p<0.05) the total daily weight gain (birth – 90days) of Bati
goats.

42
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Six month and yearling weights of Abergelle goats were heavier than mid Rift Valley goats
(7.87±1.62 and 12.85±2.55kg, respectively) reported by Tesfaye et al. (2000) and lower than
13.61±0.40 and 20.15±0.67kg, respectively of the central highland goats at Debre Berhan area
(Tesfaye et al., 2006). In the present study, single birth kids had significantly (p<0.05) higher
body weight at birth with insignificant body weight changes for the remaining ages. Except nine
month weight, kids which were born during wet season show significantly (p<0.05) higher body
weight changes across all ages. The reason might be allied with the favored feeding management
as the wet season can produce relatively ample feeds unlike the dry season.

43
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Least square means (±SE) of live body weights at different age (kg) of Abergelle goats at Bilaqu village
BWT TMWT SMWT NMWT YWT
Variables N LSM ± SE N LSM ± SE N LSM ± SE N LSM ± SE N LSM ± SE
Over all 1607 2.09±0.01 1370 6.99±0.03 1306 8.61±0.04 1072 11.4±0.12 771 14.1±0.1
CV% 13.5 17.6 16.2 35.5 16.9
Sex NS NS NS NS NS
Male 852 1.92±0.01 723 6.96±0.05 692 8.8±0.05 562 11.7±0.07 368 15.2±0.15
Female 755 1.88±0.01 647 6.92±0.05 614 8.6±0.06 510 11.5±0.25 403 14.7±0.13
Birth type * NS NS NS NS
Single 1542 2.15±0 a 1323 7.04±0.03 1262 8.62±0.04 1035 11.7±0.13 753 14.8±0.10
Twin 56 2.01±0.03a 47 6.84±0.18 44 8.78±0.21 37 11.5±0.33 18 15.1±0.79
Triplet 9 1.63±0.09b 0 - 0 - 0 - 0 -
Birth season * * * NS *
Dry 466 1.91±0.01 416 6.74±0.04 380 8.51±0.06 253 11.4±0.5 181 14.5±0.23
Wet 1141 1.95±0.00 954 7.14±0.04 926 8.9±0.05 819 11.8±0.05 590 15.4±0.11
Year * * * * ***
2014 400 1.94±0.01b 350 7.51±0.08a 315 9.36±0.10a 265 11.1±0.1b 241 12.8±0.11d
2015 243 1.97±0.01b 239 6.74±0.06b 221 8.19±0.07c 128 13.2±0.97a 90 16.9±0.30a
2016 239 1.86±0.01c 222 6.81±0.05b 213 8.26±0.07c 175 10.6±0.1c 105 15±0.27b
2017 266 1.81±0.01c 261 6.19±0.06c 260 8.60±0.06b 226 11.8±0.11ab 190 16.3±0.22a
2018 312 1.94±0.01b 298 7.45±0.08a 297 9.12±0.08a 278 11.3±0.07b 145 13.7±0.12c
2019 147 2.06±0.02a - -
Parity * NS NS * NS
1 363 1.80±0.01b 311 6.6±0.07 292 8.46±0.08 242 11.12±0.1b 157 14.11±0.21
2 288 1.86±0.01b 240 6.83±0.08 230 8.62±0.09 178 11.34±0.13a 130 14.33±0.25
3 256 1.91±0.01a 217 6.71±0.09 213 8.46±0.09 175 11.21±0.13a 116 14.23±0.25
4 308 1.90±0.01a 259 6.78±0.08 247 8.57±0.104 218 11.31±0.12a 166 14.72±0.24
5 219 1.93±0.01a 195 6.89±0.09 186 8.54±0.11 150 11.03±0.12b 116 14.2±0.23
6 117 1.94±0.02a 103 6.73±0.13 94 8.44±0.14 74 12.72±1.67a 60 14.47±0.38
≥7 40 1.94±0.04a 33 7.13±0.23 32 8.51±0.22 26 11.66±0.34a 19 15.07±0.71
Where BWT = birth weight, TMWT = three month weight, SMWT = six month weight, NMWT = nine month weight, YWT = yearling weight,
N = total population, LSM = least square means and SE = standard error, '*' = p<0.05, '**' = p<0.01, '***' = p<0.001 and NS = no
significant difference

44
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: Person correlation among weights at different ages

BWT TMWT SMWT NMWT YWT
BWT 1.00000
TMWT 0.16762* 1.00000
SMWT 0.09739 0.80393*** 1.00000
NMWT 0.00725 0.16263* 0.26482** 1.00000
YWT -0.07668 0.10141* 0.30367** 0.31093** 1.00000
Where: '*' = significantly associated at p<0.05, '**' = significantly associated at p<0.001,
‘***' = significantly associated at p<0.001

Growth trend across selection years

18
16
14
Body Weight (kg)

12 2014
10 2015
8 2016

6 2017

4 2018

2
0
Bwt Tmwt Smwt Nmwt Ywt

Figure 3: Body weight changes across selection years

Milk Production Changes across Selection Years

Milk production performance (ADMY) across the selection period/year was significantly
different (p<0.05) (Table 4). It was significantly varied during each year in an increasing rate.
In addition, the phenotypic milk production trend across years showed a positive increasing
trend (Figure 4). Lactation length, average daily milk yield and total milk yields were all
significantly affected by season and year of birth. The overall average lactation length, total
lactation milk yield and average daily milk yield were 10.5±0.1 weeks, 25.63±0.29kg and
358.04±2.97gm, respectively (Table 4). However a report on milk production performance of
indigenous goats are lacking in our country, the result of this study was slightly higher than
the report of Mohammed et al. (2012) which was 86 days, 17.73kg and 208.72gm of lactation
length, total milk yield and average daily milk yields, respectively for Arsi- Bale goat.

45
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

On the other hand, the average daily milk yield in this study was lower than the report of
Dereje et al. (2015) who reported 450g/day for Afar goat breed. Doe’s that have given birth
during wet seasons had relatively higher daily milk yield due to the availability of feed.
Average daily milk yield was significantly increased across years that can be associated with
the response to selection. Considering the milk yield, to some extent, there was inconsistent
data recording in response to flock mobility which was the most critical problem during the
peak periods of lactations as the farmers migrated the flock to distant areas for long seasons.

Table 4: Least square means (±SE) of LL (weeks), TMY (kg) and ADMY (gm) of Abergelle
goats at Bilaqu village
LL TMY ADMY
Variables N LSM ± SE N LSM ± SE N LSM ± SE
Overall 803 10.5±0.1 803 25.63±0.29 803 358.04±2.97
CV% 26.9 30.6 20.7
Season * * *
Dry 213 11.0±0.2 218 25.24±0.52 213 323.27±6.27
Wet 590 10.37±0.1 610 26.32±0.34 590 370.58±3.20
Year * * *
2014 190 10.9±0.2bc 190 21.96±0.60c 190 300.31±7.41d
bc b
2015 120 10.7±0.2 120 23.98±0.66 120 328.42±7.34c
2016 122 10.9±0.2bc 122 28.10±0.64a 122 375.13±5.87b
ab a
2017 116 11.6±0.1 116 29.33±0.46 116 362.56±5.01b
2018 255 9.7±0.2c 255 25.58±0.57b 255 389.34±3.66a
Parity NS NS NS
1 180 10.88±0.21 180 25.20±0.61 180 339.77±6.58
2 138 10.44±0.27 138 25.55±0.70 138 363.12±6.84
3 143 10.87±0.24 143 25.83±0.63 143 347.87±6.13
4 145 10.88±0.22 145 25.12±0.68 145 348.90±6.95
5 111 10.89±0.27 111 26.31±0.83 111 354.62±8.72
≥6 86 10.67±0.60 86 25.72±1.45 86 352.62±14.33
Where LL = Lactation length, TLMY = Total lactation milk yield, ADMY = Average daily
milk yield, N = total population, LSM = least square means and SE = standard error, '*' =
p<0.05, and NS = no significant difference

46
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 4: Trends of average daily milk yield across selection years

CONCLUSION AND RECOMMENDATIONS

Body weight of kids and milk yield were slightly increased during the selection years, though
the improvement was not as such faster; which needs strategic supplementation of the breeder
stock in order to catch the real genetic merit of the breed. The study gave some insight for its
great possibility of improvement of growth and milk production traits through strategic
design and implementation of selective breeding approaches. Strong consideration of the
fixed factors like birth type, season of birth and parity will become an important concern
during performance evaluation and selection. Six and nine month weight had better
association with yearling weight (r = 0.3 and 0.31, respectively), which calls for arranging the
selection age from six up to nine month which can also allow the chance of maintaining elite
kids. The kids born from the dams with parity of ≥3 had better birth and post weaning body
weight, thus keeping these doe’s as breeder stock can allow for the improvement of the
population.

ACKNOWLEDGMENT

We thank International Center for Agricultural Research in the Dry Areas (ICARDA-
Ethiopia) for collaboration, financial and technical supports for the study.

47
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Alubel Alemu. 2015. On-farm phenotypic characterization and performance evaluation of

Abergelle and central highland goat breeds as an input for designing community-based
breeding program, MSc thesis, Haramaya University, Ethiopia.
Behnke R. 2010. The Contribution of Livestock to the Economies of IGAD Member States:
study findings, application of the methodology in Ethiopia and recommendations for
further work. IGAD LPI Working Paper No. 02 -10.
Bekahgn Wondim, Mulatu Gobeze and Baye Biresaw. 2019. Feed resource availability,
livestock migration pattern and synthesis of feeding calendar at Wag-Lasta, Ethiopia.
Daagu International Journal of Basic and Applied Research Vol 1, Issue-1, pp (26-40).
Belay Derbie and Mengistie Taye. 2013. Growth performance and carcass characteristics of
central highland goats in Sekota District, Ethiopia. Agricultural Advances, 2(8), 250–
258. https://doi.org/10.14196/aa.v2i8.902
Dereje T, Mengistu U, Getachew A. and Yoseph M. 2015. A review of productive and
reproductive characteristics of indigenous goats in Ethiopia. Livestock Research for
Rural Development, 27(2), 1–19.
FARM AFRICA. 1996. Goat Types of Ethiopia and Eritrea. Physical description and
management systems. Published jointly by FARM-Africa, London, UK and International
Livestock Research Institute, Nairobi, Kenya. 76pp.
Hulunim Gatew, Halima Hassen, Kefelegn Kebede, Aynalem Haile, Raimundo Nonato,
Braga Lobo and Barbara Rischkowsky. 2019. Early growth trend and performance
of three Ethiopian goat ecotypes under smallholder management systems. Agriculture
and Food Security, https://doi.org/10.1186/s40066-018-0249-2.
Mohammed Bedhane, Aynalem Haile, Hailu Dadi and Tesfaye Alemu T. 2012. Estimates of
genetic and phenotypic parameters for milk traits in Arsi-Bale goat in Ethiopia.
Livestock Research for Rural Development, 24(6), 2012.
Mueller J. P., Rischkowsky B., Haile A., Philipsson J., Mwai O., Besbes B. and Wurzinger
M. 2015. Community-based livestock breeding programmes: Essentials and examples.
Journal of Animal Breeding and Genetics. https://doi.org/10.1111/jbg.12136
Solomon Abegaz. 2014. Design of community based breeding programs for two indigenous
goat breeds of Ethiopia. PhD Thesis. BOKU-University of Natural Resources and Life
sciences, Department of Sustainable Agricultural Systems, Division of Livestock
Sciences, Vienna, Austria.

48
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Solomon Abegaz Guangul, Sölkner J., Gizaw S., Dessie T., Haile A., Mirkena T. and
Wurzinger M. 2014. Optimizing alternative schemes of community-based breeding
programs for two Ethiopian goat breeds. Acta Agraria Kaposváriensis (2014) Vol 18
Supplement 1, 47-55.
Solomon Abegaz, Sölkner J., Solomon Gizaw, Tadele Dessie, Aynalem Haile and Wurzinger
M. 2013. Description of production systems and morphological characteristics of
Abergelle and Western lowland goat breeds in Ethiopia: implication for community-
based breeding programmes. Animal Genetic Resources/Ressources Génétiques
Animales/Recursos Genéticos Animales, 53, 69–78. https:// doi.org/ 10.1017/s
2078633613000088
Tesfaye G., Sisay L., Dereje T., Abebe M. and Solomon G. 2006. Growth and reproductive
performance of central highland goats in North Shoa and South Wollo. Proceeding of the
first annual conference on completed livestock research activity organized in Bahir Dar,
Ethiopia, 14-17 August 2006, 7-13.
Tucho T.A., A. Ragassa and L. Fita. 2000. Preliminary production and reproduction
performance evaluation of Mid Rift Valley and Boran Somali goats. In: R.C. Merkel, G.
Abebe and A.L. Goetsch (eds.). The Opportunities and Challenges of Enhancing Goat
Production in East Africa. Proceedings of a conference held at Debub University,
Awassa, Ethiopia from November 10 to 12, 2000. E (Kika) de la Garza Institute for Goat
Research, Langston University, Langston, OK pp. 182-186.
Zeleke M. Z. 2018. Environmental influences on pre-weaning growth performances and
mortality rates of extensively managed Somali goats in Eastern Ethiopia. Livestock
Research and Rural Development; 2007: 19(12). http://www.lrrd.org/ lrrd1 9/12/zele1
9186.htm.

49
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Comparative Evaluation of Lowland Sheep Breeds for their Growth, Carcass Yield and
Quality under Graded Levels of Concentrate Supplementation

Alemu Tarekegn1, Solomon Abegaz2, Kifetew Adane1, Wubie Atalel1, Yohanes Wolelaw1
and Alayu Kidane1
1
Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia
2
Woldia University, P.O. Box 400, Woldia, Ethiopia

ABSTRACT

Performance of three sheep breeds in the lowland areas of western Amhara, the Gumuz,
Rutana, and Begiet was evaluated using three levels of concentrate (400, 500, and 600 g head-
1
day-1) having the same Noug seed cake: Wheat bran ratio (75:25) under feedlot condition.
The roughage was native grass hay which was offered ad libitum. Thirty-two 8 to 9 months
old intact male sheep (12 breed-1) were randomly allotted to the dietary treatments, fed for 90
days, and slaughtered at an age of approximately 12 months. Breed and diet (concentrate
level) significantly affected average daily gain (ADG). Gumuz and Rutana sheep breeds
achieved higher average daily gain (ADG) than Begiet. Sheep fed on concentrate L2 (500 g
head-1 day-1) recorded significantly higher ADG than sheep fed on L1 and L3. As a result of
better slaughter weight (SW) achieved higher hot carcass weight (HCW) was recorded by
Gumuz and Begiet sheep breeds. Because of the greater difference in the proportion of non-
carcass components and better HCW achieved relative to slaughter body weight, better
dressing percentage (DP) was achieved by Gumuz. Breed and concentrate level did not
improve main carcass components, edible and non- edible offals. Meat quality attributes were
not influenced by breed. Among the concentrate levels, L2 was optimum to improve major
meat quality attributes like juiciness, flavor, and overall acceptance. Sheep supplemented
with concentrate L2 had the highest net return (855.06 ETB and highest MRR (14.12)
compared to the other concentrate levels. Compared to other lowland breeds Gumuz and
Rutana sheep breeds recorded highest net return (733.42 and 531.67 ETB) and highest MRR
(10.25 and 11.42 ETB), respectively. Thus, it is recommended that the use of adaptive sheep
breeds (Gumuz and Rutana) to that specific environment and use of concentrate level two
(L2) are biologically efficient and potentially profitable in the feeding of growing lowland
sheep to the area and other similar areas.

Keywords: Begiet, concentrate level, Gumuz, Rutana, sheep breed

50
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Although there exists a 30.7 million sheep population in Ethiopia (CSA, 2017), their
production and productivity are found very low. The low performance of local sheep in terms
of live weight (LW) gain and carcass yield is mainly due to inadequate nutrition associated
with reliance on sole natural pasture, crop residues and/or stubble grazing, which are
inherently low in nutrients available being subjected to great seasonal variations (Solomon et
al., 2008). Temporally abundance of forage during the short wet season is followed by long
dry periods with feed deficit leading to a cycle of live weight gain and loss of animals. Thus,
sheep often takes a longer period to attain market weight, lowering its production efficiency.
The present export market (to the Middle East countries) for Ethiopian live sheep and mutton
is demanding animals of low land origin, weighing between 25 and 30 kg at yearling age.
However, most local lambs slaughtered at this age weigh 18 to 20 kg (IAR, 1991). Thus, the
lack of consistent supply of the required animals at a younger age has remained a major
challenge for mutton and live sheep exporters.

The growth performance of animals at an early age has a strong impact on mature body
weight and the age at which it is achieved. Slow growth rate and exposure to a cycle of
weight gain and loss due to fluctuations in the annual cycle of feed supply is the main reason
for sheep to reach higher body weight at an older age. Thus, proper nutrition at a younger age
is a requirement to allow continuous growth of animals and prepare them for the finishing
phase to achieve a better growth rate and acceptable slaughter and carcass weight (McDonald
et al., 2002).

The major sheep breeds in North-Western lowland areas of the Amhara region are Gumuz,
Rutana and Begiet. There is a high preference for Gumuz, Rutana and Begiet sheep breeds
due to their adaptation to the harsh environment, prolificacy, large frame and fast growth,
especially during the early growth stage. The preferences of these breeds might be due to the
breeds’ lowland background, adaptation of the buyers to the conformation of the animals, and
the taste of the meat (Asfaw et al., 2011). Though there exists a large sheep population with
special merit, very few efforts were made so far in improving the production and productivity
of the sheep breeds in the area. Nevertheless, available information on body weight gain and
carcass yield most lowland breeds like Gumuz, Rutana and Begiet sheep in response to
different regimes of feeding is scanty.

51
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A study by Alemu et al. (2018) demonstrated that supplementation of Gumuz sheep with
agro-industrial by-products improved daily body weight gain, dressing percentage and hot
carcass weight. Though this study generates some information, it did not provide detailed
information about the maximum level of concentrate feed required and how the manipulation
of the feeding regimes of Gumuz sheep will improve feed utilization and body weight gain to
achieve the desired carcass weight that is attractive for the export market. Information on the
sheep breeds, especially Gumuz, Rutana and Begiet growth performance, carcass
characteristics and sensory attributes of meat, when supplemented with different levels of
concentrate feed, is scarce. Therefore, the present study was designed to study the effects of
genotype and different levels of concentrate supplementation on growth, carcass
characteristics, and meat quality of three sheep breeds in the low land areas of North Western
Amhara.

MATERIALS AND METHODS

Description of the Study Area

The study was conducted at the Gende wuha, Gumuz sheep breed improvement station,
which is found in West Gondar Zone of the Amhara National Regional State (ANRS),
Ethiopia. The area is located between 12o46’45’’N and 36o24’20’’E, and at about 906 km
northwest of Addis Ababa, and about 158 km west of Gondar Town. The agro-ecological
map of the area lies under the moist kola agro-ecological zone (AEZ). The altitude of Gende
wuha is about 745 m.a.s.l. The mean annual rainfall of Gende wuha ranges from 850 to 1100
mm, and it receives a unimodal rainfall. The minimum annual temperature ranges between
22oC and 28oC. The daily maximum temperature becomes very high from March to May,
during which the temperature can reach as high as 43oC. The mean annual temperature is
about 31oC (ILRI, 2005).

Feed Preparation and Feeding Management

Locally available mixed sward hay under the station was properly harvested and used as a
basal diet to the study animals. The basal diet was manually chopped to the size of about 1 - 5
cm, thoroughly mixed and stored under shade to maintain its quality. Noug seed cake and
Wheat bran were purchased from oil extracting and milling industries and kept properly. Hay
was fed ad libitum at 20% refusal rate; whereas the daily concentrate supplements were
limited to the respective treatment level in g DM day-1.

52
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The concentrate mix was 74% Noug seed cake, 25% Wheat bran and 1% salt on a DM basis.
The concentrate mix was formulated to contain at least the recommended amount of CP and
energy (18% CP and 9 MJ ME/kg of DM) for intensive feeding management (Flint, 2005).
Water was available at all the time to the experimental animals, while supplements were
offered twice a day in two equal portions at 08:00 AM and 16:00 PM.

Experimental Animals and their Management

A total of 36 sheep from three breeds (Gumuz, Rutana and Begiet) were used to the study.
The age of the animals was between eight to nine months, and the age of animals was
determined based on the information obtained from the owners. The lambs were ear tagged
and quarantined for 21 days to get them adapted to the environment and to check their health
conditions. The sheep were dewormed, sprayed, and vaccinated against known parasites and
diseases during the quarantine period. The animals were kept under a shed in individual pens
of 1.2 x 0.8 m dimension with free access to clean water and fed the experimental diet for 90
days.

Experimental Design and Treatments

The experiment was arranged in a completely randomized block design (RCBD) with 3 x 3
factorial arrangements (three genotypes and three concentrate levels). Animals within a
genotype were blocked according to their initial body weight and randomly assigned to one
of the concentrate levels, each level consisting of four animals. The three concentrate levels
were 400, 500 and 600g head-1 day-1.

Feed Intake and Body Weight Measurements

Daily weights of concentrate and hay offered and refused were recorded to derive the daily
feed intake amount. The study animals were weighed every ten days in the morning before
watering and feeding during the feeding trial by using salter balance having a sensitivity of
0.1kg. The initial weight (IW) and final weight (FW) were recorded twice for two
consecutive days and the average was taken as a net weight gain. Average daily gain (ADG)
was calculated as the difference between final body weight and the initial body weight of the
sheep divided by the total number of feeding days. Total gain (TG) was calculated as the
difference between final and initial weights. Feed conversion efficiency (FCE), which is the
measure of feed utilization, was calculated as a unit of body weight gain per unit of feed
consumed.

53
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Slaughtering Procedures, Carcass Characteristics and Meat Quality

At the end of the growth trial, all experimental sheep from each feeding treatment have fasted
for 12 hrs overnight with free access to water (PRIMEFACT 340, 2007). The next morning,
slaughter body weight (SBW) was taken right before the animals killed, and all sheep
slaughtered to determine carcass characteristics. After flaying, the weight of offal like head,
skin, and feet, heart, lungs and trachea, liver with gallbladder, spleen, testis, penis, kidneys,
visceral fat, reticulo-rumen, omaso-abomasum, small and large intestine were recorded.

The full alimentary tracts weighed, and then the contents were emptied and reweighed to
determine the weight of the contents. Empty body weight (EBW) was calculated as slaughter
weightless gut content. Hot carcass weight (HCW) was recorded immediately after slaughter
and computed by excluding contents of thoracic, abdominal and pelvic cavities, head, and
skin with fetlock. Dressing percentage (DP) computed as a proportion of hot carcass weight
to slaughter weight (Gilmour et al., 1994).

The rib-eye muscle area (REA) of each animal was determined by tracing the cross-sectional
area of the 11th and 12th ribs after cutting perpendicular to the backbone rib eye area based on
the recommendation of Torell and Suverly (2004). The left and right rib-eye muscle areas
were traced on a transparent waterproof paper and the area was measured by using a 1cm2
grid square having 5 by 5 dots within a grid. Fat thickness was measured in the middle-upper
part of the rib eye area by using a graduated ruler having a sensitivity of 1 mm.

Sensory Evaluation

Fresh samples of meat from each treatment from loin, ribs, and leg were taken and thawed for
4 hrs, cut into chops of an average of 40g, and labeled for identification. Meat cubs from each
cut were cooked in water at a temperature of 80oC for 60 minutes in a circulating water bath
using a gas cooker as described by (Fasae et al., 2010). Nine consumer panelists were used to
evaluate sensory parameters of meat cuts and soup in a balanced design. They were instructed
to chew on a sample and drink soup from each treatment and score it for tenderness,
juiciness, flavor, connective tissue amount, and overall acceptability.

54
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The panelists were score each sample on a five-point hedonic scale (tenderness; 5 =
extremely tender, 1 = very tough; Juiciness, 5 = extremely juicy, 1 = very dry; Flavor, 5 =
extremely intense, 1 = bland; connective tissue amount, 5 = none, 1 = moderately abundant;
Overall acceptability, 5 = excellent, 1 = poor) (Iwe, 2002). Panelists were eating a bit of
bread and drinking still water at the beginning and between samples just to make the palate
conditions similar and to reduce flavor carryovers.

Cost- Benefit Analysis

Cost of all variable inputs, buying and selling prices of lambs, and labour cost for hay
chopping were considered to determine the net income of production. The direct purchasing
price of lambs was used as an initial price for lambs to compute the cost-benefit analysis. At
the end of the experiment, three experienced lamb dealers have estimated the selling price of
each experimental sheep before slaughtering. Monetary values of all other variable inputs
were considered at the prevailing market price.

According to Upton (1979) procedure, net income (NI) was obtained as the amount of money
left when total variable costs (TVC) subtracted from the total returns (TR) of the purchasing
and selling price of sheep in each treatment before and after the experiment carried on.

NI = TR - TVC

The change in net income (∆NI) was calculated as the difference between changes in total
return (∆TR) and the change in total variable costs (∆TVC), and this is to be used as a
reference criterion for decision on the adoption of a technology.

∆NI = ∆TR - ∆TVC

The marginal rate of return (MRR) that measures the increase in net income (∆NI) associated
with each additional unit of expenditure (∆TVC) was calculated and expressed in ratio as:

MRR = ∆NI / ∆TVC

55
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Analysis

Statistical analysis was performed to determine the effect of genotype, supplement level and
the interaction between genotype and supplement level. The data collected on feed intake,
body weight change, feed conversion efficiency, carcass parameters and sensory attributes of
meat was subjected to analysis of variance (ANOVA) using the General Linear Model
(GLM) procedure of SAS (2003) according to a 3 × 3 factorial arrangement with breed and
diet as main effects in Randomized Complete Block Design. When the difference was
significant (P<0.05), Tukey honestly significant difference test was used to locate differences
between the treatment means. The statistical model was:

Yijkl = μ + Bi + Gj + Fk + (G×F) jk + eijkl

Where: Yijkl is the response variable, μ is the overall mean, Bi is the effect of block, Gj is the
effect of genotype, Fk is the effect of feeding level, (G×F)jk is interaction between genotype
and feeding level and eijkl is the random error.

RESULTS AND DISCUSSION

Dry Matter Intake

Mean daily dry matter intake (DMI) of sheep breeds (Gumuz, Rutana and Begiet) fed with
hay as a basal diet and supplemented with different levels of concentrate is presented in Table
1. The daily hay, supplement and total dry matter intake of different sheep breeds were found
not significant (P>0.05). This might be due to the sheep breeds used for this study are from
the same origin, their weight is also relatively the same and thus the animals are expected to
feed on their body weight bases. Significant (P<0.01) difference was observed among sheep
breeds in their total dry matter intake as a percent of body weight. Rutana sheep breed had
significantly (P<0.01) higher total dry matter intake than the other breeds as a result of its
relatively lower body weight.

For indigenous Gumuz yearling sheep, DM intake in the range of 2.7 - 3.62% body weight
was reported by Alemu et al. (2018) when they were supplemented with different levels of
concentrate, which is by far lower than the result obtained in this study. This is due to the
differences in type and level of concentrate used at different times which ultimately affect the
amount of total dry matter intake and body weight.

56
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Mean daily dry matter intake of lowland sheep breeds (Gumuz, Rutana and Begiet) supplemented with different levels of concentrate at
Gende wuha
Sheep breed (B) Concentrate level (C) P value
Parameters Gumuz Rutana Begiet L1 L2 L3 B C B* C
Dry matter intake
Hay DM intake (g/day) 563.55 564.95 565.82 591.19a 582.17a 520.97b 0.9936 0.0033 0.5718
Supplement DM intake (g/day) 500.00 500.00 500.00 400.00c 500.00b 600.00 a
- 0.0000 -
b
Total DM intake (g/day) 1063.50 1065.00 1065.80 991.20 1082.20a 1121.00a 0.9934 0.0000 0.5718
b
DM intake (% BW) 4.31 4.63a 4.19b 4.22b 4.24b 4.68 a
0.0093 0.0029 0.7733
0.75
DMI (per kg W ) (g/day) 36.12 37.17 61.09 43.71 44.75 45.92 0.0624 0.9810 0.7421
*a, b, c = means within rows having different superscript letters are significantly different; BW = body weight; DM = dry matter; DMI = dry
matter intake; P = probability; L1 = 400g Supplement (75% NSC: 25% WB); L2 = 500g Supplement (75% NSC: 25% WB); L3 = 600g
Supplement (75% NSC: 25% WB).

57
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Total daily dry matter intake of different sheep breeds supplemented with concentrate L2 and
L3 was significantly (P<0.001) higher than sheep supplemented with L1 (Table 1). This was
probably due to the addition of extra CP as a result of concentrate level increase that can
stimulate efficient rumen fermentation, more passage rate and intake, which increased the
basal diet intake and resulted in better total DMI (McDonald et al., 2002). Feed intake can be
maximized if the feed provides all the nutrients required by appropriate rumen microbes and
by the tissue of the animal.

The total DM intake as a percent of body weight in the current study was significantly
(P<0.01) different among concentrate levels. Concentrate L3 recorded significantly (P<0.01)
higher total DM intake as a percent of body weight than other levels. This could be due to the
higher concentrate intake achieved. Supplementary concentrate levels markedly affect total
DM intake per live body weight (Tadesse et al., 2003). Daily total dry matter intake per unit
metabolic body weight (g/kg W0.75) was not significant (P>0.05) among sheep fed different
concentrate levels.

Body Weight Change and Feed Conversion Efficiency

The body weight (BW) parameters of different sheep breeds and sheep in different
concentrate levels are presented in Table 2. Final body weight was significantly (P<0.05)
affected by breed and concentrate level. Associated with its numerical highest initial body
weight recorded, the relatively highest final body weight was achieved by Begiet sheep
breed. Among the concentrate levels sheep fed on L2 had significantly higher (P<0.05) final
body weight than sheep consumed L1. This might be due to the BW gain tended to increase
for the higher CP intakes as a result of increasing the concentrate level. Mulu et al. (2008)
also noted that supplementation with high levels of protein source supplements resulted in
higher final body weight (BW) than the low level of supplementation. Different sheep breeds
and sheep fed on different concentrate levels significantly (P<0.05) affect average daily gain
(ADG). Gumuz and Rutana sheep breeds achieved 1.46 and 1.77 kg extra body weight
compared to Begiet by growing at an average of 29.17 and 35.5 g greater rate per day,
respectively. The observed body weight gain variation between breeds could be attributed to
their adaptation to different environments. Sheep fed on L2 recorded significantly (P<0.05)
higher ADG than sheep fed on L1 and L3.

58
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The results of average daily body weight gain of sheep supplemented with different levels
were higher than those reported by Alemu et al. (2018) and Fentie and Solomon (2008) which
were in the range of 44.26 - 86.67 and 70.11 - 82.44 g day-1 for Gumuz sheep fed a basal diet
grass hay and supplemented with 400g day-1 head-1 of Noug seed cake and Wheat bran at
different proportions and for Farta sheep fed a basal diet of natural grass hay when
supplemented with 300g day-1 head-1 with Noug seed cake, Wheat bran and their mixtures,
respectively. This could be due to the differences in breed, concentrate level and ratio,
environment and age of animals used at different times.

Table 2: Mean initial, daily and final body weight of lowland sheep breeds (Gumuz, Rutana
and Begiet) supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Parameters Gumuz Rutana Begiet L1 L2 L3 B C B* C
Initial BW (kg) 20.02 18.33 22.53 20.06 20.46 20.35 0.0561 0.8805 0.2321
ab
Final BW (kg) 24.78 23.40b 25.83a 23.81 b
25.82 a
24.38ab 0.0438 0.0345 0.7764
a
Net gain (kg) 4.76 5.07a 3.30b 3.75 b
5.36 a
4.03b 0.0170 0.0316 0.5397
a
ADG (g/d) 95.17 101.50a 66.00b 75.00 b
107.17 80.50b
a
0.0170 0.0316 0.5397
a
FCE 0.09 0.10a 0.06b 0.08 ab
0.10a 0.07b 0.0169 0.0415 0.6038
*a, b = Means within the same row not bearing a common superscript letters differ
significantly; ADG = average daily body weight gain; BW = body weight; FCE: feed
conversion efficiency; P = probability; L1 = 400g Supplement (75% NSC: 25% WB); L2 =
500g Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75% NSC: 25% WB).

Feed conversion efficiency (FCE) was significant (P<0.05) between different sheep breeds.
Rutana and Gumu sheep breeds recorded better FCE than Begiet. This is due to better daily
body weight gain achieved than feed consumed as a result of genetic difference. Significant
(P<0.05) difference was observed in FCE between sheep fed on different concentrate levels.
The observed higher FCE in L2 may be due to higher daily body weight gain achieved than
feed consumed. The current result was comparable to the FCE recorded in Gumuz sheep fed
natural pasture hay and supplemented with 400 g day-1 of concentrate with a proportion of
75% Noug seed cake: 25% Wheat bran (Alemu et al., 2018).

59
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 1: Body weight change over time of lowland sheep breeds (Gumuz, Rutana and
Begiet) supplemented with different levels of concentrate

The overall trends of body weight changes of the concentrate levels across the feeding trial
periods are presented in Figure 1. The relative increase in body weight change achieved in all
concentrate levels with a slight overlying of L2 indicated the fact that animals assigned in
each concentrate level received continual optimum nutrients required for production. Slight
daily body weight gain achieved during the last ten days of the experimental period was an
indication of no further extending the feeding time to be effective.

Carcass Parameters

The mean values of carcass characteristics of three sheep breeds (Gumuz, Rutana, and
Begiet) supplemented with different levels of concentrate in the current experiment are given
in Table 3. Slaughter body weight (SW), empty body weight (EBW), hot carcass weight
(HCW), and dressing percentage as a proportion of SW were significantly (P<0.05) affected
by breed. Begiet breed had higher (P<0.05) EBW than other breeds. The higher EBW
achieved by this breed might be due to the lower proportion of gut content and BW increase
since EBW is the difference between SW and gut content (Jadish, 2004). As a result of better
SW achieved by breeds significantly (P<0.05) higher HCW was recorded by Gumuz and
Begiet. Dressing percentage (DP) of Gumuz sheep breed was significantly (P<0.05) higher
than the rest of sheep breeds. The reason for variation in DP among genotypes in the present
study was because of the greater difference in the proportion of non-carcass components such
as gastrointestinal tract (GIT) and due to better HCW achieved relative to slaughter body
weight.

60
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The dressing percentage of the current study result is within the range of the average dressing
percentages (ADP) of tropical sheep (40 - 50%) reported by (William et al., 2003). The fat
thickness and rib-eye muscle area did not significantly (P>0.05) affected by genotype. There
was no significant (P>0.05) difference observed in most of the carcass parameters among
sheep receiving different concentrate levels.

Table 3: Carcass characteristics of lowland sheep breeds (Gumuz, Rutana and Begiet)
supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Carcass parameters Gumuz Rutana Begiet L1 L2 L3 B C B*C
No. of animals slaughtered 9 9 9 9 9 9
Slaughter BW (kg) 24.77ab 23.32b 25.24a 23.79 25.01 24.30 0.0416 0.6101 0.6448
Empty BW (kg) 20.24b 19.27b 22.01a 20.04 20.78 20.69 0.0101 0.6063 0.6347
Hot carcass weight (kg) 11.02a 9.76b 10.57a 10.29 11.23 10.89 0.0140 0.0899 0.5918
Dressing percentage (%)
Slaughter BW base 44.48a 41.85b 41.88b 32.25 44.90 44.81 0.0192 0.0652 0.3518
Empty BW base 54.45a 50.65b 48.02b 51.35 54.04 52.63 0.0079 0.0831 0.2626
Fat thickness (mm) 5.67 7.61 5.39 7.28 5,83 5.56 0.3622 0.5445 0.5320
Rib-eye muscle area (cm2) 13.4 10.89 12.01 11.4 12.67 12.23 0.0647 0.4467 0.2244
*a, b = Means within the same row not bearing a common superscript letters differ significantly;
BW = body weight; P = probability; L1 = 400g Supplement (75% NSC: 25% WB); L2 = 500g
Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75% NSC: 25% WB).

Main Carcass Components

Main carcass components did not significantly (P>0.05) affected by genotype and different
concentrate levels.

Table 4: Main carcass components (gram) of lowland sheep breeds (Gumuz, Rutana and
Begiet) supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Main carcass components Gumuz Rutana Begiet L1 L2 L3 B C B*C
Fore- quarter 2032.1 1907.2 1883.7 1949.2 1770.9 2103.0 0.7355 0.2854 0.8461
Neck region 979.8 935.7 1034.1 972 1039.6 938.1 0.7482 0.7284 0.8280
Sternum (Brisket) 479.9 357.9 471.7 435 452.3 422.1 0.1575 0.9028 0.5574
Thoracic and lumbar region 812.9 709.7 834.2 781.2 809.2 775.3 0.2826 0.9050 0.5672
Rib- eye muscle 716.4 458.1 573.7 561.1 585.4 601.7 0.0851 0.9310 0.6718
Abdominal muscle 738.3 568.3 620.7 621.4 616.2 689.7 0.1043 0.5719 0.7092
Hind- quarter 2170.2 2002.8 1973.6 2107.3 1901.8 2137.4 0.5096 0.3803 0.1893
Pelvic (Rump) region 918.3 784.1 945.6 876.1 923.1 848.8 0.1919 0.7121 0.9602
Tail fat weight 262 167.9 221.8 194.11 217 240.6 0.1866 0.6444 0.8711
Ribs 916 786.7 912.8 877.4 895.2 842.8 0.2789 0.8361 0.4927
TMCC (Kg) 10.03 8.68 9.48 9.38 9.21 9.60 0.2445 0.8813 0.9435
*P = probability; TMCC = total main carcass components; L1 = 400g Supplement (75%
NSC: 25% WB); L2 = 500g Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75%
NSC: 25% WB).

61
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Edible Offal Components

Associated with the genetic difference among the edible offals, the effect of genotype was
significant (P<0.05) for testicle size and weight, while there was no significant (P>0.05)
difference exhibited in the rest of the edible offals. Associated with the physiologically early
maturing nature of the breeds significantly (P<0.05) higher testicle size and weight were
recorded by Gumuz and Rutana.

Table 5: Edible offal components (gram) of lowland sheep breeds (Gumuz, Rutana and
Begiet) supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Edible offals Gumuz Rutana Begiet L1 L2 L3 B C B*C
Blood 912.9 1041.9 1021.4 1091.2 927.8 957.3 0.2957 0.1576 0.7873
Heart 119.2 108.3 151.1 129.3 98 151.3 0.6076 0.4896 0.3995
Liver 431.6 394.8 403 436.2 408.9 384.2 0.6975 0.5259 0.7973
Kidney 86.3 91.4 88.4 89.3 90.2 86.7 0.5634 0.7386 0.3414
Tongue 253.6 105 169.9 136.1 114.6 277.8 0.4610 0.3407 0.7888
Reticulo-rumen 498.9 534.9 530 594.7 491.3 477.8 0.8493 0.2024 0.4847
Omaso-abomasum 279 318 279.7 272.1 378.7 243.9 0.8640 0.1648 0.8531
Hind gut 908.3 826 1167.8 856.3 926.7 1191.1 0.0870 0.2212 0.2869
Testicles 306.2a 257.7a 234.7b 364.2 317.4 270.9 0.0137 0.1227 0.2095
Kidney fat 108.9 102.9 82.1 95.7 98.8 99.3 0.6359 0.9912 0.3324
Pelvic fat 27.8 45.4 35.8 38.1 24.3 46.6 0.2918 0.1485 0.1797
Pancreas (g) 33.33 33.2 31.7 30 36.2 32 0.9780 0.7751 0.9386
Omental and mesenteric
fat(g) 209.8 114.7 153.3 182.8 150.3 143.7 0.1524 0.6738 0.9717
Total edible offals (kg) 3.89 3.72 4.11 3.95 3.75 3.95 0.4551 0.6574 0.6732
Total usable product (kg) 16.01 14.54 15.87 15.25 15.17 16.01 0.2768 0.6435 0.8915
Total usable product (%SW) 63.46 60.18 59.31 61.42 58.87 63.65 0.4225 0.5090 0.8335
*a, b = Means within the same row not bearing a common superscript letters differ
significantly; P = probability; SW= slaughter weight; L1 = 400g Supplement (75% NSC:
25% WB); L2 = 500g Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75% NSC:
25% WB).

Non- Edible Offal Components

Except for gallbladder with bile and gut content most individual non- edible offals and total
non-edible offals (kg) did not significantly (P>0.05) affected by genotype and different
supplement levels. Associated with their genetic difference, significantly (P<0.05) higher
gallbladder with bile was recorded by Gumuz and Rutana sheep breeds than Begiet.

62
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 6: Non-edible offal components (gram) of three sheep breeds (Gumuz, Rutana and
Begiet) supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Non-edible offals Gumuz Rutana Begiet L1 L2 L3 B C B*C
Head without tongue (kg) 1258.4 1508 1167.1 1357 1516.1 1060.4 0.2509 0.1058 0.6797
Skin (kg) 2.1 2.14 2.28 1.92 2.21 2.39 0.6862 0.1218 0.7352
Penis (g) 195.8 70.7 207.7 94.2 183.1 196.8 0.3425 0.5518 0.1552
Feet (g) 744.6 740 575.7 684.7 764.6 611 0.1961 0.3412 0.7954
Lung with trachea (g) 436.9 402.1 371.3 430.7 387.9 398.8 0.2474 0.4691 0.5275
Spleen (g) 105.1 70.9 105 113.6 77.2 90.2 0.3489 0.3966 0.7991
Gall bladder with bile (g) 17.67a 18.56a 8.44b 13.44 12.78 18.44 0.0131 0.2039 0.4768
Bladder (g) 62.9 57 33.3 60.78 45.44 47 0.2240 0.6281 0.5435
Total non-edible offals(kg) 5.28 5.37 4.98 5.04 5.51 5.08 0.6667 0.5085 0.8696
Gut content (kg) 3.03a 2.57b 1.9c 2.65a 2.76a 2.09b 0.000 0.005 0.0000
*a, b, c = Means within the same row not bearing a common superscript letters differ
significantly; P = probability; SW = slaughter weight; L1 = 400g Supplement (75% NSC:
25% WB); L2 = 500g Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75% NSC:
25% WB).

Meat Sensory Attributes

Sensory evaluation results of cooked meat are presented in Table 7. Meat quality attributes
were not significantly (P>0.05) influenced by breed. This might be due to the animals used
for the experiment being managed in the same manner. Swan et al. (1998) reported that the
breed did not affect the flavor and juiciness of cooked goat Longissimus dorsal (LD) muscle.
The flavor and overall acceptance were significantly (P<0.05) higher in sheep fed concentrate
L2 than sheep supplemented with L1 and L3. This could be due to higher total fat content
achieved as a result of bodyweight improvement of animals as fat has a positive relationship
with overall acceptability.

Table 7: Mean carcass sensory attributes of lowland sheep breeds (Gumuz, Rutana and
Begiet) supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C) P value
Sensory attributes Gumuz Rutana Begiet L1 L2 L3 B C B*C
Juiciness 3.92 3.94 3.84 3.74 4.15 3.81 0.8144 0.0659 0.1441
Overall tenderness 3.9 3.83 3.7 3.6 4.04 3.78 0.6296 0.1248 0.7270
Flavor intensity 3.77 3.7 3.58 3.42b 4.11a 3.53b 0.6261 0.0071 0.3719
Connective tissue amount 3.96 4.02 4.02 3.84 4.29 3.87 0.9254 0.0631 0.5395
b
Overall acceptance 3.9 3.69 3.7 3.52 4.15a 3.63b 0.2409 0.0005 0.1902
*a, b = Means within the same row not bearing a common superscript letters differ
significantly; P = probability; L1 = 400g Supplement (75% NSC: 25% WB); L2 = 500g
Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75% NSC: 25% WB).

63
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Partial Budget Analysis

The result of the partial budget analysis was shown in Table 8. The result of the partial
budget analysis revealed that sheep fed on concentrate L2 and use of Gumuz and Rutana
sheep breeds resulted in a higher profit margin. Generally, sheep supplemented with an
optimum level of concentrate (L2) had a required CP intake, and the use of adapted sheep
breed (Gumuz and Rutana) had superior ADG as a result of this, had a higher sale price to
earn a higher net return. The marginal rate of return for concentrate L2 supplemented sheep
was 14.12 ETB and for Gumuz and Rutana sheep breeds were 10.25 and 11.42 ETB,
respectively. This indicates that to attain the required BW by the use of a required level of
supplement feeding and by the use of adapted sheep breed, each additional unit of 1 ETB
increment per sheep to purchase supplement feed resulted in a profit of 14.12 ETB for
concentrate L2, 10.25 ETB for Gumuz and 11.42 ETB for Rutana sheep breed, respectively.

Table 8: Partial budget analysis results of lowland sheep breeds (Gumuz, Rutana and Begiet)
supplemented with different levels of concentrate at Gende wuha
Sheep breed (B) Concentrate level (C)
Variables Gumuz Rutana Begiet L1 L2 L3
Number of animals 9 9 9 9 9 9
Purchase price of sheep (ETB head-1) 1650 1915 1824 1863 1863 1863
Total feed consumed (kg head-1) 50.72 50.85 50.92 53.21 52.4 46.89
Total variable cost (ETB) 465.58 465.33 464.86 417.63 470.94 507.25
Gross income (ETB) 2849.00 2912 2284 2383 3189 2542
Total return (ETB) 1199.00 997 460 520 1326 679
Net return (ETB) 733.42 531.67 -4.86 102.37 855.06 171.75
MRR (ratio) 10.25 11.42 - - 14.12 0.77
*ETB = Ethiopian Birr; MRR = marginal rate of revenue; L1 = 400g Supplement (75%
NSC: 25% WB); L2 = 500g Supplement (75% NSC: 25% WB); L3 = 600g Supplement (75%
NSC: 25% WB).

CONCLUSION AND RECOMMENDATIONS

The study demonstrated the effects of genotype and levels of concentrate on growth, carcass
yield and quality of selected indigenous sheep breeds at Gende wuha. Generally, the present
study indicated that Gumuz and Rutana sheep breeds can achieve better daily gain than
Begiet. This is due to their adaptation to the area these breeds have had the potential to
achieve faster growth than Begiet. Concentrate level two (L2) provided optimum level of
nutrients to support maintenance and growth compared to other levels and resulted in higher
FCE.

64
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Among the three genotypes, Rutana sheep had better FCE than other breeds. As a result of
better SW achieved higher HCW was recorded by Gumuz and Begiet. Because of greater
difference in the proportion of non-carcass components and better HCW achieved relative to
slaughter body weight, better dressing percentage (DP) was achieved by Gumuz sheep breed.
Breed and concentrate level did not improve main carcass components, edible and non-
edible offals. Meat quality attributes were not influenced by breed. Among the concentrate
levels, L2 was optimum to improve major meat quality attributes like juiciness, flavor and
overall acceptance. Moreover, it was concluded that supplementation of sheep with the
required level of concentrate (L2) and the use of proper sheep breed (Gumuz and Rutana) is
potentially profitable in feeding of growing lowland sheep compared to other concentrate
levels and other breeds and could be recommended. Hence, according to the results of this
study for better animal performance and to produce quality meat, feeding of growing Gumuz
and Rutana sheep and use of concentrate (L2) were biologically efficient and profitable.

ACKNOWLEDGMENT

The authors gratefully acknowledged the Ethiopian Institute of Agricultural Research

(EIAR), for financing this research project.

REFERENCES

Alemu Tarekegn, Yeshambel Mekuriaw and Firew Tegegne. 2018. Growth and Carcass
Characteristics of Gumuz Sheep Supplemented with Different Proportions of Noug Seed
(Guizotia abyssinica) Cake and Wheat Bran. In: Mesfin Lakew, Likawent Yeheyis and
Mulugeta Alemayehu (Eds.). 2018. Proceeding of the 10th Annual Regional Conference
on Livestock Completed Research Activities 13-16 March, 2017, Amhara Agricultural
Research Institute, Bahir Dar, Ethiopia, Pp. 41-52.
Asfaw Negassa, Shahidur R. and Berhanu Gebremedhin. 2011. Livestock Production and
Marketing: The Ethiopia Strategy support Program (ESSP) Working Paper 26.
International Food Policy Research Institute, Addis Ababa, Ethiopia.
CSA (Central Statistics Authority). 2016/17 (2009 E.C.). Report on livestock and livestock
characteristics (private peasant holdings), Statistical Bulletin, 585, Volume 2, Addis
Ababa, Ethiopia.
Fasae A., Adu I. F., Aina A. B. J. and Dipeolu. M. A. 2010. Growth performance, carcass
characteristics and meat sensory evaluation of West African Dwarf sheep fed varying
levels of maize and cassava hay. Tropical Animal health and Production. 43, 2
Fentie Bishaw and Solomon Melaku. 2008. Feed Utilization and Live Weight change of Farta
sheep Supplemented with noug seed (Guizotia abyssinica) cake, wheat bran and their
mixtures, http://www.springerlink.com/index/64685366r4427 p 22, Retrieved on 15 Jan
2019.

65
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Flint, M. 2005. Management guide for lot feeding goats. http//www.dpi.qld.gov.au/sheep

/15153.html. Accessed on 10 January 2019.
Gilmour, A. R., Cullis, J. J., Fogarty ,N. M. and Banks, R. 1994. Genetic parameter for
ultrasonic fat depth and eye muscle measurement in live Poll Dorset sheep. Aust. J. of
Agric. Res. 45: 1281-1291.
IAR (Institute of Agricultural Research). 1991. Strategic Planning for Livestock Research in
the Institute of Agricultural research (IAR). Addis Abeba, Ethiopia.
ILRI (International Livestock Research Institute). 2005. Metema pilot learning site diagnosis
and program design. ILRI, Addis Ababa, Ethiopia.
Iwe, M. O. 2002. Handbook of sensory methods of analysis, 1st Ed. Rejoins communication
services limited, 70-71.
Jadish, P. 2004. Goats, Sheep and Pig production and management. 3rd (edition). Kalyam
Pub, New Delhi, India, 451p.
McDonald, R. E., Edward, R. A., Greenhalgh, J. F. D. and Morgan, G. A. 2002. Animal
nutrition. 6th edition, Pearson Educational Limited, Edinburgh, Great Britain.
Mulu Moges, Berhan Tamir and Alemu Yami. 2008. The Effects of Supplementation of
Grass Hay with Different Levels of Brewer’s Dried Grain on Feed Intake Digestibility
and Body weight Gain in Intact Wogera Lambs. East African Journal of Sciences.
Volume 2 (2) 105-110.
PRIMEFACT 340. 2007. Dressing percentages for cattle. www.dpi.nsw.gov.au , Accessed on
2 September 2018.
SAS. 2003. SAS user’s guide: statistics. Version 9.1. Carry, North Carolina. USA. Statistical
Analysis System Inc.
Solomon Bogale, Solomon Melaku and Alemu Yami. 2008. Potential use of crop residue as
livestock feed resource under smallholder farmers’ conditions in bale highlands of
Ethiopia. Tropical and subtropical Agro ecosystems. 8, 107-114.
Swan, J. E., Esguerra, C. M. and Farouk, M. M. 1998. Some physical, chemical and sensory
properties of chevon products from three New Zealand goat breeds. Small Ruminant
Research 28 (3): 273–80.
Tadesse Bekele, Zelalem Yilma, Yohannes Gojjam and Alemu G/Wold. 2003. Nutrient
utilization efficiency of lactating cross bred dairy cows supplemented with different
levels of concentrate diets. In: Proceedings of the 11th National Conference of Ethiopian
Society of Animal Production (ESAP). 21-23 August 2003, Addis Ababa, Ethiopia, pp.
341-345.
Torell, R. and Suverly, N. 2004. Navada Market Lamb Carcass Merit Program. Fact Sheet-
04-02. http://www.meatupdate.csiro.au/data/Advances in Meat Tech III 07-82. pdf.
Accessed on 11 December 2018.
Upton, M. 1979. Farm management in Africa: The principle of production and planning.
Oxford University press, UK, 282-298 pp.
William, J., Payne, A. and Wilson, R. T. 2003. An introduction to animal husbandry in the
tropics. 5th Edition, 815p.

66
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Effect of Pigeon Pea Dried Leaves and Concentrate Mix Supplementation on the
Growth, Feed Utilization, Carcass Yield and Market Weight of Central Highland Goats

Solomon Tiruneh1, Zeleke Tesema1, Bimrew Asmare2 and Firew Tegegne2

1
Sirinka Agricultural Research Center, P.O. Box 74, Woldia, Ethiopia
2
Bahir Dar University, College of Agriculture and Environmental Sciences, P.O. Box 5501,
Bahir Dar, Ethiopia

ABSTRACT

The study was conducted to evaluate intake, digestibility, market weight and carcass
characteristics of male Central Highland goats (CHG) fed different proportions of pigeon pea
leaves hay and Noug seed cake as a protein supplement. A total of 24 yearling intact males
with an initial mean body weight of 21.4 ± 1.12 kg were used for this study. The experiment
was conducted in a randomized complete block design (RCBD) using six goats per treatment.
The experimental goats were blocked based on their initial body weight and divided into four
treatment groups. The dietary treatments were T1 = grass hay ad libitum + 100% Noug seed
cake, T2 = grass hay ad libitum + 65% Noug seed cake + 35% pigeon pea hay, T3 = grass
hay ad libitum + 35% Noug seed cake + 65% pigeon pea hay and T4 = grass hay ad libitum +
100% pigeon pea leaf hay on a dry matter (DM ) base. The total DM intake for T3 goats
(902.3 g/day) was higher (P<0.05) than that of other treatment groups. Besides, goats in T3
and T4 also fed the highest (P<0.01) average NDF (611.7 g/day). However, the lowest ADF
intake (335.6g/day) was recorded for T1. Consequently, the highest average daily weight gain
(77.8 g/day) and total weight gain (6.5 kg/84 days) were observed for goats kept in T3 diet.
The feed conversion ratio (FCR) among treatments was found to be similar (P>0.05). The
average dressing percentage (empty weight basis) ranged from 54.25 to 54.94% and the
dressing percentage (slaughter weight basis) was between 45.25 and 47.6%. Central Highland
goats attained a market weight (25 - 30 kg) within 84 days of feeding using pigeon pea leaves
as a protein supplement. Therefore, supplementation of Central Highland goat with 65% of
pigeon pea and 35% of Noug seed cake as protein supplements mix in addition to 50% Wheat
bran was found to be a very promising supplement.

Keywords: Carcass, Central Highland goat, intake, market weight, weight gain

67
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Ethiopia has different agro-ecological zones ranging from semi-arid to arid environments
which are suitable for goat production. The goat population of Ethiopia is estimated at
30.2 million (CSA, 2016). Goats play important roles in communities by improving the
livelihood of resource-challenged farmers by creating alternative employment opportunities,
enhancing family income by the sale of live animals, skin and manure. Moreover, goats act as
an income buffer to the risks associated with erratic climatic changes (Rumosa et al., 2009),
and they also contribute to human nutrition in the form of meat and milk.

Despite their importance, the productivity is low due to different technical and non-technical
constraints. Feed shortage is one of the technical constraints and feed production covers
requirements only in exceptionally good years; the deficit reaching 35% in normal years and
70% in bad years (FAO, 2005). The situation is exacerbated during the dry season when
natural pastures usually dry out and are overgrazed, resulting in low contents of protein and
energy in the fodder. Therefore, it is important to seek different feed sources that could be
used as dry season supplements for goats.

Previous works have reported that leaves of multipurpose trees growing in the backyards of
the farmers can be used as a cheap source of protein for supplementing livestock (Tibebu et
al., 2009; Denbela et al., 2018). One such fodder legumes is pigeon pea (Cajanus cajan)
whose leaves are an important fodder and serves as a valuable source of feed for farm
animals (Foster et al., 2009). Pigeon pea leaf is an excellent source of protein ranging from
16.3 - 27% (Alexander et al., 2007). In addition, pigeon pea remains one of the most drought-
tolerant and adaptable legumes in the tropics (Netsanet and Yonatan, 2015). It is therefore
important to supplement the major available fodder with some amount of concentrates in
order to improve the intake and digestibility of such poor quality feed resources. Therefore,
this study was conducted to evaluate the intake, digestibility, market weight and carcass
characteristics of male Central Highland goats (CHG) fed different proportion pigeon pea
leaves hay and noug seed cake as a protein supplement.

68
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of the Study Area

The study was conducted at Sirinka Agricultural Research Center (SARC) research site,
which is located 508 km away from Addis Ababa. The site is located at an altitude of 1850
meter above sea level and at 11˚45'00''N, 39˚36'36''E. The rainfall pattern is bimodal, with the
two-rainfall season, Belg (Feburary/March - April) and Meher (July – October/November)
and the mean annual rainfall amount is on average about 950 mm. The area is a moderately
warm temperature zone with mean daily temperature ranges from 16 – 21oC.

Experimental Animals and Their Management

A total of 24 (9 - 12 months age) intact male Central Highland goats were purchased from
local markets. The animals were quarantined for 21 days before adaptation started. All the
goats were ear-tagged, vaccinated for the protection of common goat diseases and treated
against internal and external parasites during quarantine time. The experimental animals were
randomly allocated into four treatment groups and each group contains six animals. The
treatment feed gradually introduced into the animals’ ration after 15 days of adaptation
period. The ration used in this experiment consisted of pigeon pea leaf hay, Wheat bran,
Noug seed cake and salt. Goats were offered 3.0% of their body weight per head according to
the recommendation of Kearl (1982). All the experimental goats had got natural grass hay
and water ad libitum throughout the experiment period.

Feed Preparation

Pigeon pea leaves were harvested from the Jari trial site. The re-growths were harvested at
about 50% flowering stage, air-dried and stored for feeding trials. The grass was harvested at
a 10% heading stage from the research center pasture area, dried and stored in recommended
moisture content. Concentrate feeds were purchased from the factory and mixed according to
the treatment settled. Samples of all feeds were taken and oven-dried for 65°C for 72 hrs to
determine the dry matter percentage of the feeds.

69
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Experimental Design and Treatments

The experiment was conducted in a randomized complete block design (RCBD). Goats were
grouped into six blocks based on their live weight, and one goat from each block was
randomly assigned to one of the four treatments. Feed allowed were adjusted weekly based
on the live weight of the animals The treatments were grass hay ad libitum + 50% WB + 50%
Noug seed cake (T1), grass hay ad libitum + 50% WB + 32.5% Noug seed cake + 17.5%
pigeon pea hay (T2), grass hay ad libitum + 50% WB + 17.5% Noug seed cake + 32.5%
pigeon pea hay (T3) and grass hay ad libitum + 50% WB + 50 % pigeon pea leaf hay (T4) on
DM base.

Feed Intake and Live Weight Measurement

The supplementations were continued until the entire treatment group goat’s attained live
weight within the range of 25 - 30kg (Endashaw et al., 2013). The total amounts of feed
offered and refused were daily measured to determine the quantity of feed intake throughout
the experimental period. Every morning the refusals were collected, weighed on an individual
basis and the samples were bulked for laboratory analysis. The body weight (BW) was
measured every seven day interval with a suspended weight balance of 200g precision during
the morning. Feed conversion efficiency was calculated as the proportion of daily weight gain
in gram and feed intake in gram. At the end of the feeding trial, the value of each animal was
estimated at the prevailing market price.

Digestibility Trial

The digestibility was done after the feeding trial was completed. The animals were adapted to
the attached fecal bags for three days followed by a digestibility trial for seven days. The
amount of voided feces per goat/day was weighed and sub-sampled 20% of the daily
collected. About 10% of the total sample was taken and placed in the freezer at -20oC until
the end of data collection. Samples of feces were dried and milled using a cross-beater mill
through 1mm sieve and kept in plastic bottles until analysis. The apparent digestibility
coefficient (DC) was calculated as:

Nutrient in feed − Nutrient in faeces

Apparent Digestibility (%) = × 100
Nutrient in feed

70
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Carcass Evaluation

Carcass evaluation was done after the feeding and digestibility trials were done. After
overnight fasting, all goats were slaughtered for carcass analysis. Prior to slaughter, body
weight was measured to determine the slaughter weight. The empty body weight of each
animal was determined by subtracting the weight of gut content from the slaughter weight.
Hot carcass weight was computed by excluding the contents of thoracic, abdominal, pelvic
cavities, head, skin with feet, and tail of the animal. Dressing percentage was calculated as
the proportion of hot carcass weight to slaughter and empty body weight. The weights of
edible and non-edible components were recorded. Each carcass was cut between the 12th and
13th ribs perpendicular to the backbone to measure the cross-sectional area of the rib-eye
muscle area.

Chemical Analysis of Samples

Samples of feed offered, refusals and feces were subjected to the analysis of dry matter (DM)
and ash according to AOAC (1990). The neutral detergent fiber (NDF), acid detergent fiber
(ADF) and acid detergent lignin (ADL) were determined according to Van Soest and
Robertson (1985). The crude protein (CP) content was calculated by multiplying nitrogen
content with a factor 6.25.

Partial Budget Analysis

Before the slaughtering of goats, their selling price was estimated. Wheat bran, Noug seed
cake and goat purchasing and pigeon pea collection and preparation costs were considered as
a variable cost. The total return was obtained by calculating the difference between the
estimated selling prices and the purchasing price of an experimental goat. The net return was
calculated according to Ibrahim and Olaloku (2000) as follow:

∆NI = ∆TR − ∆VC

Where, NI = net income; TR = total return; VC = variable cost.

71
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Statistical Analysis

Data were analyzed using the general linear model procedure of the Statistical Analysis
System (SAS, 2002). If there was a significant difference between treatments at a 5%
significance level, mean separation among treatments was done using the Tukey HSD tests.
The model was presented as follow:

Yijk = µ + Bi + Tj + eijk

Where; Yijk is response variables such as average daily gain (ADG), DM intake, feed and
conversion efficiency (FCE); μ is overall mean; Bi is block effect; Tj is effect of the jth dietary
treatment; eijk is random error.

RESULTS AND DISCUSSION

Chemical Composition of Feeds

The chemical compositions of grass hay and the supplements used in the current study are
shown in Table 1. The concentrate feeds were low in fiber and rich in CP. The ADF and NDF
content of wheat bran were lower than any of the other ingredients used to formulate the
supplemental diet. The grass hay had better CP content (70g/kg DM) which can fulfill the CP
required to support optimal microbial activity in the rumen (McDonald et al., 2010). The CP
content of the grass hay is comparable with the report (9.9 to 10.1%) of Wondwosen et al.
(2010). However, the CP content of Noug seed cake in the present study (31.9%) is higher
than the report (28.9%) of Wondwosen et al. (2010). Based on its CP content, the grass hay
used in this experiment could support adequate rumen microbial function or maintenance
requirement of adult ruminants and may also support optimum DM intake, since roughage
feeds with more than 70 - 80 g CP/kg DM do not depress intake and digestibility of feeds
(Van Soest, 1994).

Table 1: Chemical composition (%) of experimental feeds used in the current study
Experimental feeds DM Ash CP NDF ADF
Grass Hay 89.0 12.3 9.60 64.4 30.2
Pigeon Pea 89.0 10.1 24.5 53.3 29.3
Wheat bran 88.0 9.10 15.9 30.1 23.6
Noug Seed Cake 89.0 10.1 31.9 42.2 30.3
DM = dry matter, CP = crude protein, NDF = neutral detergent fiber, ADF = acid detergent
fiber

72
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Dry Matter and Nutrient Intakes

There were significant (P<0.05) differences in dry matter intake, crude protein, neutral
detergent fiber and acid detergent fiber intake among the treatments (Table 2). The highest
average DMI (902.3 g/kg DM) and crude protein intake (160.5 g/day) were recorded for T3.
The daily DMI proportion to live weight of the Central Highland goat is within the range of
2.6 – 3.2% reported by William et al. (2013) for Small East African goats and their crosses
with Norwegian goat fed hay and supplemented with concentrate. The average DMI (902.3
g/day) obtained in the present study is greater than the result (822.09 g/day) reported for Boer
x Woyto-Guji goats supplement with commercial concentrate with Acacia nilotica pod meal
(Denbela and Deribe, 2017).

Table 2: Effect of experimental diets on feed intake (g/day) and growth performance of goats
Parameters T1 T2 T3 T4 SEM SL
b b a b
TDMI(g)/d/head 722.2 783.3 902.3 786.4 22.5 ***
c b a b
TCPI(g)/d/head 119.0 142.6 160.5 139.2 5.28 ***
TNDFI(g)/d/head 474.8c 535.5b 611.7a 582.7a 22.7 ***
b a a a
TADFI(g)/d/head 266.7 332.0 335.6 342.0 9.58 ***
ab
means within a row not bearing a similar superscript letter significantly differ; *** = P < 0.001;
SEM = standard error of the mean; SL = significant level; DMI = Dry Matter Intake; CPI = Crude
Protein Intake; NDFI = Neutral Detergent Fiber; ADFI = Acid Detergent Fiber

Dry Matter and Nutrient Digestibility

Apparent DM digestibility coefficient (% DM) of Central Highland goat is presented in Table

3. T2, T3 and T4 had higher CP, NDF and ADF apparent digestibility than T1. The DM
digestibility in this study was ranged from 67.72 to 70.50% and this finding is agreed with the
result (45.02 - 69.89%) reported for Boer x Woyto-Guji goats supplemented with Acacia
nilotica pod meals (Denbela and Deribe, 2017). However, the digestibility of CP (71.24 -
72.54%) is smaller than the result (83.06%) reported for Boer x Woyto-Guji goats
supplemented with Acacia nilotica pod meals (Denbela and Deribe, 2017) and the result
(74.9%) reported for Central Highland goat fed concentrate mix (Mekonnen et al., 2014).

73
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: Effect of dietary treatments on nutrient digestibility

Digestibility (%) T1 T2 T3 T4 SEM SL
DM 67.9 67.7 70.5 69.3 0.83 NS
CP 68.1b 71.2a 72.4a 72.5a 1.09 *
NDF 62.6b 67.9a 67.8a 68.9a 1.06 ***
ADF 53.2c 63.3a 58.6b 60.8ab 1.37 ***
a, b, c
mean within a row not bearing similar superscript letters significantly differ. * = P<0.05; *** =
P<0.001; NS = not significant; SEM = standard error of the mean; SL = significant level

Weight Gain and Feed Conversion Efficiency

The average daily weight gain and total weight change were significantly different (P<0.05)
among treatments (Table 4). Goats in T3 had the highest ADG and TWG than goats in the
other treatments. This superiority may be due to higher DM and CP intake of goats in this
treatment. The weight gain of Central Highland goats in the present study is higher than the
value (32.1 g/d) reported for the same breed by Mekonnen et al. (2014), 42.1 g/day reported
for Central Highland goat feed concentrate mixture (Dereje et al., 2016) and higher than the
value 33 - 49 g/day reported for Harrege Highland goats fed different levels of leftover khat
(Mesignaw et al., 2012). The observed better performance might be due to higher CP intake
from the supplements.

The total body weight change through 84 day feeding period for T3 in the current study was
6.5 kg which is in line with the result (6.25 kg) reported for Boer x Woyto-Guji goats
supplemented with Acacia nilotica pod meal (Denbela and Deribe, 2017). The feed
conversion efficiency (FCE) of Central Highland goats was found to be similar (P>0.05)
among treatments.

Table 4: Weight gain and feed efficiency of Central Highland goat

Parameters T1 T2 T3 T4 SEM SL
Initial weight (kg) 21.6 21.2 21.3 21.6 1.12 ns
Final weight (kg) 25.2 25.5 28.6 26.2 1.26 ns
Total weight gain (kg) 3.00b 4.90ab 6.50a 3.90b 0.69 *
ADG (g/day) 35.7b 58.5ab 77.8a 47.0b 8.28 *
FCE (g ADG/g DDMI) 6.77 5.95 7.64 3.90 1.24 ns
FCE = Feed conversion efficiency; ADG = Daily body weight gain; DDMI = Daily dry matter intake

74
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Carcass Characteristics

All carcass parameters among treatments were found to be similar (P>0.05) (Table 5).
According to Belay and Mengistie (2013), the HCW of Central Highland goat feed
concentrates mix was 12.5 kg which was comparable with the current study. But the EBW
(24.9 kg) reported by the same author was greater than the current study (20.42 - 22.28 kg).
The average DP as a percentage of the empty weight of Central Highland goats in this study
(54.25 - 54.94%) was comparable to the report of Dereje et al. (2016) for three Ethiopian
indigenous goat breeds (51.2 - 54.4%) fed concentrate mix and greater than the DP reported
for Sidama goat (49.3 - 51.2%) and Central Highland goats (40.5%) (Wondwosen et al.,
2010; Belay and Mengistie, 2013).

Rib-eye area of Central highland goat recorded under this study ranges from 11.2 to 12 cm2.
The result obtained in the current study was higher than the result (6.9 cm2) noted for the
Central Highland goat feed concentrate mix. According to the report of Wolf et al. (1980)
(Cited in Dereje et al., 2016), a larger rib-eye area is associated with higher production of
lean in the carcass and higher lean to bone ratio. This indicates the high proportion of lean in
the Central Highland goat.

Table 5: Carcass characteristics of male Central Highland goats

Carcass traits T1 T2 T3 T4 SEM SL
Slaughter Weight (kg) 23.9 23.8 25.7 24.5 1.22 NS
Empty Body Weight (kg) 20.5 20.4 22.3 20.5 1.01 NS
Hot carcass Weight (kg) 11.1 11.2 12.3 11.1 0.89 NS
Dressing percentage
Slaughter weight basis (%) 46.5 46.7 47.6 45.2 2.40 NS
Empty body weight basis (%) 54.2 54.6 54.9 54.3 4.00 NS
Carcass primal cuts
Hind quart (kg) 4.20 4.40 4.60 4.50 0.22 NS
Fore quart (kg) 6.20 6.20 7.20 6.40 0.48 NS
Leg (kg) 3.50 3.60 3.80 3.70 0.21 NS
Loin (kg) 0.76 0.78 0.80 0.75 0.06 NS
Leg + Shoulder + Neck (kg) 4.20 4.10 4.70 4.20 0.32 NS
Rib + Chest (kg) 2.10 2.10 2.40 2.20 0.19 NS
Thin cut (g) 684.5 645.5 629.2 568.2 54.8 NS
2
Rib eye area (cm ) 11.3 11.4 11.2 12.0 6.88 NS
NS = not significant; SEM = standard error of mean; SL = significant level

75
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Non-Carcass Components

Table 6 shows non-carcass components of Central Highland goat feed pigeon pea leaves and
concentrate mixture. The effects of feeding pigeon pea and concentrate mix were not shown
to have a significant difference (P>0.05) for all non-carcass parameters except for blood in
which higher blood weight was observed for T1 and T3. This result was comparable with the
result (1010 g) of a similar breed feed concentrate mix (Mekonnen et al., 2014).

Table 6: Non-carcass components of Central Highland goats

Parameters T1 T2 T3 T4 SEM SL
Kidney (g) 77.7 87.1 82.7 83.7 19.93 NS
Head (g) 1449.5 1365.8 1548.8 1505.8 71.65 NS
Liver with bile (g) 536.8 544.2 534.6 469.7 28.67 NS
Heart(g) 131.0 130.6 143.2 122.1 8.48 NS
Empty gut (g) 825.8 867.6 857.5 846.4 29.09 NS
Blood (g) 1042.5a 969.1ab 1036.5a 898.9b 37.02 *
Total edible offal 4063.3 3964.5 4203.4 3976.7 0.182 NS
Full gut (kg) 4.30 4.30 4.30 4.90 0.33 NS
Intestine (kg) 2.10 2.40 2.00 2.10 0.09 NS
Kidney fat (g) 85.0 58.0 55.9 73.3 19.93 NS
Abdominal fat (g) 262.8 235.2 244.9 147.1 51.38 NS
a, b
means within a row not bearing a similar superscript letter significantly differ. * =P<
0.05; NS = not significant; SEM = standard error of mean; SL = significant level

Number of Days Required to Attain Market Weight and Economic Analysis

The study revealed that Central highland goats managed under intensive management
achieved market weight requirement within 84 days of feeding (Figure 1). However, a longer
period of time (168 days) than the current study was reported for intact yearling Afar goats
under semi-intensive feeding management supplemented with concentrate mix and lucenea
leaf (Endashaw et al., 2013). This might be due to small feed intake, the difference in nutrient
composition and management systems. Feeding of yearling Central Highland goats with a
proportion of 65% dried pigeon pea leaves and 35% of Noug seed cake as a protein source
and wheat bran as an energy source resulted in the highest net income (254.84 ETB).

76
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

T1 T2 T3 T4
28.0
27.0
26.0
Body weight (kg)

25.0
24.0
23.0
22.0
21.0
20.0
W0 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W11 W12
Feeding period (week)

Figure 1: Growth trend of Central Highland goat supplemented with different levels of
pigeon pea leaf hay as a protein source

CONCLUSIONS

The experimental feeds used in this experiment had high nutritional qualities that were able to
support the maintenance level. Central Highland goats managed under intensive conditions
fed a maximum of 3.1% of their body weight per day on a dry matter base. A high amount of
CP, NDF and ADF intake in this study had no adverse effect on the total intake of goats.
Goats fed 65% pigeon pea and 35% of Noug seed cake as a protein source had the highest
amount of DMI. Moreover, they attained the market live weight (25 to 30 kg) at yearling age
when they fed for 84 days. The higher value for the ribeye area of Central Highland goat
indicates the quality of the carcass produced from them. Most of the parameters were not
significantly different among treatments. However, based on the intake, weight gain and
partial budget analysis, supplementation of goats with 65% of pigeon pea and 35% of Noug
seed cake as a protein source is preferable. Further research on the effect of feeding pigeon
pea dried leaves on meat quality of goats should be done.

ACKNOWLEDGMENTS

The authors would like to thank ARARI for funding this research work and all staff of
livestock research directorate in Sirinka for their valuable contribution to this study. We are
very thankful to Mr. Asfaw Bisrat, Mr. Tesfaye Zewdie and Mr. Ashenafie from Debre
Birhan Agricultural Research Center for their cooperation during carcass evaluation and
chemical analysis of samples.

77
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Alexander G., Ravi D., Reddy R.Ch, Saxena K.B., Hanson J., Upadhyaya H.D. and Blümmel
M. 2007. Forage yield and quality in pigeon pea germplasm lines. An open access
Journal. 3(1).
AOAC. 2005. Official methods of analysis, Association of Official Analysis Chemist,
Washington D.C. 15th Ed.
Belay Deribe and Mengistie Taye. 2013. Growth performance and carcass characteristics of
central highland goats in Sekota District, Ethiopia. Journal of Agricultural Advances.
2(8): 250-258.
CSA (Central Statistical Agency). 2016. Livestock and livestock characteristics (private
peasant holdings), federal democratic republic of Ethiopia, agricultural sample survey.
Statistical bulletin 583 volume II, August 2016 Addis Ababa, Ethiopia.
Denbela Hidosa, Adugna Tolera and Ajebu Nurfeta. 2018. Effect of lablab and pigeon pea
leaf meal supplementation on performance of goats fed a basal diet of haricot bean
haulms. Tropical Animal Health and Production https://doi.org/10.1007/s11250-018-
1554-2.
Denbela Hidosa and Deribe Gemiyo. 2017. Replacement of Commercial Concentrate with
Acacia nilotica Pod Meal on Feed Intake, Digestibility and Weight Gain of Boer x
Woyto-Guji Crossbred Goats. American Journal of Agriculture and Forestry. 5(6): 192-
197.doi: 10.11648/j.ajaf.20170506.13
Dereje Tadesse, Mengistu Urge, Getachew Animut and Yoseph Mekasha. 2016. Growth and
carcass characteristics of three Ethiopian indigenous goats fed concentrate at different
supplementation levels. Journal of Springer Plus. 5:41.
Endashaw Terefe, Yibra Yaqob, Kidanie Dessalegn, Abebe Tafa, Ashebir Kifle, Weldegebrel
Gebregziabher and Weldegebrel Tesfamariam. 2013. Market weight and carcass
characteristics of intact yearling afar goats under semi-intensive feeding management.
Int. J. Livest. Prod. 4(6): 95-101.
FAO (Food and Agriculture of the United Nations). 2005. Legume trees and other Fodder
Trees as protein sources for Livestock. FAO Animal Production Health. pp. 95-108.
Foster J.L., Adesogan A.T., Carter J.N., Blount A.R., Myer R.O. and Phatak S.C. 2009.
Intake, digestibility, and nitrogen retention by sheep supplemented with warm-season
legume hays or soybean meal. J. Anim. Sci. 87: 2891–2898.
Mc Donald P., Edward R.A., Greenhalgh J.F.D. and Morgan C.A. 2002, Animal Nutrition
book Sixth Edition. Pearson Educational Limited, Edinburg Gate, Harlow. P669.
Mekonnen Tilahun, Kebede Kefelegn, Girma Abebe and Goetsch A.L. 2014. Feed intake,
digestibility, weight gain, and slaughter characteristics influenced by genetic percentage
of Boer in goats and Dorper in sheep in the central highlands of Ethiopia.
Mesiganaw Wallie, Mekasha Y, Urge M, Abebe G. and Goetsch A.L. 2012. Effects of form
of leftover khat (Catha edulis) on feed intake, digestion, and growth performance of
Hararghe Highland goats. Small Rumin Res. 102: 1–6

78
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Netsanet Beyero and Yonatan Kassu, 2015. Participatory Evaluation of Dual Purpose Pigeon
Pea (Cajanus Cajan) Leaves for Sheep Feeding. Journal of Biology, Agriculture and
Healthcare. 5 (13):2015.
Rumosa-Gwaze F.G.; Chimonyo M. and Dzama K. 2009. Communal goat production in
Southern Africa: A review. Trop. Anim. Health Prod.41: 1157–1168.
SAS. 2002. SAS user's guide version 9.1: Statistics. Cary, NC: SAS Institute Inc.
Tibebu Manaye, Adugna Tolera, and Tessema Zewdu. 2009. Feed intake, digestibility and
body weight gain of sheep fed Napier grass mixed with different levels of Sesbania
sesban. Livestock Science. 122(1): 24-29
Van Soest P.J. and Robertson, J.B. 1985. Analysis of Forages and Fibrous Foods a
Laboratory Manual for Animal Science. Cornell University, Ithaca, NY.
Van Soest, P.J. 1994. Nutritional ecology of the ruminant. 2nd Edition, Cornell University
Press, Ithaca, 476.
William A.H., Kifaro G.C., Safari J.G. and Mushi D.E. 2013. Effect of concentrate
supplementation levels on growth and slaughter characteristics of SEA and SEA ×
Norwegian goats under on-farm conditions. Tropical Animal Health and Production
45(8). DOI: 10.1007/s11250-013-0436-x
Wondwosen Alemu, Solomon M, and Adugna T. 2010. Supplementation of cottonseed,
linseed, and noug seed cakes on feed intake, digestibility, bodyweight, and carcass
parameters of Sidama goats. Tropical Animal Health and Production 42(4): 623-31.

79
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Early Finishing of Grazing Crossbred Sheep under Graded Level of Concentrate

Supplementation

Ayele Abebe1, Asfaw Bisrat1, Teferra Mekonnen1, Beneberu Tefera1, Sisay Lemma1, Tesfaye
Zewdie1, Tesfa Getachew1 and Tesfaye Getachew2

1
Drbre Birhan Agricultural Research Centre, P.O. Box 112, Debre Birhan
2
International Centre for Agricultural Research in the Dry Areas, P.O. Box 5689, Addis Ababa

ABSTRACT

The experiment was conducted to evaluate the potential of early finishing ability of Awassi x
Washera, Awassi x Bonga and Dorper x Menz F1 crossbred ram lambs supplemented with
graded level of concentrate on top of grazing. Fifty four ram lambs 7±1 months aged were
used for the study. Randomized complete block design with 3 × 3 factorial arrangement used
for this study and initial body weight was used for blocking. The treatments were grazing +
300 g commercial concentrate CC (T1), 600 g CC (T2), and 900 g (T3). Data were analyzed
using the GLM procedure of SAS and the cost-benefit ratio (CBR) was calculated by
considering the total cost and benefits obtained. The feeding trial was conducted for 90 days
with 15 days of acclimatization period. The body weight gain significantly varied (P<0.001)
among the treatment groups; however, no significant (P>0.05) difference was detected among
breeds. Also, there was no interaction (P>0.05) among breeds and rations. The body weight
gain was linearly increased as the levels of concentrate supplementation increased. The mean
ADG for Awassi × Bonga crossbred at 900 g/d concentrate supplementation was 186.8g.
There was a significant difference (P<0.05) among breeds on some carcass (loin, tail,
hindquarter, skin, and liver) characteristics measured, whereas, levels of concentrated
supplementation had a significant effect on all carcass parameters measured. The CBR for all
treatment is higher than 1, and treatment T3 had the highest value (1.83). The profit margin
revealed that investing in the three levels of feed could keep 50.81%, 61.62%, and 64.67% of
their earnings after covering expenses for T1, T2, and T3, respectively. Hence, an increased
level of concentrate supplementation improves body weight gain and carcass yield of the
studied animals. The current result showed that the possibility of early finishing of crossbred
ram lambs under a semi-intensively managed system to be implemented by organized sheep
fatteners, investors, women, and youth groups.

Keywords: concentrate, cost-benefit, crossbred lambs, finishing

80
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

In Ethiopia, sheep rearing is the major component of livestock production. It generates major
household income for smallholder farmers. However, in most cases, they are sold unfinished
and before attaining marketable weight or kept for a long period for fattening (>2yrs).
Keeping animals for a long period in the farming system increases feed cost and increasing
load on grazing land.

In the past few decades’ efforts have been made to improve the growth performance of local
sheep breeds through selection, crossbreeding, and improved feeding management. Attempts
to rapidly intensify sheep production in Ethiopia include; crossbreeding and improved
nutrition. Commercial fat lamb production involves terminal crossing for the production of
crossbred lambs continuously, which can be sold as prime lambs or finished under short-term
intensive feeding. The terminal crossing provides opportunities to exploit the faster growth
rate of the sire breed and reproduction performance of the local breed. This strategy is also
suitable for conserving the adapted indigenous sheep resources. Usually, yearling male lambs
have been selected for fattening using 400 g concentrate supplementation. However, research
results show the possibility of early finishing of crossbred sheep under concentrate
supplementation (Ayele et al., 2016). Hence, it is important to evaluate fast-growing sheep
breeds, such as crossbred sheep, under a higher level of concentration supplementation.

Objectives

• To evaluate the fattening and early finishing potential of crossbred ram lambs fed on
different levels of concentrate supplementation
• To assess economic feasibility of early finished crossbred ram lambs

MATERIALS AND METHODS

Study Area

The experiment was conducted between May and July 2017 at Debre Birhan Agricultural
Research Centre (DBARC). DBARC is located 120 km from Addis Ababa and it is located at
09º35'45'' to 09º36'45''N latitude and from 39º29'40'' to 39º31'30''E longitude (NSRC, 2006).

81
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Experimental Animal Management

Fifty four ram lambs aged 7±1 (mean±SD) months were used for this study. They were
allowed to graze on natural pasture for 8 h and fed natural pasture hay overnight as a basal
diet. Experimental animals grazed in a group and were delivered half of the concentrate in
their respective experimental groups during the mid-day and the rest supplement was
delivered in the night. Animals in each group of the experiment were fed on a group basis.
All experimental sheep were sprayed against the external parasite and were also dewormed
against the internal parasite as per the protocol of the research center. The concentrate feed
was bought from Addis Ababa (Kality feed processing plant, personal communication) and
the feed had 21% crude protein and 14 MJ/kg metabolizable energy on a dry matter basis.
The feeding trial was conducted for 90 days with 15 days acclimatization period.

Awassi x Washera 50% Dorper X Menz 50% Awassi X Bonga 50%

Figure 1: Experimental sheep breeds

Experimental Design and Treatments

Randomized complete block design with 3 × 3 factorial arrangement was used for this study
and initial body weight was used for blocking. The treatments were supplementation of
different levels of commercial concentrate (CC) to grazing ram lambs; grazing + 300 g CC
(T1), 600 g CC (T2) and 900 g (T3). Data was analyzed following GLM procedure of SAS
(version 9). The model used for the experiment is:

Yijkl = µ + Ti + Bj + Wk+ eijkl

Where: Yijkl is the response variable, µ is the overall mean, Ti is the treatment effect, Bj is the
breed effect, Wk is the block effect (initial body weight) and eijkl is the random error

82
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Body Weight Measurement

Bodyweight was measured at the beginning of the experiment and then after fortnightly. All
bodyweight measurements were taken after overnight fasting. Average daily gain (ADG) was
calculated as the difference between the final and initial BW divided by the number of
feeding days.

Carcass Evaluation

Carcass parameters were analysed at the end of feeding trial. All experimental sheep were
slaughtered after overnight fasting.

Cost Benefit Analysis

All information regarding items and their associated costs were recorded and accordingly the
gross benefit and total costs incurred for respective treatments captured. From such data
profit margins and the cost best ratio (CBR) were calculated.

RESULTS AND DISCUSSION

Body Weight Change

The body weight gain significantly varied (P<0.001) among the treatment groups, however,
no significant (P>0.05) difference was detected among breeds (Table 1 and Figure 2). Also,
there was no interaction (P>0.05) between breed and ration (Table 2). The body weight gain
was linearly increased as the levels of concentrate supplementation increased; indicating the
nutrient requirement of fast-growing crossbred lambs will not be met at a lower level of
supplementation. The mean ADG observed in this study was lower than the value reported
for 50% Dorper crossbred sheep supplemented with a similar level of concentrate (129 Vs
102.3), however, the ADG for 900g supplemented group found in this study was considerably
higher (162 Vs 129 g) (Mekonnen et al., 2014; Tefera et al., 2018). Though there was no
interaction between breeds and levels of concentrate supplementation, the highest ADG was
recorded for Awassi × Bonga crossbred at the level of 900g/d concentrate supplementation.
The mean ADG for Awassi × Bonga crossbred at 900 g/d concentrate supplementation, in the
current study, was 186.8g, ranging between 150 and 218g, which is uncommon for pure and
crossbred sheep in Ethiopia. Since, in this study, the total dry matter intake was not measured
it is difficult to recommend 900g concentrate supplementation.

83
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Estimated least-square means (LSM±SE) of body weight change of grazing Awassi ×
local and Dorper x local crossbred ram lambs supplemented with graded levels of concentrate
Breed IWT (Kg) BCS FWT (Kg) TWG (Kg) ADG (g)
Awassi × Bonga 50% 24.9±1.21 3.3a 35.9±1.50 11.0±0.55 122.7±6.09
Awassi × Washera 50% 24±1.38 2.6b 33.1±1.70 9.3±0.62 103.7±6.92
Dorper × Menz 50% 22.7±1.09 3.7a 33.1±1.34 10.3±0.49 114.8±5.45
SL P>0.05 P<0.05 P>0.05 P>0.05 P>0.05
Level of concentrate
300 g 22.4±1.16 2.3±0.17a 28.6±1.43a 6.2±0.52a 69.3±5.80a
b b b
600 g 24.6±1.25 3±0.18 34.4±1.54 9.9±0.56 109.9±6.27b
c c c
900 g 24.7±1.25 3.9±0.18 39.3±1.55 14.6±0.56 162±6.28c
SL P>0.05 P<0.001 P<0.001 P<0.001 P<0.001
R2 0.08 0.49 0.37 0.72 0.72
CV% 20.96 23.05 18.17 22.25 22.25
ADG: Average daily weight gain; BCS: Body condition score; IWT: Initial body weight; FWT: Final
body weight; TWG: Total weight gain; LS: Level of significance

Table 2: Estimated least-square means (LSM±SE) of effects of different levels of concentrate

supplementation and breed on body weight performance of grazing ram lambs.
Breed Level of concentrate (g) FWT (Kg) TWT (Kg) ADG (g)
Awassi × Bonga 300 27.4±2.3 5.7±0.81 63.1±9.0
600 37.5±2.5 11.4±0.88 126.1±9.7
900 44.6±3.0 16.81±1.07 186.8±11.9
Awassi × Washera 300 30±3.0 5.9±1.07 65.6±11.9
600 33.1±3.0 9.2±1.07 102.1±9.7
900 37.3±2.7 13.1±0.96 145±10.6
Dorper × Menz 300 29.2±2.2 7.3±0.76 80.6±8.4
600 32.5±2.5 9.2±0.87 102.3±9.7
900 37.1±2.3 14.3±0.81 158.9±9.0
CV% 17.7 20.8 20.8
R2 0.34 0.72 0.72
SL P>0.05 P>0.05 P>0.05
ADG: Average daily weight gain; FWT: Final body weight; TWG: Total weight gain; LS: Level of
significance

Carcass

The effect of graded level concentrate supplementation on carcass characteristics of crossbred

sheep grazed on natural pasture hay is presented in Table 3. There was a significant difference
(P<0.05) among breeds on some carcass characteristics measured, whereas, levels of concentrated
supplementation had a significant effect on all carcass parameters measured. There was no
interaction between breed and levels of concentrate supplementation on all carcass characteristics
measured. Only loin (P<0.05), tail (P<0.01), a hind quarter (P<0.05), skin (P<0.001), and liver
(P<0.05) were significantly different among breeds. Though higher hot carcass weight was
recorded for Awassi × Bonga 50% crossbred sheep, no significant difference (P>0.05) among
breeds.

84
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The hot carcass weight found in the current study was considerably greater than reported for
Ethiopian local breeds supplemented with different levels of concentrate supplementation
(Eshete et al., 2013; Getahun, 2014; Shashie et al., 2018). Though it was less compared to
Awassi × Bonga 50% crossbred sheep, the hot carcass weight reported for Menz local sheep
supplemented with different level of dried brewers grain (Nake et al., 2018) and 25% Dorper
crossbred sheep supplemented with forage legumes (Tefera et al., 2018) were comparable
with the mean hot carcass weight of Awassi × Washera 50% and Dorper × Menz 50%
crossbred sheep in the current study. In this study, hot carcass weight increased as the level of
concentrate supplementation increase and the highest hot carcass weight was recorded for
900 g/d/head supplementation. Similarly, Getahun, (2014) reported HCW increased for local
sheep breeds as the level of concentrate supplementation increases.

Hence, the average hot carcass weight (10kg) of Ethiopian sheep breed, which is the second
from the last in sub-Saharan countries (FAO, 2009) could be improved through crossbreeding
and improved feeding management.

Figure 2: Trend of growth among the different treatment groups for all breeds

85
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: Effect of breed and graded levels of concentrate supplementation on carcass characteristics of grazing ram lambs LSM(±SE)

Fore quarter Shoulder and

Breed Dressing (%) HCW (Kg) Loin (Kg) Rib (Kg) Tail (Kg) Hind quarter Skin Liver neck
Awassi × Bonga 50% 43.1 (0.82) 16.4 (0.86) 1.2 (0.07)a 3.0 (0.17) 1.1 (0.13)a 7.6 (0.40) 7.4 (0.40)a 4.2 (0.17)a 0.6 (0.02)a 4.5 (0.22)
Awassi × Washera 50% 41.1 (0.89) 13.9 (0.94) 1.0 (0.07)b 2.6 (0.18) 0.9 (0.15)a 6.6 (0.44) 6.2 (0.43)b 3.7 (0.18)a 0.5 (0.03)b 3.9 (0.24)
Dorper × Menz 50% 42.9 (0.70) 14.3 (0.74) 1.1 (0.06)ab 2.8 (0.14) 0.5 (0.12)b 7.1 (0.34) 5.8 (0.34)b 3.1 (0.14)b 0.5 (0.02)b 4.2 (0.19)
SL P>0.05 P>0.05 P<0.05 P>0.05 P<0.01 P>0.05 P<0.05 P<0.001 P<0.05 P>0.05
Level of concentrate
300 g 38.3 (0.77)b 11.1 (0.80)c 0.8 (0.06)b 2.1 (0.16)b 0.5 (0.13)b 5.5 (0.37)b 4.8 (0.37)c 2.9 (0.16)c 0.4 (0.22)c 3.2 (0.21)c
600 g 43.5 (0.84)a 15.4 (0.88)b 1.2 (0.07)a 3.0 (0.17)a 0.7 (0.14)b 7.3 (0.41)a 6.6 (0.41)b 3.7 (0.17)b 0.5 (0.23)b 4.3 (0.23)b
900 g 45.3 (0.82)a 18.0 (0.86)a 1.3 (0.07)a 3.3 (0.17)a 1.3 (0.14)a 8.5 (0.40)a 7.9 (0.40)a 4.5 (0.17)a 0.6 (0.23)a 5.1 (0.22)a
SL P<0.001 P<0.001 P<0.001 P<0.001 P<0.001 P<0.0001 P<0.0001 P<0.001 P<0.001 P<0.001
Grand mean 42.2 14.5 1.0 2.74 6.96 6.27
R2 0.52 0.49 0.47 0.47 0.46 0.49
CV% 7.57 23.6 24.2 24.0 22.7 25.3

86
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Cost Benefit Analysis

Since there is no significant difference between crossbreeds (Table 1 and 2), the comparison
was done with the three-level of feeds. In this analysis, the cost-benefit ratio for all treatment
is higher than 1, and treatment T3 has the highest value (1.83) which indicates that this
feeding level benefits significantly outweigh its costs. Moreover, investing in this feed level
could expect Birr 1.83 for each Birr 1 of costs (Table 4). The profit margin, which determines
what percentage of revenue keeps, revealed that investing in the three levels of feed could
keep 50.81%, 61.62%, and 64.67% of their earnings after covering expenses for T1, T2, and
T3, respectively. Hence, investing in T3 is better than the other treatments (Table 4).

Table 4: Cost benefit analysis of early finishing performance of crossbred sheep

supplemented with three level of concentrate
Items and their cost Units T1 T2 T3
Average initial price of sheep Birr/head 1500 1500 1500
Average initial weight kg/head 22.4 24.6 24.7
Average final weight kg/head 28.6 34.4 39.3
Average weight gain kg/head 6.2 9.9 14.6
Average daily gain gm/head 69.3 109.9 162
Average hot carcass weight Kg 11.1 15.4 18.0
Selling price of mutton at butcher retails Birr/kg 300 300 300
Gross Benefit (GB) Birr/head 3330 4620 5400
Commercial concentrate for 90 days kg/head 27 54 81
Selling price of commercial concentrate Birr/kg 5 5 5
Cost of commercial concentrate Birr/head 135 270 405
Medication cost Birr/head 3 3 3
Total Cost Birr/head 1638 1773 1908
Net Benefit (NB) Birr/head 1692 2847 3492
Profit margin % 50.81 61.62 64.67
Benefit Cost Ratio (BCR) 1.03 1.61 1.83
Note: Feed levels - T1= grazing + 300 g/head/day, T2= grazing + 600 g/head/day and T3 = grazing +
900 g/head/day

CONCLUSION AND RECOMMENDATION

Increased level of concentrate supplementation improves body weight gain and carcass yield
of Awassi x Bonga, Awassi x Washera and Dorper x Menz ram lambs. The current result
proved the possibility of early finishing of crossbred ram lambs. Farmers, investors, youth
groups, and women in the highland areas could potentially use this early finishing practice
following a semi-intensive management system.

87
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Ayele Abebe, Shambel Besufikad, Tefera Mekonnen, Solomon Gizaw and Abiro Tigabe.
2016. Growth performance of Dorper X Menz 50% Lambs under grazing with
concentrate supplementation. Proceedings of the 8th Annual Regional Conference on
Completed Livestock Research Activities Amhara regional agricultural research
institute, Bahir Dar, Ethiopia. Feb 13-20,2014.
Eshete, T., Gizaw, S. and Seifu, E. 2013. Effect of inclusion of tossign (Thymus serrulatus) in
concentrate mix supplementation on performance and sensory quality of meat of Menz
sheep. Trop Anim health prod 45 (1):177-84 doi:10.1007/s11250-012-0189-y.
FAO. 2009. FAO.STAT Data, Accessed 25 October 2018. http:// www. fao.org/ docrep /012/
i068e. pdf.
Getahun Kebede. 2014. Effect of concentrate supplementation on performance of Ethiopian
Lowland Afar and blackhead Ogaden lambs. Animal and veterinary science 2 (2): 36-41
Hassen Y. Solkner J. and Fuerst-Waltl B. 2004. Body weight of Awassi and indigenous
Ethiopian sheep and their crosses. Small Rumin. Res, 55, 51-56.
Mekonnen T., Kefelegn K., Girma A. and A. L. Goetsch. 2014. Feed intake, digestibility,
weight gain, and slaughter characteristics influenced by genetic percentage of Boer in
goats and Dorper in sheep in the central highlands of Ethiopia, Trop Anim Health Prod,
DOI 10.1007/s11250-013-0532-
Nake Ziku, Tefera Mekonen, Aschalew Tsegahun, Beneberu Tefera, Ayele Abebe and Asfaw
Bisrat. 2018. Effects of Different Proportions of Dry Brewery Grain and Concentrate
Mixture Supplementation on Performance of Yearling Menz Lambs Fed Natural Pasture
Hay as a Basal Diet. In: Solomon Abegaz, Likawent Yeheyis and Mesfin Lakew (Eds.).
2019. Proceedings of the 11th Annual Regional Conference on Completed Livestock
Research Activities April 30 - May 5, 2018, Amhara Agricultural Research Institute,
Bahir Dar, Ethiopia.
NSRC (National Soil Research Council), 2006. Soils of the farms of Debre-Birhan
Agricultural Research centre and its testing sites. Technical report , NSRC, Addis
Ababa, Ethiopia, 57p.
Shashie A., Mengistu U., Getachew A. and Mohammed Y. 2018. Comparative slaughter
performance and carcass quality of three Ethiopian fat-tailed hair sheep breeds
supplemented with two levels of concentrate. Tropical animal health and production, 45
(6).
Tefera Mekonen, Leulseged Alemayehu, Mesfin Eshete, Ashenafi Kebede, Ayele Abebe,
Asfaw Bisrat, Beneberu Tefera, Aschalew Tsegahun, and Tesfa Getachew. 2018.
Potential of Dry Brewery Grain to Replace Commercial Concentrate for Fattening of
50% Dorper Crossbred Sheep on Natural Pasture Based Basal Diet. In: Solomon
Abegaz, Likawent Yeheyis and Mesfin Lakew (Eds.). 2019. Proceedings of the 11th
Annual Regional Conference on Completed Livestock Research Activities April 30 -
May 5, 2018, Amhara Agricultural Research Institute, Bahir Dar, Ethiopia.

88
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Milk Yield and Composition of Fogera Cows Fed with Napier Grass and Concentrate
Feed at Andassa Livestock Research Center

Addisu Bitew1, Mekonnen Tilahun1, 2, Mulugeta Meseret1, Shigdaf Mekuriaw1, 3, Assemu

Tesfa1, Eyasu Lakew1, Yeshwas Ferede1 and Molla Haile1

1
Andassa Livestock Research Center, P.O. Box 27, Bahir Dar, Ethiopia
2
Graduate School of Chinese Academy of Agricultural Sciences, Beijing, China
3
United Graduate School of Agricultural Sciences, Arid Land Research Center, Tottori
University, Japan

ABSTRACT

The experiment was conducted at Andassa Livestock Research Center (ALRC) to evaluate
the effect of Napier grass and concentrate supplementation on feed intake, milk yield, milk
composition, and economic feasibility of Fogera cows. Twenty six cows above second parity
were randomly assigned to three treatment groups; T1 = Grazing plus hay during the dry
season, T2 = Hay plus Napier grass and T3 = Hay plus concentrate feed. The trial was
conducted from calving until drying off dates (April 2018 to February 2019). There was a
significant difference between treatment groups in total dry matter intake. Concentrate
supplemented group had a significantly higher (P<0.01) total dry matter intake than the
Napier grass supplemented group. Milk yield of Fogera cows improved through concentrate
and Napier grass supplementation. Concentrate supplemented groups had significantly higher
(P<0.01) average daily milk yield (4.39 liters) compared to Napier grass supplemented (2.73
liters) and control groups (1.75 liters). Except solid not fat and minerals other parameters of
milk composition had no significant variation at concentrate and Napier grass supplementary
feeding. Concentrate supplemented cows (T3) resulted in a net benefit of 6039.60 ETB and a
cost-benefit ratio of 1.68. Thus, it is recommended that supplementation of concentrate feed
is biologically efficient and potentially profitable in the feeding of Fogera cows.

Keywords: Concentrate, Napier grass, Milk yield, Fogera cows

89
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

In developing countries, cattle production is oriented towards multi-commodity production

systems that produce milk, meat, manure, and draught power. Milk production is dominated
by dual-purpose breeds and restricted suckling systems in which milk and calf production are
combined (Preston and Vaccarro, 1989). Fogera cattle are among the indigenous cattle breeds
of Ethiopia, which are found in the Amhara National Regional State. The breed is well
adapted to the marsh areas of the region around Lake Tana and known by its triple use: as a
source of draught power, meat and milk (Addisu et al., 2010). Fogera cows provide on
average two liters of milk per cow per day at Andassa Livestock Research Center under an
extensive management system. While, in 2012/13, the average milk yields per cow per day at
the national level was about 1.32 liters under low levels of nutrition and management (CSA,
2013).

Poor production may be attributed to the sub-optimal performance of local cattle as a result of
unplanned breeding, inadequate feeding, management and disease control measures. The
environmental factors may suppress the animal's true genetic ability and create a bias in the
selection of animals (Lateef et al., 2008). Breeding programs alone are insufficient to meet
this goal; therefore, animal management at the farm level, including feeding must improve.
The search for strategies that can improve animal performance has been an on-going effort,
and the use of energy or protein supplements is common (Silva et al., 2011). Napier grass is
cultivated at ALRC forage development sites throughout the year (irrigated during the dry
season) to supplement milking cows. In addition, concentrate feed is prepared at ALRC feed
mill and given as supplemental feed for milking cows. However, the level of supplementation
and response in terms of milk yield to these feeds were not studied. Therefore the objectives
of the study were:

• To evaluate the milk yield and milk composition of Fogera cows fed with Napier grass
and concentrate supplementation.
• To evaluate the economic feasibility of feed supplementation on milk yield of Fogera
cows.

90
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of Study Area

The study was conducted at Andassa Livestock Research Center (ALRC), located in Amhara
National Regional State (ANRS), Ethiopia. It is south of Bahir Dar and 22 km far on the way
to Tis Abay. The center is situated 11o29'N latitude and 37o29'E longitude at an altitude of
1730 meters above sea level. The mean annual rainfall is 1434 mm and the mean maximum
and minimum temperature ranged from 27.9 to 13.1oC and humidity ranged from 95%
throughout the rainy season to 35% during the dry season. The topography of the area varies
from river valley plain to gentle slope grassland. In general, the area is characterized by dark
clay soil, which is seasonally waterlogged (Yihalem, 2005).

Animals, Experimental Feeds and Feeding

The experimental animals were twenty-six cows at the second and above lactations that were
selected from ALRC Fogera cattle improvement herd. The cows were checked for major
contagious diseases and drenched with anthelmintic and sprayed against external parasites
periodically. The experimental cows were kept under individual pens. Calves were allowed to
suckle their dams for the first four days to get colostrum, and thereafter they suckled their
dams for about 1-2 minutes to stimulate milk letdown.

Natural pasture hay was harvested from ALRC pasture land, baled and stored for feeding to
Fogera cows. In this pasture land, the dominant grass species were Cynodon, Hyperhenia,
Andropogon, Paspalum, Cetaria, Elusin, Eragrostis, Sporobulus, and Trifolium (Yihalem,
2005). Napier grass was harvested on average at the one-meter height from the forage
multiplication site of ALRC. The Napier grass was irrigated once per week during the dry
season for better biomass production. The Concentrate mixture (maize grain 40%, wheat bran
28%, noug seed cake 30% and salt 2%) was prepared at ALRC feed mill. Animals were given
a fifteen days adaptation period to the experimental house and feed. The experimental cows
were randomly assigned under the three (control, Napier grass, and concentrate
supplemented) dietary treatments.

91
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Treatment feeds used were:

T1 = Grazing + hay during dry season

T2 = Grass hay + Napier grass
T3 = Grass hay + Concentrate mix

The treatment feeds were offered in two equal portions twice a day at 7:00 am and 5:00 pm.
The experimental animals were given water freely. Basal diet (hay) was offered ad libitum
and adjusted daily by allowing 20% of refusals from the previous day’s intake (Treatment 2
and 3). For treatment 1, hay was offered on average 3 kg/cow for the group. A periodic
adjustment of treatment feeds offer was made for each cow as per the average milk
production. The level of experimental feed was adjusted to a half kilogram of concentrate
with one liter of milk and one kilogram of Napier for one liter of milk on a dry matter basis.
Cows were given one kg of concentrate feed above their requirement for an anticipated
increase in milk yield. Cows in treatment 1 were allowed to graze throughout the whole
experimental period. While cows in treatment 2 and 3 were allowed to graze during the wet
season only in the morning and their respective experimental feeds were given in the
afternoon.

Chemical Composition Analysis of Feeds

The feed samples were analyzed for DM (method no. 930.15) according to the procedures of
AOAC (2000). The neutral detergent fiber (NDF) and acid detergent fiber (ADF) were
determined according to the method of (Van Soest et al., 1991). The Crude Protein (CP)
content was calculated as N x 6.25 after determining the total nitrogen and carbon of the
feedstuffs by dry combustion in carbon and nitrogen, CN-coder, Yanagimoto Co. Ash content
was determined by incineration at 600°C for 2 hr in a muffle furnace (AOAC, 1990). The
contents of minerals were analyzed by Inductive Coupled Plasma Mass Spectrometer (ICP
MS) after digestion of the feed samples with a mixture of nitric acids according to the
procedures described by AOAC (1995). All the chemical analysis was conducted at Shimane
University Animal Nutrition Laboratory, Japan.

Milk Composition Analysis

Fresh milk samples of 100 ml taken from each experimental cow using a graduated bottle
were used to analyze milk composition using a Lacto scan at the ALRC animal health
laboratory room.

92
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Partial Budget Analysis

Partial budget analysis was conducted to select a cost-effective feeding technology among the
three feed treatments. The direct market price of milk, concentrate, and labor was used to
compute the partial budget analysis. The real market price of Napier grass and grass hay was
not used while their estimated market prices were taken from the report done by Misganaw et
al. (2018) in the same area. The labor costs for Napier grass harvesting, feeding, milking, and
data collection were considered in the analysis.

Data Collection and Statistical Analysis

Basal and supplementary feeds offered and refused were weighed daily. All the cows were
hand milked twice a day (at 7:00 AM and at 5:00 PM) from calving until drying off dates
(April 2018 to February 2019). Milk yield measurements were taken using graduated jars
during the entire study period. Lactation length ranged from 238 days to a maximum of 329
days. Feed intake, milk yield, and milk composition data were subjected to the analysis of
variance (ANOVA) using the General Linear Model (GLM) procedure of SPSS (2011). The
least significance difference (LSD) test was used to determine any significant difference
between means. The models used for data analysis were:-

Yij = μ + Ti + eij (for feed intake and milk composition).

Yijkl = μ + Ti + Lj + Sk + eijkl (for milk yield)
Where;
Y = the observation on feed intake, milk yield and milk composition
μ = the overall mean
Ti = effect of ith feed treatment
Lj = effect of the jth lactation stage
Sk = effect of the Kth season of lactation
eijkl = random error

RESULTS AND DISCUSSION

Chemical Composition of Experimental Feeds

The chemical composition of grass hay, Napier grass and concentrate mixture used in the
study is shown in Table 1. The chemical analysis results of the present study showed a
considerable difference in CP content between the basal and supplemental diets. The
supplemental diets (Napier grass, 127.8 g/kg DM and concentrate mix, 152.2 g/kg DM) have
better CP content than the basal diet (66.5 g/kg DM).

93
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The grass hay and concentrate feed CP content in the present study were found comparable to
the value reported by Yihalem (2005) and Misganaw et al. (2018) respectively, whereas the
CP content of the Napier grass (12.8% of DM) is higher than the value of CP content (9.9%
of DM) reported by Misganaw et al. (2018) in the same area. The CP content of the grass hay
was less than the critical value (70 g/kg) required by ruminant animals for optimum activity
of rumen microorganisms (NRC, 2001) and thus supplementing animals with other feed
ingredients is essential. However, the Napier grass and concentrate feed that used in the farm
have a higher CP level than the minimum CP level (70 g/kg) required for an adequate rumen
function in ruminants (Van Soest et al., 1991).

The fiber component, NDF and ADF content was higher in the grass hay than the Napier
grass and concentrate feed. The NDF content of grass hay was close to the value reported by
Misganaw et al. (2018), while the NDF content of Napier grass was lower to the value
reported by Misganaw et al. (2018) in the same area, and Muia (2000) in Kenya. The
threshold level of NDF in tropical grass beyond which dry matter intake of cattle is affected
is 600 g/kg (Meissner et al., 1991) suggesting that Napier grass and concentrate feed in the
present study have low, while grass hay has NDF content above the threshold level (Table 1).
The macro and micro mineral composition of the experimental feeds are shown in Table 2.
The concentration of calcium, magnesium, potassium, and phosphorus was sufficiently high
to meet the requirements of the cows in all feed types, while sodium concentration was low
and therefore needed to be supplemented. The micro minerals; iron, manganese, cobalt and
zinc were found in high concentrations in the feed ingredients, while copper was found in low
concentrations (0.7 mg/kg DM in hay to 2 mg/kg DM in grazing pasture). Copper
concentration should be 12 mg/kg of dietary DM to meet the daily copper requirement of a
cow (NRC, 2001).

Table 1: Chemical composition of the experimental feeds

Feed characteristics Grass Hay Napier grass Concentrate Mix
DM (g/kg product) 863.7 973.6 910.4
Crude protein (g/kg DM) 66.5 127.8 152.2
Crude ash (g/kg DM) 140 146 104.4
NDF (g/kg DM) 694 423.12 412
ADF (g/kg DM) 398.4 340.5 177.7

94
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Macro and micro mineral composition of the experimental feeds

Macro Mineral (g/kg DM) Micro Mineral (mg/kg DM)
Feed types Ca K Na Mg P Al Mn Fe Co Cu Zn
Concentrate 5.01 27.71 20.57 7.35 12.98 202 17 60 1.0 1.0 77
Hay 2.82 32.14 0.27 5.63 3.74 188 45 65 1.2 0.7 24
Napier grass 2.80 24.08 0.46 10.14 4.48 355 9 62 1.4 1.1 14
Grazing Pasture 4.06 48.43 1.11 7.08 9.71 594 29 766 5.0 2.0 36

Feed Intake (Kg/Day) of Fogera Cows

Summary of the mean dry matter intake (DMI) (kg/day/cow) of Fogera cows fed on Napier
grass and concentrate mixture is presented in Table 3. Total dry matter intake significantly
(P<0.01) varied among treatment groups. The concentrate supplemented group had
significantly higher total dry matter intake than the Napier supplemented group. However,
there is no significant difference in grass hay dry matter intake between Napier and
concentrate supplemented groups. Similar low total dry matter intake was reported by
Misganaw et al. (2018) in cows given fresh Napier grass ad libitum, which had higher
moisture content, than those given grass hay and concentrate mixture. The cows' lower dry
matter intake may be attributed to the lower dry matter content of the Napier grass.

Table 3: DM feed intake (kg/day/cow) of Fogera cows

Treatment Grass Hay Napier grass Concentrate Total
NS **
1 3.0 ± 0.03 3.0 ± 0.03c
2 3.38 ± 0.03 2.48 ± 0.01 5.86 ± 0.03b
3 3.43 ± 0.03 3.35 ± 0.01 6.78 ± 0.03a
LSD (0.05) 5.81
Means within column with different letters are significantly different; ** = significant at P < 0.01

Milk Yield

There was a significant difference (P<0.01) in milk yield between control (grazing), Napier
and concentrate supplemented groups (Table 4). The concentrate supplemented group had
significantly higher milk yield (4.39 liters/day) than the Napier supplemented (2.73
liters/day) and control groups (1.75 liters/day). The difference in milk yield between
concentrate supplemented and control (grazing) groups is more than double, which indicates
the Fogera cow’s milk yield could be improved through concentrate feed supplementation.

95
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Similar to the present study, Sairanen (2014) reported milk yield of cows increased when the
amount of concentrate supplementation increased. Adebabay et al. (2011) and Sallam et al.
(2017) also reported that supplementation of cows with concentrate feed ingredients resulted
in increment of milk production compared with those left on natural grazing alone. The daily
milk yield of Fogera cows has also been affected by lactation stages and seasons (Table 4).
The average daily milk yield was higher at early and mid-lactation stages and declined at the
late lactation stage. Cows could be dried earlier in the late lactation stage when the level of
milk production is low which will have an effect on the onset and intensity of post parturient
oestrus (Addisu and Hegde, 2003).

The average daily milk yield was higher during the wet season compared to the dry season of
lactation. The difference in milk yield was attributed to the availability of green feed because
cows were allowed to graze on pasture land during the wet season. There was a significant
difference (P<0.01) in 200 days of milk yield between control (grazing), Napier and
concentrate supplemented groups (Table 4). Under the existing management system, Fogera
cows are partially milked (two teats) until calves are weaned at an average age of eight
months. Feed supplementation increased the 200 days milk yield which implies calves will
suckle more milk and therefore will have better growth. This period covered early, mid and
beginning of late lactation stages.

Table 4: Milk yield of Fogera cows under different feeding systems

Treatment Number of records AMY (liters/day) 200 days milk yield (liters)
Overall 6901 2.96 ± 0.01 646.53 ± 29.53
Feeding group ** **
1 (Grazing) 1630 1.75 ± 0.02c 375.75 ± 54.86c
2 (Napier grass) 2411 2.73 ± 0.01b 570.79 ± 47.51b
a
3 (Concentrate) 2860 4.39 ± 0.01 993.04 ± 50.79a
Lactation stage **
1 (Early) 1959 3.40 ± 0.02a
2 (Mid) 2288 2.98 ± 0.02a
3 (Late) 2654 2.50 ± 0.01b
Lactation season **
1 (Dry) 3494 2.66 ± 0.02b
2 (Wet) 3407 3.26 ± 0.02a
CV 23.3
Means within column with different superscripts are significantly different; ** = significant
at P < 0.01, AMY = average milk yield

96
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The Fogera cow’s lactation curve over different feed treatments is shown in Figure 1. The
peak average daily milk yield was attained between weeks 11 to 13, 20 to 22, and 10 to 12 for
treatment 1, 2 and 3, respectively, which declined thereafter. In contrast to this study, Rekik
et al. (2003) reported that multiparous cows reach their peak of production earlier in the
lactation (5 – 6th week of lactation). The highest milk yielding cow was from a concentrate
supplemented group with an average daily milk yield of 5.37 liters/day through a lactation
length of 329 days (Figure 2). This result indicated the presence of elite Fogera cows in the
herd that respond to feed supplementation.

Figure 1: Lactation curve of Fogera cows at different feed treatments

Cow ID 2440
Daily Milk Yield (lts)
Early lactation 6.16
Mid lactation 6.06
Late lactation 4.52
Mean 5.37
Highest 7.9
Lowest 2.9
Lactation milk Yield (lts)
1767

Lactation length (days)

329

Figure 2: Highest milk yielder Fogera cow during the experiment

97
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Milk Composition

The milk composition of Fogera cows that are fed with different feed treatments is presented
in Table 5. No significant variation was observed between feed treatment groups for milk
composition parameters, except solid not fat and minerals. Concentrate supplemented group
had higher (p<0.05) solid not fat and minerals than the Napier supplemented group. In an on-
farm feeding trial conducted in Bure district, Adebabay et al. (2011) reported a significant
difference between control and supplemented group in terms of increased milk-fat and total
solids while treatment effects were not significantly different for increased milk protein,
solids-not-fat, and ash contents.

Table 5: Milk composition of Fogera cows at different feed treatments

Variable Fat Lactose SNF Protein Minerals
Overall mean 4.24±0.07 4.33±0.03 7.86±0.05 2.87±0.02 0.64±0.01
Feed groups NS NS * NS *
ab ab
1 (Grazing) 4.06±0.15 4.33±0.05 7.86±0.09 2.89±0.05 0.64±0.01
b b
2 (Napier grass) 4.22±0.11 4.26±0.04 7.72±0.07 2.82±0.04 0.62±0.01
a a
3 (Concentrate) 4.42±0.12 4.41±0.04 8.0±0.07 2.88±0.04 0.65±0.01
CV 11.8 7.2 7.1 6.9 7.3
Means within column with different superscripts are significantly different; * = significant at P <
0.05

Partial Budget Analysis

The partial budget analysis result showed that T3, T2, and T1 gave an average net benefit
from milk sales 6039.60, 2251.85 and 1994.25 ETB, respectively in two hundred days (Table
6). Concentrate feeding gave the highest benefit when compared to other treatments.

Table 6: Partial budget analysis at different feed treatments of Fogera cows (200 days)
Amount used and costs
T1 T2 T3
Cost Cost Cost
Costs Unit Amount ETB Amount ETB AmountETB
Concentrate Kg 0 0 0 0 670.004690.00
Napier grass Kg 0 0 496.00 1736.00 0 0
Hay Kg 600.00 1800.00 676.00 2028.00 686.002058.00
Labor cost ETB 1842.00 2546.00 2108.00
Total cost (A) ETB - 3642.00 - 6310.00 - 8856.00
Gross benefit (B) Liter & ETB 375.75 5636.25 570.79 8561.85 993.04 14895.60
Net benefit (B -A) ETB - 1994.25 - 2251.85 - 6039.60
Cost benefit ratio (B/A) ETB - 1.54 1.35 - 1.68
T1 = Grazing + Grass hay (dry season), T2 = Grass hay + Napier grass, T3 = Grass hay +
Concentrate, ETB = Ethiopian birr

98
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATION

Milk yield of Fogera cows improved through concentrate and Napier grass supplementation.
The difference in average daily milk yield between concentrate supplemented and controlled
(grazing) groups is more than double that indicates milk yield of Fogera cows could be
greatly improved through concentrate supplementation. Feed supplementation also increased
the 200 days milk yield which implies calves will suckle more milk, and therefore will have
better growth. However, except solid not fat and minerals other parameters of milk
composition had no significant variation with concentrate and Napier grass supplementary
feeding. Concentrate feeding gave the highest benefit when compared to other treatments.
Thus, it is recommended that the supplementation of concentrate feed is biologically efficient
and potentially profitable in the feeding system of Fogera cows.

ACKNOWLEDGEMENTS

The authors would like to thank the Agricultural Growth Program (AGP II) project under
ARARI (Amhara Agricultural Research Institute) for the financial support and those who
participated in this experiment for their valuable contribution in experimental cow’s feeding,
milking and data collection.

REFERENCES

Addisu Bitew and Hegde B. P. 2003. Reproductive and growth performance of Fogera cattle
and their F1 Friesian crosses at Metekel ranch, Ethiopia. Proceedings of the 10th Annual
Conference of the Ethiopian Society of Animal Production (ESAP) held in Addis Ababa,
Ethiopia.
Addisu Bitew, Mengistie Taye, Adebabay Kebede, Getinet Mekuriaw, Asaminew Tasew,
Tezera Mulugeta and Gebeyehu Goshu. 2010. Milk yield and calf growth performance
of cattle under partial suckling system at Andassa Livestock Research Center, North
west Ethiopia. 22 (8). http://www.lrrd.org.
Adebabay Kebede, Firew Tegegne, Zeleke Mekuriaw and Azage Tegegne. 2011.
Characterization of Milk Production Systems in Bure District, Ethiopia. The IUP Journal
of Life Sciences, Vol. 5, No. 2, pp. 40-53, 2011. Available at SSRN: https: //ssrn .com
/abstract=2013728.

99
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

AOAC. 1990. Official Methods of Analysis: Changes in Official Methods of Analysis Made
at the Annual Meeting. Supplement, Association of Official Analytical Chemists.
AOAC. 1995. Official methods of analysis of AOAC International (16th edn): edited by
Patricia A. Cunniff, AOAC International, 1995. (North America) (elsewhere) (xxvi+
1899 pages) ISBN 0 935 584 54 4. Elsevier.
AOAC. 2000. Official methods of analysis of AOAC international 17th edition. AOAC
international, Gaithersburg, USA, 43-44.
CSA (Central Statistics Authority). 2013. Agricultural sample survey, 2012/13 (2005 E.C).
Volume II, report on livestock and livestock characteristics (private peasant holdings).
Statistical bulletin, 570, April 2013, Addis Ababa, Ethiopia.
Lateef, M., Gondal, K. Z., Younas, M., Sarwar, M., Mustafa, M. I. and Bashir, M. K. 2008.
Milk production potential of pure bred Holstein Friesian and jersey cows in subtropical
environment of Pakistan. Pakistan Vet. J., 2008, 28(1): 9-12.
Meissner, H.H., Kster, H.H., NIieuwoudt, S.H. and Coertze, R.J. l99l. Effect of energy
supplementation on intake and digestion of early and mid-season ryegrass and
Panicum/Smuts finger hay, and on in Sacco disappearance of various forage species. S.
Afr. J. Anim. Sci. 21, 33.
Misganaw Walie, Beyadglign Hunegnaw, Mekonnen Tilahun, Tewodros Bimerew,
Wondimeneh Mekonnen, Lisanework Molla, Adebabay Kebede and Asresu Yitayew.
2018. Development of Napier Grass Based Feeding Package for Crossbred Dairy Cows
at Andassa Livestock Research Center. P. 214-227. In Mesfin Lakew, Likawent Yeheyis
and Mulugeta Alemayehu (Eds.). 2018. Proceeding of the 10th Annual Regional
Conference on Livestock Completed Research Activities 13-16 March, 2017, Amhara
Agricultural Research Institute, Bahir Dar, Ethiopia.
Muia, J.M.K. 2000. Use of Napier grass to improve smallholder milk production in Kenya.
PhD Thesis Wageningen University, Wageningen, the Netherlands.
NRC. 2001. Nutrient requirements of dairy cattle: 2001, National Academies Press.
Preston, T. R. and Vaccarro L. 1989. Dual purpose cattle production systems. In: C. J. C.
Phillips (ed.) New Techniques in Cattle Production. pp 20–32. Butterworths Scientific,
London.
Rekik, B., Gara, A.B., Hamouda, M.B. and Hammami, H. 2003. Fitting lactation curves of
dairy cattle in different types of herds in Tunisia. Journal of livestock production
science, 2003, 83, 2-3,309-315.

100
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sairanen, A. 2014. Milk production and physiological response to concentrate

supplementation of dairy cows grazing timothy meadow fescue swards. Doctoral thesis
University of Helsinki, Finland.
Sallam, A. B., Musa, A.M., Tibin, J. B. and Jadalla, N.A.E. 2017. Effects of Level of
Nutrition on Milk Yield and Composition of Cows Fed Natural Grazing and Offered
Different Supplements, in South Kordofan, Sudan. Journal of Scientific and Engineering
Research, 2017, 4(4):239-242.
Silva, T.M., Oliveira, R.L. and Barbosa, L.P. 2011. Preliminary study on meat quality of
goats fed levels of licury oil in the diet. Asian Australian Journal Animal Science, v.24,
n.8, p.1112-1119.
SPSS Inc. Released. 2011. SPSS for Windows, Version 20. Chicago, SPSS Inc.
Van Soest, P.V., Robertson, J. and Lewis, B. 1991. Methods for dietary fiber, neutral
detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. Journal of
dairy science, 74, 3583-3597.
Yihalem Denekew. 2005. The effect of stage of harvesting on botanical composition, dry
matter yield (DMY) and nutritive value of natural pasture at Andassa Livestock
Research Center, Ethiopia. MSc. theses, Alemaya University, Ethiopia.

101
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Examining Preservation and Feeding Effects of Brewery Spent Grain on Dairy Cattle at
Gondar Town, Amhara Region, Ethiopia

Alemu Tarekegn, Kifetew Adane, Yosef Amsalu, Yengusie Demsew, Tsegaye Asredie

Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia

ABSTRACT

The objectives of the study were to assess the current production potential, storage condition and
utilization of brewery spent grain and to determine the effects of feeding brewery spent grain on
production potential and health of dairy cattle. The study was conducted at Gondar town from
July 2018 to January 2019. A total of 84 randomly selected dairy producers were interviewed
using a pre-tested structured questionnaire and through direct observations. Microbiological and
chemical analyses were carried out for brewer’s spent grain samples collected at five stages of
contamination (factory level, dispatch-level one day after arrival, 1 to 30 days stay ensiled, 30 to
60 days stay ensiled and dried sample). The results revealed that there was high variation in
utilization of brewery spent grain to dairy cattle producers across the farm types. Estimated daily
roughage, concentrate, total feed DM, daily brewer’s spent grain and mineral (salt) consumption
was highly variable in amount and proportion of brewery spent grain to other concentrate was
significant (3.0 to 3.2 kg DM cow-1 d-1) both in medium and large farms. However, there were
great variations in households’ response towards the long term effect of brewery spent grain
feeding on milk lactation yield, milk composition and health complications of dairy cows
(P<0.05). Though, majority (96.13%) of interviewed households felt positive improvements in
lactation yield and milk composition, some other farmers (72.06%) were also perceived to have
some health complications due to consumption of brewery spent grain. Commonly used brewery
spent grain preservation techniques included ensiling (96.14%) and sun drying (0.88%). No
Aflatoxin, Ochratoxin and Salmonella spp. contamination detected in the entire brewery spent
grain samples. High level of (>1,100MPN/g) E. coli was recorded from samples collected at 30 to
60 days after ensiling and at dried samples. Significant (6.7 x 103 to 6.5 x 107CFU/g) counts of
fungi were observed as the storage time increases from day 0 (factory level) to at 30 to 60 days
after ensiling and at dried sample. The vitamin A content of brewery spent grain samples was
(<0.10mg/100g) which was below the recommended level for dairy cows. To avoid vitamin A
deficiency supplementation with cheap and locally available vitamin A sources should be done.
Since wet brewery grain is delivered in bulk, appropriate feed storage and handling systems need
to be available as a mitigation measure for some dairy health complications.

Keywords: aflatoxin, brewery spent grain, crossbred dairy cattle, ochratoxin, smallholder

102
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Brewing industry generates relatively large amounts of by-products and wastes; spent grain,
spent hops and yeast being the most common by- product in the brewing process. Spent grain
is the most abundant brewing by-product, corresponding to around 85% of total by-products
generated. Depending on the type of beer to be produced, Brewer’s Spent Grain (BSG) may
consist of the residues from malted barley, maize, sorghum, rice and wheat (Reinold, 1997).
BSG is available at low or no cost throughout the year, and is produced in large quantities as
long as there is demand for beer. These materials can be fed to cattle in the wet form (wet
brewer’s grains) or dried (dried brewers grains). Wet BSG from a brewery factory contains
77 - 81% (w/w) water (Russ et al., 2005). Due to its high moisture and fermentable sugar
contents, BSG is a very unstable material and is liable to deteriorate rapidly due to microbial
activity. Several methods have been proposed to prolong BSG storage time as a result of their
high moisture content. Drying is a possible alternative for BSG preservation with the
advantage that it also reduces the product volume, and therefore, decreases transport and
storage costs (Santos et al., 2003).

The chemical composition of BSG varies according to barley variety, harvesting time,
malting and mashing conditions, and the quality and type of adjuncts added in the brewing
process (Santos et al., 2003); but in general, BSG is considered as a lignocellulosic material
rich in protein and fiber, which account for around 20 and 70% of its composition,
respectively. BSG also contains different vitamins such as biotin (0.1), choline (1800), folic
acid (0.2), niacin (44), pantothenic acid (8.5), riboflavin (1.5), thiamine (0.7) and pyridoxine
(0.7) ppm (Huige, 1994).

According to Huige (1994), BSG is an excellent feed ingredient for ruminants since it can be
combined with inexpensive nitrogen sources, such as urea, which can be synthesized to
essential amino acids in the digestive system. Traditionally spent grain is used to feed cattle
being a valuable supplement to existing feed due to its high protein content. Currently, the
primary market for BSG is as dairy cattle feed (Kaur and Saxena, 2004). In addition to its
high nutritional value, BSG is reported to promote increased milk production without
affecting animal fertility (Reinold, 1997). When BSG incorporated into the diet of cows; milk
yield, milk total solid content and milk fat yield will increase (Belibasakis and Tsirgogianni,
1996).

103
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

On the other hand if the material is not properly stored and utilized it is believed to be a
source of harmful molds and bacteria that can cause disease (Boateng et al., 2015). Among
this microbial spoilage moldy feed, afflation, and mycotoxin are of the contaminants of BSG
that can cause mycoticplacentitis which is a common source of abortion or death of a calf
shortly after birth and the birth of weak and deformed calves. In this regard, producers in the
study area are complaining the product as a source of abortion, calf mortality and blindness.
However there is a lack of organized and detailed information to the overall positive and
negative contribution of BSG in the dairy sector in the study area. Thus, the objectives of the
study were to assess the current production potential, storage condition and utilization of
brewery spent grain by smallholder dairy producers and to determine the effects of feeding
brewery spent grain on the health and production potential of dairy cows in the study area.

MATERIALS AND METHODS

Description of the Study Area

The study was conducted in Gondar town and its surroundings, which is found in the North
West part of Amhara National Regional State (ANRS), Ethiopia. The town is located
between latitude 12o54’31”N and longitude 37o41’05”E, and at about 748 km North West of
Addis Ababa and at about 180 km away from the regional capital Bahir Dar city. Gondar
town is bordered by the districts of East Dembia to the south, Gondar Zuria to the east, Lay
Armachiho to the north and west. The agro-ecological map of the town reveals that the study
area lies under moist woina dega agro-ecological zone (AEZ). The altitude of Gondar town is
about 2133 m.a.s.l. Gondar town is characterized by a uni-modal type of rainfall pattern with
annual average rainfall of 1037 mm. The minimum annual temperature ranges between 12
and 15oC. The daily maximum temperature becomes high during the months of March to
May, during which the temperature can reach as high as 34.8oC. The mean annual
temperature is about 20oC (World Weather Information Service - Gondar, 2018).

Sampling Technique and Sample Frame

A multistage sampling procedure was employed to select respondents. At the first stage,
Gondar town was selected purposely based on the distribution of dairy cattle and high
utilization potential of brewery spent grain. In the second stage six kebeles were purposely
selected from the town based on their comparative advantages over the others in dairy cattle
population and brewery spent grain utilization experience and potential.

104
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The target population for the study was the entire list of households in Gondar town who own
dairy cattle. A list of 122 smallholder dairy producers in the town was obtained from Gondar
town administration agricultural development office. The number of households sampled in
the study area was determined by N = 0.25/SE2, where N = number of sampled households,
SE = standard error (Arsham, 2005). Considering standard error of 0.05 with 95% confidence
interval as follows, N = 0.25/ (0.05)2 = 84. Then a total of 84 smallholder dairy farmers were
randomly selected after determining the total population engaged in dairying and feeding the
brewery spent grain in each kebele using simple random sampling technique. Proportional
sampling was used to fix the sample size for each kebele. The randomly selected
farms/households were stratified again into three groups based on the number of dairy cattle
owned as: small farms owning 1 - 8 dairy cattle, medium farms 9 - 15 animals and large
farms with more than 15 animals (Azage et al., 2013).

Data Sources and Methods of Data Collection

In this study, both primary and secondary sources of data were used. The primary data was
collected from sample respondents through semi-structured questionnaire surveys and key
informant interviews. Secondary data from various sources were collected through desk
review. The questionnaire for the survey was prepared in English, and translated into the
local ‘Amharic’ language. Before the actual survey the questionnaires were pre-tested to
check its appropriateness and correctness of generating all the necessary information to meet
all the stated objectives and fine-tuned accordingly. Data was collected on socio-economic
characteristics of the respondents, feed sources and utilization, brewery spent grain storage,
utilization and its effect on dairy cattle and major challenges in dairy cattle production.
Individuals whom the researcher believed them as resource persons to give the intended
information were purposely selected and interviewed using a prepared checklist.

Sample Collection and Laboratory Analysis

Brewer’s spent grain samples were collected at five stages considering the common stage of
contamination (at factory level, at dispatch-level one day after arrival, at 1 to 30 days stay
ensiled, at 30 to 60 days stay ensiled and dried sample). For each stage of sample collection,
samples were collected from five representative substrate sources.

105
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Samples collected at each stage of contamination were pooled together, homogenized and
quartered to obtain 500g laboratory samples for microbiological analysis (Salmonella spp.,
E.coli, Yeast and Mold) and 1000 g laboratory samples for chemical analysis (Total
Aflatoxin, Ochratoxin A and Vitamin A) according to the recommendation of Bless Agri.
Food Laboratory Services Private Limited Company. Samples were properly packed in bags
and immediately sent to the laboratory.

In microbiological analysis, detection of Salmonella spp. was carried out using the method
described in SR EN ISO 6579: 2003 + AC: 2006 standard and SR EN ISO 16649-2: 2007
standard, respectively (International Organization for Standardization, 2007). The
quantitative enumeration of Salmonella spp. per gram of sample (CFU/g) was expressed as
described by the standard.

Total combined yeasts and moulds count (TYMC) was determined using the method
described in SR ISO 21527-1: 2009 standard (International Organization for Standardization,
2009). The quantitative enumeration of fungi as colony-forming units per gram of food
(CFU/g) was performed using the surface-spread method described by (Pitt and Hocking,
1997). Ten grams of each sample were homogenized in 90 ml distilled water solution for 30
minutes in an orbital shaker. Serial dilutions (from 10 - 2 to 10 - 5) were made, and 0.1 ml
aliquots were inoculated in duplicates onto the media dichloran that rose Bengal
chloramphenicol agar (DRBC) for estimating total culturable fungi Abarca et al. (1994) and
dichloran 18% glycerol agar (DG18) that favors xerophilic fungi development. The plates
were incubated at 25oC for 5 - 7 days. All samples were also inoculated onto Nash and
Snyder agar (NSA) to enumerate Fusarium species (Nelso et al., 1983). Nash-Snyder plates
were incubated at 24oC for 7 days under a 12 h cold white/12 h black fluorescent light
photoperiod. Only plates containing 10 - 100 CFU were used for counting, with results
expressed as CFU per gram of sample.

Escherichia coli population was determined using the procedure described by (AOAC, 1996).
AOAC confirmatory method for analysis of E.coli from each gas positive LST tube, a 3ml
loopful of turbid broth was transferred to a test tube containing an inverted Durham
fermentation tube and approximately 10ml of sterile EC broth. These tubes were incubated in
a water bath at 44.5oC and examined for gas formation at 12, 24 and 48 hrs, respectively
(AOAC, 1996).

106
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The results were recorded and the most probable number (MPN) of E.coli was calculated
with the use of an MPN table. The number of gas positive tubes at each dilution level was
recorded and a 3 number code calculated, signifying the number of positive tubes at each
level. This code was recorded to an MPN table from which an MPN value was selected. The
most probable number per gram of sample was calculated as follows:

MPN value from table X dilution factor of middle = organisms/g of sample

100 set of tubes

The higher pathogen contents were stabilized to the required standard (<103 MPN/g) set by
(AOAC, 1996).

The extraction of Aflatoxin (AF) and Ochratoxin A (OTA) was determined according to
Soares and Rodriguez- Amaya procedure (Soares and Rodriguez-Amaya, 1989). Quantitative
evaluation of Aflatoxin (AF) and Ochratoxin A (OTA) was made using high performance
liquid chromatography (HPLC). The detection limit of the technique for AF was 1.0 μg/kg
and the limit of detection for OTA was 2.5 μg/kg.

Level of Vitamin A in the feed sample was determined by using High-Performance Liquid
Chromatographic (HPLC) methods (AOAC, 2007). During the analysis to provide accurate
feed measurements of vitamin A activity low acetinic glassware, nitrogen, and/or vacuum, as
well as avoiding excessively high temperature were considered. Brewery spent grain samples
of 40mg were transferred to a 100ml digestion flask containing a stirring bar. For
saponification of the sample 10ml of ethanolic pyrogallol solution and 40ml ethanolic KOH
were added. The flask was wrapped in aluminum foil and stirred at room temperature for 18
hr, or at 70oC using the reflux vessel and then diluted to volume with an ethanolic pyrogallol
solution. 3ml of digestate was pipetted into a 15ml centrifuge tube and 2ml of deionized H2O
was added. Vitamin A was extracted with 7ml of hexane: diethyl ether (85:15, v/v).
Extraction was repeated 2X with 7ml portions of extractant. After extractions the organic
layer were transferred to a 25ml volumetric flask. 1ml of hexadecane solution (i.e., 1ml
hexadecane in 100ml hexane) was added and diluted to volume with hexane. 15ml of diluted
extract was pipetted into a test tube and evaporate under nitrogen and then dissolve the
residue in 0.5ml of heptane.

107
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Management and Analysis

The computer software Excel was used for data management and entry. All the collected
survey data were coded and entered into the computer with Excel. Data was analyzed using
Statistical Package for Social Sciences (SPSS version, 22, 2002). Descriptive statistics such
as means, frequency distribution and percentages were used. Furthermore, one-way ANOVA
was used to examine differences between continuous variables. Differences were considered
significant at P<0.05.

RESULTS AND DISCUSSION

Socio- Economic Characteristics of Respondents

Household characteristics of the respondents in the study area are shown in Table 1. Overall,
the mean age of the household heads was found to be 50.70±2.01 years and there was no
difference (P>0.05) among farm sizes. The result shows that farmers with medium age were
involved in dairy production in Gondar town. The overall mean family size was 4.93±0.40
persons/household and there was no difference (P>0.05) among the farm types. This result
was lower than the findings of Asaminew and Eyassu (2009) for Bahir Dar Zuria (8.2) and
Mecha districts (7.2).

Overall, 78.91 and 21.09% of the farmers were male and female-headed households,
respectively and there was no significant difference (P>0.05) among farm sizes. The
proportion of female-headed households in the present study was lower than the 33% for
Addis Ababa (Azage, 2004). In this study, the proportions of female- headed households
were relatively few (18.52%) in medium farm sizes.

The current study results revealed that 89.29% of the dairy producers were married, while
divorced, widowed and unmarried were 2.38, 3.57 and 4.76%, respectively. There was more
number of married dairy producers in all farm types that range from 88.89 to 89.47%.
Significant difference (P<0.05) was observed among farm types practicing dairying both as a
main and sideline activity. Results of the current study revealed that the majority of the
respondents (66.67%) do their dairying as a side business to supplement their income. In line
with the current finding, Smith and Hogan (1999) reported that urban dairy farming can be a
part-time activity where household members work in other sectors of the urban economy.

108
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Regarding educational level of respondents, 39.29, 16.67, 5.95, 11.90, 14.26, 10.71 and
1.19% of the interviewed households had illiterate, basic education, primary school, junior
secondary, senior secondary, diploma, and degree and above, respectively. However,
statistically there was no significant variation (P>0.05) among the farm sizes in terms of
educational levels. Majority of the respondents in the present study were literate and which is
important to understand extension messages and to realize the importance of new
technologies. The current finding is supported by the report of Ofukou et al. (2009) who
stated that dairy producers with better reading and writing levels usually adopt new
technologies more rapidly than illiterate ones.

Table 1: Socio- economic characteristics of respondents

Farm type
Variables Small (N = 38) Medium (N = 27) Large (N = 19) Total (N = 84) P value
Age (mean ±SE) 49.14±1.8 52.01±2.01 50.94±2.21 50.70±2.01 0.508
Family size (mean 4.84±0.29 5.03±0.31 4.92±0.61 4.93±0.40 0.328
±SE)(%)
Sex 0.885
Male 76.32 81.48 79.95 78.91
Female 23.68 18.52 21.05 21.09
Marital status (%) 0.954
Married 89.47 88.89 89.47 89.29
Divorced 5.25 0.00 0.00 2.38
Widowed 0.00 2.40 5.26 3.57
Unmarried 5.26 3.70 5.26 4.76
Educational level (%) 0.899
Illiterate 42.11 37.06 36.84 39.29
Basic education 18.42 14.81 15.79 16.67
Primary school 2.63 11.11 5.26 5.95
Junior secondary 7.89 14.81 15.79 11.90
Senior secondary 18.42 14.81 5.26 14.26
Diploma 7.89 7.40 21.05 10.71
Degree and above 2.63 0.00 0.00 1.19
Dairy practice 0.037
Main activity 26.32b 29.63b 52.63a 33.33
Side line activity 73.68a 70.37a 47.37b 66.67
*
a, b = Means with different superscript letters in the same row are significantly different at
P< 0.05; N = number of respondents; SE = standard error.

Brewery Spent Grain Production, Availability, Supply and Marketing

Dashen is the only brewery factory which is operating at Gondar town. Commonly produced
byproduct feeds at the factory are wet brewery spent grain and Brewer’s spent yeast but
supply of spent yeast to dairy producers so far was not to the level expected. The actual total
amount of BSG produced from the factory in the same fiscal year was 2602.5 tons (Getu et
al., 2018).

109
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Almost all BSG produced are daily supplied to dairy cooperatives according to their order of
registration at the factory. Availability of BSG in the study area was subjected to seasonal
fluctuations. Accordingly, the quantity of BSG produced and supplied to user communities
during the long rainy season relatively drops owing to the chilling weather conditions that
suppresses beer consumption and subsequent beer production. The supply and marketing of
BSG in the study area is presented in Table 2. There is high variation in the frequency of
BSG supply to dairy producers across the farm types (P<0.001). Majority of the sample
respondents (56.49%) were receiving BSG in every three weeks interval. This can be
speculated to be the existing high demand for BSG associated with better profitability of the
dairy business in the study area.

Table 2: Price per quintal and frequency of BSG distribution to dairy producers’ at Gondar
town
Farm type
Variables Small (N = 38) Medium (N = 27) Large (N = 19) Total (N = 84) P value
Frequency of Every 2 week 16.06 11.86 5.71 11.21c
distribution Every 3 week 62.27 45.76 61.43 56.49a
Every 4 week 16.12 33.9 28.57 26.20b
Above 4 5.55 8.48 4.29 6.10c 0.000
Price per quintalweeks
of BSG on
DM basis (mean ± SE) 135.41±1.08 132.36±0.97 136.57±0.67 134.78±0.91 0.253
*a, b, c
= Means with different superscript letters in the same column are significantly different
at P< 0.001; DM = dry matter; N = number of respondents.

BSG was sold at a very low factory gate price of 0.31ETB/kg on a wet basis but the dairy
cooperatives included transportation and some profit margins as an administrative cost. Mean
purchasing price for a quintal of brewers’ grain on DM basis across the farm types was
134.78±0.91 ETB. The price did not show any variability (P>0.05) among the farm types.

Daily Feed Intake of Lactating Dairy Cows

The average quantity of roughage feed (grass hay) consumed by a lactating crossbred cow
was estimated at 4.03 kg d-1, with cows in medium and large farm types daily consuming
considerably higher (P<0.05) roughage than those cows in small farms (Table 3). Similarly,
estimated daily concentrate and total feed DM consumed by lactating cows in large farms
was significantly higher (P<0.05) than cows in small and medium farms. The higher total
feed DM intake by dairy cows at large farms can be linked to the higher ratio of supplement
to roughage in the total diet.

110
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In the farms, there was significant variation (P<0.001) in the daily brewer’s spent grain and
mineral (salt) consumption of individual cows with values being higher for medium and large
farms. The single most frequently utilized mineral source in all farm types in the study area
was “table salt”. The amount of estimated daily brewery spent grain consumed when
expressed as percentage of the total daily concentrate and feed DM intake was also
significantly higher (P<0.05) for lactating cows both in medium and large farm types. The
observed variations among the farm types for daily BSG intakes in the present study might be
partly associated with the difference in their engagement dairying as a main activity as a
means of living.

Table 3: Estimated daily feed intake (kg DM cow-1 d-1) of lactating dairy cows in different
farm types at Gondar town
Farm type
Feed intake Small (N = 38) Medium (N = 27) Large (N = 19) Total (N = 84) P value
Roughage 3.51±0.17b 4.3±0.16a 4.3±0.11a 4.03±0.15 0.042
Total concentrate* 4.52±0.21c 4.9±0.17b 5.2±0.16a 4.87±0.18 0.021
Brewery grain 2.2±0.21b 3.0±0.26a 3.2±0.14 a 2.80±0.15 0.008
Mineral (salt) 0.04±0.07b 0.07±0.04a 0.08±0.04a 0.06±0.05 0.009
Total feed DM intake 8.03±0.29b 9.2±0.24a 9.5±0.22a 8.91±0.25 0.036
Proportion of brewer’s grain
To total concentrate (%) 48.67 ±2.63b 61.22±2.19a 61.54±2.77a 57.15±2.53 0.031
To total DM (%) 27.40 ±1.87b 32.61±1.43a 33.68±1.37a 31.23±1.56 0.043
*a, b, c
= Means with different superscript letters in the same row are significantly different at
P< 0.01 and P< 0.05; DM = dry matter; N = number of respondents, * = Total concentrate
includes brewery grain.

The average amount of daily roughage DM intake (4.03±0.15 kg-1) to a crossbred dairy cow
was relatively higher than (3.5 kg) reported in urban dairy farms at Bedele, Debre Birhan and
Sebeta (Getu et al., 2018). Discrepancy between the present findings and previously reported
values could be attributed to the differences in the availability and cost of roughage feeds in
different areas. On the other hand, the quantity of concentrate and total dry matter intake fed
to a crossbred cow in the study area was consistent to the value (8.47 and 4.9 kg/day)
reported by the same authors.

Perceptions of Respondents on the Effects of Feeding BSG for Dairy Cattle

The perception of respondents on the effects of feeding BSG on dairy cattle in the study area
is presented in Table 4. There was high variation (P<0.05) on the perception of long term
effects of feeding BSG in producers response in the study area.

111
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Overall results from the current study indicated that 96.13 and 89.28% of the interviewed
households felt positive improvements on milk yield and composition, respectively as a result
of BSG supplementation. This is due to the greater consumption of brewery grain that
contains higher digestible proteins and crude fat. Brewers’ grain is mainly classified as a
protein supplement and hence its introduction in dairy cows diet has been observed to
increase milk yield and milk fat due to the greater amount of digestible proteins and crude fat
consumption by cows (Biljana, 2013).

Table 4: Households’ perception on the effects of feeding BSG for dairy cattle (%) in the
study area
Farm type
Small Medium Large Total
Variables (N = 38) (N = 27) (N = 19) (N = 84) P value
Change in milk Milk yield increases 92.11 96.30 100.00 96.13a 0.023
yield Milk yield decreases 2.63 0.00 0.00 0.88b
No change 5.26 3.70 0.00 2.99b
Change in milk Milk fat increases 94.74 88.89 84.21 89.28a 0.013
composition Milk becomes watery 0,00 3.70 5.26 2.99c
No change 5.26 7.41 10.53 7.73b
Health related Still birth 13.16 7.40 0.00 6.83c 0.042
Problems Abortion 10.53 22.22 21.05 17.93b
encountered Calf mortality 15.79 14.81 26.32 18.97b
Calf blindness 7.89 18.52 15.79 14.07b
Diarrhoea 10.53 7.41 10.53 9.49c
Birth of weak calves 5.26 3.70 5.26 4.74c
NA 36.84 25.93 21.05 27.94a
*a, b, c
= Means with different superscript letters in the same column are significantly different
at P< 0.05; NA = not applicable; N = number of respondents.

Overall mean result from the present study also indicated that the majority (72.06%) of the
respondents from the surveyed households reported to encounter some health complications
associated with feeding BSG. According to the interviewed households’ calf mortality,
abortion and calf blindness in their order of importance were the major health problems they
encountered related to BSG feeding. On the other hand, 27.94% of responding producers
from the surveyed households did not encounter any major health problems in response to
feeding BSG. The root cause for calf blindness mentioned by respondents in the study area
could still be speculated to be mycotic effects arising from over consumption of spoiled BSG.
Poor health condition of dairy cattle in the study area could be linked to over consumption of
badly stored BSG that have been exposed to fungal and yeast growth of pathogenic
importance. Mould and yeast have been reported to inflict heavy loss in production and
health performances of farm animals (McDowell, 2000; Boateng et al., 2015).

112
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Brewery Spent Grain Preservation Practices in the Study Area

The BSG preservation practices of BSG by the sample respondents in the study area are
summarized in Table 5. In general, about 97.01% of the responding households across the
study area conserve wet spent grains. Preservation practice of respondents’ significantly
(P<0.001) vary in their brewery grain preservation techniques. Majority of sample
respondents (96.14%) practice a combination of salting and ensiling to increase the shelf life
of BSG. Sun drying was the second preservation technique only practiced by (0.88%)
respondents in the study area. Aside from improving its shelf life and lowering the storage
volume, drying the BSG could enable users to further explore its incorporation in the rations
of other animals.

Table 5: Major wet brewery paint grain preservation practices and facilities for preservation
in the study area (% of respondents)
Farm type
Small Medium Large Total
Preservation practices and facilities (N = 38) (N = 27) (N = 19) (N = 84) P value
BSG preservation Salting and ensiling 92.11 96.3 100 96.14a 0.0000
mechanisms Sun drying 2.63 0.00 0.00 0.88b
NA 5.26 3.70 0.00 2.99b
Facilities for Wooden upright silo 52.63 62.96 42.1 52.56a 0.0000
Preservation Concrete trench silo 28.95 22.22 57.89 36.35b
Metal made barrel 18.42 14.81 0.00 11.08c
*a, b, c
= Means with different superscript letters in the same column are significantly different
at P< 0.001; NA = not applicable, N = number of respondents.

The most commonly used BSG storage facilities in the study area were wooden made upright
silo, concrete trench silo and metal made barrel in their order of importance (Table 5). The
type of facilities used for BSG preservation was significantly (P<0.001) highly variable
among farm types. The majority of the small and medium producers preferred to utilize least
cost wooden made plastic lined up right silo while larger producers due to its efficiency and
long lasting usage they mostly preferred to use concrete made trench silo for BSG
preservation. Though metal made barrel was the most efficient facility due to its high cost
only a small number of dairy producers (11.08%) used it as a preservation facility. Regarding
knowledge about preservation techniques, only few (13.10%) of the sample respondents
reported to receive information related to BSG preservation techniques from their cooperative
and neighbours.

113
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

According to our personal observations, the majority of producers claimed the absence of
technical training related to preservation and storage techniques. A rough estimate of 2.6%
salt on BSG DM was used to conserve BSG during ensiling in the study areas. This appears
to be higher than the 1.3 - 2.5% salt used to effectively conserve BSG (IPRO’s, 2011).

Analysis of the Mycobiota

The mean detected levels of BSG E. coli (MPN/g), Salmonella spp. (CFU/g) and total yeast
and mould count (CFU/g) is presented at Table 6. The level of E. coli contamination was
highly variable between BSG samples collected at different sampling stages. Samples which
were collected at factory level, at dispatch (1 day after arrival) and 1 to 30 days after ensiling
recorded significantly lower E. coli (MPN/g) than other sampling stages, which implies no
characteristic growth of the strain was observed at these stages. Collected samples at 30 to 60
days after ensiling and dried samples recorded higher concentration of E. coli (MPN/g) which
is higher than the required standard for higher pathogen contents (<103 MPN/g) set by
(AOAC, 1996). This could be due to reduced levels of pH that can favour the proliferation of
E. coli. Reduced the minimal pH (4.40 to 5.00) in tomato juice at which growth of E. coli
could be initiated (AOAC, 1996). This might lead to a generalization that substrates at 30 to
60 days after ensiling and dried products can be the cause of infantile diarrhoea
and calf scours which agrees with the survey result.

Brewer’s grains samples collected at different stages did not show any Salmonella spp.
contamination. This indicates that the environment was not convenient for the propagation of
the pathogen.

Table 6: Microbiological characteristics of brewery spent grain samples collected at different

stages in Gondar town
Pathogens detected
Sample E. coli Salmonella Total yeast and mould
type Sample collection stage (MPN/g) spp. (CFU/g) count (CFU/g)
Brewery At factory level <3 ND 4.0 x 102
spent grain At dispatch (1 day after arrival) <3 ND 6.7 x 103
At 1 to 30 days after ensiling <3 ND 7.8 x 104
At 30 to 60 days after ensiling >1,100 ND 1.0 x 105
Dried sample >1,100 ND 6.5 x 107
*
CFU = colony-forming units; ND = not detected.

114
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The total yeast and mould counts in the brewery spent grain samples collected at different
stages in the study area were highly variable that ranges from 4.0 x 102 to 6.5 x 107 CFU/g.
All the samples collected at different stages except samples collected at factory level were
recorded at much higher total yeast and mould counts than the detection limit of the method
(1 x 103 CFU/g). These results suggest a high fungal activity that could affect the palatability
of feed and reduce the animal’s nutrients absorption, determining a low-quality substrate
(Martins and Martins, 2001). Comparing counts of the different sampling stages, higher
counts were observed as the storage time increases from day 0 (sample at factory level) to at
30 to 60 days after ensiling and at dried sample as well. This could be due to the presence of
continual aerobic exposure during taking out of the product at feeding that can favor the
growth of aerobic bacteria. Total yeast and mould counts from samples collected at 1 to 30
days and at 30 to 60 days after ensiling and at dried sample were relatively much higher than
the limit of fungal colonies established as a hygienic quality standard (1 x 104 CFU/g)
according to Good Manufacturer Practices guidance (GMP, 2008). In line with the current
study Cavaglieri et al. (2009) obtained counts ranging between 1 x 103 and 1 x 106 CFU/g
for the brewery spent grain samples.

Analysis of Mycotoxins and Vitamin A

Table 7 shows the Aflatoxin (AFs), Ochratoxin A (OTA) and Vitamin A levels found in the
brewery spent grain samples collected at different stages. Vitamin A concentration in the
brewery spent grain mg/100g did not show any variability across different sampling stages.
This indicates that there was no condition observed that can affect the concentration of
Vitamin A across different sample collection stages. Brewery spent grain samples collected at
different stages recorded Vitamin A below the recommended limit 30 IU/100g or
0.75mg/100g of brewery spent grain sample. According to NRC (2001) vitamin A
requirement for lactating dairy cows is about 72,000 IU or 1800 mg/day. However, lactating
dairy cows in the study area are actually receiving Vitamin A below the daily required level,
which indicates a serious Vitamin A deficiency in dairy farms practicing brewery spent grain
based feeding. This may result in calf blindness, diarrhea, abortion, stillbirths, calf mortality
due to respiratory and other infections and birth of weak calves in dairy farms, which agrees
with the survey result.

115
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In cattle, signs of vitamin A deficiency include reduced growth rate; diarrhea; lacrimation;
night blindness; blindness from corneal opacity; abnormal bone growth; low conception
rates; abortion; stillbirths; weak, blind or stillborn calves; and increased susceptibility to
respiratory and other infections (McDowell, 2000).

Table 7: Mycotoxins and vitamin A levels detected in brewery spent grain collected at
different stages at Gondar town
Chemicals detected
Sample Vitamin A Total Aflatoxin Ochratoxin A
type Sample collection stage mg/100g (µg/kg) (µg/kg)
Brewery At factory level <0.10 ND ND
spent grain At dispatch (1 day after arrival) <0.10 ND ND
At 1 to 30 days after ensiling <0.10 ND ND
At 30 to 60 days after ensiling <0.10 ND ND
Dried sample <0.10 ND ND
*
ND = not detected.

There were no detectable levels of AFs and OTA contamination in any of the analyzed
breweries spent grain samples collected at different stages in the study area. This implies that
none of pathogenic microorganisms colonizes the substrate at factory level, during dispatch,
during storage and at drying that can produce AFs and OTA. This could be due to the
presence of high temperature that can suppress mycotoxin production as a result of
microorganisms’ growth. Microorganisms’ near-freezing temperatures, they produce large
quantities of mycotoxins with minimal mold growth (Joffe, 1986). Generally, brewery spent
grain samples collected at different stages in the study area showed AFs and OTA levels
below the recommended limits 20µg/Kg and 50µg/Kg, respectively. Results of our study did
not agree with Batatinha et al. (2007) and Cavaglieri et al. (2009) who found FB1 in brewer’s
grains, at levels that ranged from 198 to 295 µg/Kg and from 564 to 1383 µg/Kg,
respectively. Gonzalez Pereyra et al. (2011) found the mean levels of AFs to be 104 - 145
µg/Kg for brewery spent grain collected from Argentinean breweries which is by far higher
than the result we obtained. Rosa et al. (2008) found higher amounts of OTA in samples of
brewer’s grains. Variations in the AFs and OTA production could be attributed to the
differences in environmental temperature that favor the growth of microorganisms’ which are
responsible for mycotoxins production.

116
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

From this study result it can be concluded that availability, conservation and proper
utilization of BSG are major challenges which the producers in the study area were seen to be
confronted with. Though, BSG believed to increase milk yield and composition, other
problems associated with the use of BSG include deterioration of the by-product and health
related issues. Improper use of salt as preservative, lack of awareness and absence of
technical training related to preservation and storage techniques of the by-product remain key
limiting factors. No detectable levels of AFs, OTA and Salmonella spp. contamination
observed in any of the analyzed breweries spent grain samples collected at different stages in
the study area. Significantly higher E. coli was recorded from samples collected at the level
of 30 to 60 days after ensiling and from dried samples that may cause infantile diarrhea
and calf scours. As a result of aerobic exposure during taking out of the product higher counts
of fungi were observed as the storage time increases from day 0 (sample at factory level) to at
30 to 60 days after ensiling and at dried sample. Brewery spent grain samples collected at
different stages in the study area recorded Vitamin A below the recommended level for dairy
cows. Setting the appropriate level of salt used to conserve BSG should be investigated.
Training should be given to the dairy producers on the proper preservation and utilization of
brewery spent grain to overcome negative impacts of feeding the by-product. Since WBSG
are delivered in bulk, appropriate feed storage and handling systems need to be available. To
avoid the presence of Vitamin A deficiency on brewery spent grain based feeding farms
vitamin A source of supplement should be provided.

ACKNOWLEDGMENT

The authors would like to acknowledge Agricultural Growth Program II for funding the
research project.

REFERENCES

Abarca, M. L., Bragulat, M. R., Castell´a, G. and Caba˜nes, F. J. 1994. “Ochratoxin A

production by strains of Aspergillus niger var. Niger,” Applied and Environmental
Microbiology, vol. 60, no. 7, pp. 2650–2652.
Arsham, H. 2005. Questionnaire design and survey sampling. http://home.ubalt.edu /ntsbarsh/
stat-data/surveys.htm.
Asaminew Tassew and Eyassu Seifu. 2009. Smallholder dairy system and emergency of dairy
cooperatives in Bahir Dar Zuria and Mecha Woredas, Northern Ethiopia. World J Dairy
Food Sci 4(2):185–192.

117
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Association of Official Analytical Chemists. 1966. Official Methods of Analysis. J. Assoc.

Off. Anal. Chemists. 49: 46-250.
Association of Official Analytical Chemists. 2007. Official methods of analysis, 18th ed.,
2005; Current through revision 2, 2007 (On-line). International, Gaithersburg, MD.
Azage Tegegne, Berhanu Gebremedhin, Hoekstra D, Berhanu Belay, and Yoseph Mekasha.
2013. Smallholder dairy production and marketing systems in Ethiopia: IPMS
experiences and opportunities for market-oriented development. IPMS of Ethiopian
farmers project working paper 3 1, Nairobi, Kenya, 65 pp.
Azage Tegegne. 2004. Urban livestock production and gender in Addis Ababa, UA
Magazine, 4, 30–31. In: Proceedings of the 14th annual conference of the Ethiopian
society of animal production (ESAP). September 5–7, 2006.Addis Ababa, Ethiopia.
Batatinha, M. J. M., Simas, M. M. S., Botura, M. B., Bitencourt, T. C., Reis, T. A. and
Correa, B. 2007. “Fumonisins in brewers grain (barley) used as dairy cattle feed in the
State of Bahia, Brazil,” Food Control, vol. 18, no. 5, pp. 608–612.
Belibasakis, N. G. and Tsirgogianni, D. 1996. Effects of wet brewers grains on milk yield,
milk composition and blood components of dairy cows in hot weather. Animal Feed
Science and Technology 57, 175–181.
Biljana, V, K., Ranko, P., Milun, R. and Dusan. 2013. Economic analysis of feed ingredients
in dairy cow ration. IV International Symposium. Agrosym 2013. University of
Kragujevac. Faculty of Agronomy. Serbia.
Boateng, M., Okai, D. B., Frimpong, Y. O. and Zeebone, Y. Y. 2015. Wet brewers’ spent
grains and wet brewers’ spent yeast: problems associated with their usage and suggested
solutions: a case study of the Ejisu-Juaben Municipality of Ghana. J. Liv. Res. for Rural
Dev. 27 (1).
Cavaglieri, L. R., Keller, K. M. and Pereyra, C. M. 2009. “Fungi and natural incidence of
selected mycotoxins in barley rootlets,” Journal of Stored Products Research, vol. 45, no.
3, pp. 147–150.
Getu Kitaw, Birhan Tamir, Getnet Assefa and Getachew Animut. 2018. Production,
Preservation, and Utilization Patterns of Brewery Spent Grain in Ethiopia. Ethiop. J.
Agric. Sci. 28(3) 1-17.
Gonzalez Pereyra, M.L., Rosa, C.A.R., Dalcero, A.M. and Cavaglieri, L.R. 2011. Mycobiota
and mycotoxins in malted barley and brewer’s spent grain from Argentinean breweries.
Letters in Applied Microbiology. 53, 649–655.
Huige, N. J. 1994. Brewery by-products and effluents, in: Hardwick, W.A. (Ed.), Handbook
of Brewing. Marcel Dekker, New York, pp. 501–550.
International Organization for Standardization (ISO). 2007. SR ISO 16649-2:2007.
Microbiology of food and animal feeding stuffs: horizontal method for the enumeration
of beta-glucuronidase-positive Escherichia coli. Part 2: colony-count technique at 44
degrees C using 5-bromo-4-chloro-3-indolyl beta-D-glucuronide. Geneva: ISO.
International Organization for Standardization (ISO). 2009. SR ISO 21527-1:2009:
microbiology of food and animal feeding stuffs: horizontal method for the enumeration
of yeasts and moulds. Part 1: colony count technique in products with water activity
greater than 0, 95. Geneva: ISO.

118
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

IPRO’s 340. 2011. Business study of alternative uses for brewers’ grain. Final project report.
25pp.
Joffe, A.Z. 1986. "Fusarium Species: Their Biology and Toxicology." John Wiley and Sons
Inc., New York.
Kaur, V. I. and Saxena, P. K. 2004. Incorporation of brewery waste in supplementary feed
and its impact on growth in some carps. Bioresource Technology 91, 101–104.
Martins, H. M. and Martins, M. L. 2001. “Mycological quality evaluation of bovine
feedstuffs (Portugal: 1996–1999),” Revista Portuguesa de Ciˆencias Veterinarias´, vol.
96, pp. 85–88.
McDowell, L. R. 2000. Vitamins in Animal and Human Nutrition, 2nd ed., Iowa State
University Press, Ames, IA.
National Research Council. 2001. Nutrient requirements of dairy cattle. 7th rev. ed. ed. Natl.
Acad. Press, Washington DC.
Nelson, P. E., Toussoun, T. A. and Marasas, W. F. O. 1983. Fusarium Species: An Illustrated
Manual for Identification. The Pennsylvania State University Press. Philadelphia Pa.
USA.
Ofukou, A. U., Egho, E. O. and Enujeke, E. 2009. Integrated Pest Management (IPM)
adoption among farmers in the Central Agro-ecological Zone of Delta State, Nigeria.
Adv Biol Res 3(1–2):29–33.
Pitt, J. I. and Hocking, A. D. 1997. Fungi and Food Spoilage. 2nd edition. Chapman & Hall.
Cambridge.
Reinold, M. R. 1997. Manual pra´tico de cervejaria, first ed. Aden Editora e Comunicac¸o˜es
Ltda, Sa˜o Paulo. 214p.
Rosa, C. A. R., Cavaglieri, L. R. and Ribeiro, J. M. M. 2008. “Mycobiota and naturally-
ochratoxin A in dairy cattle feed from Rio de Janeiro State,” World Mycotoxin Journal,
vol. 1, no. 2, pp. 195–201.
Russ, W., Mo¨rtel, H. and Meyer-Pittroff, R. 2005. Application of spent grains to increase
porosity in bricks. Construction and Building Materials 19, 117–126.
Santos, M., Jime´nez, J. J., Bartolome´, B., Go´mez-Cordove´s, C. and Del Nozal, M. J. 2003.
Variability of brewers’ spent grain within a brewery. Food Chemistry 80, 17–21.
Smith, K. L. and Hogan, J. S. 1999. A world standard for milk somatic cell count: is it
justified? Bull Int Dairy Fed 345:7–10.
Soares, L. M. V. and Rodriguez-Amaya, D. B. 1989. “Survey of aflatoxins, ochratoxin A,
zearalenone, and sterigmatocystin in some Brazilian foods by using multi-toxin thin-
layer chromatographic method”. Journal of the Association of Official Analytical
Chemists. vol. 72, no. 1, pp. 22–26.
SPSS. 2002. Statistical packages for social sciences, SPSS 22 for windows. SPSS Inc.
Chicago, Illinois.
World Weather Information Service – Gondar. 2018. World Meteorological Organization.
Archived from the original on 23 October 2018. Retrieved 31 March 2019.

119
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Effect of Substitution of Noug Seed Cake with Pigeon Pea Dried Leaves on Feed Intake
and Weight Gain of Old Oxen Fed a Basal Diet of Teff Straw

Abito Asres, Solomon Tiruneh and Negus Belayneh

Sirinka Agricultural Research Center, P.O. Box 74, Woldia, Ethiopia

ABSTRACT

The experiment was conducted to evaluate the substitution of noug seed cake with dried
pigeon pea leaves on feed intake and weight gain of local old oxen. A total of 25 local old
oxen (Bos indicus) with 7 - 8 years old and an initial mean body weight of 300.2±19.25 kg
were used for the study. The experiment was conducted in a randomized complete block
design (RCBD) using five oxen per treatment for 105 feeding and 15 acclimatization days.
The experimental oxen were blocked based on their initial empty body weight and randomly
assigned into five treatment groups. The feeding trial was carried out for 105 days including
15 days of acclimatization period. The dietary treatments were T1 = Teff straw ad libitum +
Wheat bran + Pigeon pea (25%) + Noug seed cake (75%), T2 = Teff straw ad libitum +
Wheat bran + Pigeon pea (50%) + Noug seed cake (50%), T3 = Teff straw ad libitum +
Wheat bran + Pigeon pea (75%) + Noug seed cake (25%), T4 = Teff straw ad libitum +
Wheat bran + Pigeon pea (100%) and T5 = Teff straw ad libitum + Wheat bran + Noug seed
cake (100%). The crude protein content of Pigeon pea dried leaves (24.47%), noug seed cake
(31.92%) and wheat bran (15.98%) can satisfy the ruminant production requirement. The
total DM, teff straw and nutrient intake were significant (P<0.01, P<0.001, P<0.001),
respectively between treatment groups, but supplement intake was not significant (P>0.05)
among treatment groups. The total DM intake ranged from 7.29 to 8.09 kg/d/head with mean
intake per treatment for 8.02, 8.09, 7.29, 8.08 and 7.80 kg/d/head of T1, T2, T3, T4 and T5,
respectively. There were no significant differences (P>0.05) in initial body weight gain, final
body weight gain, total body weight gain and average daily body weight gain among
treatment groups. The average daily weight gain (675.56 g), gross benefit (2159.96 ETB) and
marginal revenue (1.11 ETB) of T2 was higher than that of other treatment groups.
Therefore, supplementation of old oxen with 50% pigeon pea + 50% noug cake + wheat bran
+ teff straw as a basal diet was found to be better in body weight gain and benefit.

Keywords: Body weight gain, Fattening, Noug seed cake, Old oxen, Pigeon pea

120
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Cattle fattening activity is one of the potential sources for an employment opportunity and to
increase the volume and quality of meat produced and to alleviate poverty in Ethiopia
(Kerstin et al., 2011). Cattle fattening has gained prominence as an important business project
of the livestock industry in Ethiopia to make use of cheap, plentiful farm by-products
(Habtamu et al., 2008). Moreover, fattening offers the opportunity to exploit the vast cattle
and meet the export market in the region as well as Ethiopia’s domestic market (Habtamu,
2012). The fattening activity can be undertaken at any level of livestock value chain ranging
from smallholder farmers rearing livestock for multiple functions (Sarma and Ahmed, 2011),
and small scale commercial fattening to large scale commercial fattening (Tsegay and
Mengistu, 2013). Beef cattle fattening is a common practice in Ethiopia and special attention
was given by the government to boost red meat supply through cattle fattening (Agmas and
Adugna, 2018).

In Ethiopia, productivity of beef is very low which might be related to feed shortage, diseases
and parasites, drought, shortage of grazing land, market access, veterinary services, extension
services and use of low performance local breeds (Matawork, 2017). Among the major
constraints, shortage and low quality feed could be considered as an important factor. Feed is
a major factor that affects livestock productivity and contributes to nearly 65% of the
phenotypic expressions of the animals’ performance and the same percent of running costs in
pen feeding (Yeshitila et al., 2009). Therefore, to mitigate problems related to shortage of
feed particularly protein deficiency in ruminant animals, use of locally available fodder
species that can replace purchased concentrates. Among candidate fodder species the pigeon
pea is the one.

Pigeon pea (Cajanus cajan) is a plant that is cultivated for human food and animal feed. It
exists as a wide range of cultivars, and their flexibility for use in animal rearing systems have
made pigeon pea popular, especially for small-scale farmers. Pigeon pea is grown widely in
India and in parts of Africa and Central America. The main producers of pigeon pea in the
world are India, Uganda, Tanzania, Kenya, Malawi, Ethiopia, Mozambique, the Dominican
Republic, Puerto Rico, the West Indies in the Caribbean and Latin America region, Indonesia
and the Philippines and Australia (ABD et al., 2018). Pigeon pea is relatively resistant to
drought, improves soil fertility and prevents soil erosion on steep land. It’s in Eastern Africa
countries (Damaris, 2007).

121
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Pigeon pea (Cajanus cajan) has relatively good crude protein (CP) content 16.3 - 27%
(Ahmed and Nour, 1997). The dry matter and CP value of pigeon pea leaf-twig mixed were
904 g/kg and 236 g/kg, respectively (Shumuye et al., 2016). Pigeon pea has a potential to
produce high biomass ranging from 40.0 to 57.6 t DM/ha. Cattle feeding based on pigeon pea
pasture have gained in weight from 0.7 to 1.25 kg/head /day at a carrying capacity of 1 to
3.75 beasts/ha, and live weight gains of 200 to 500 kg/ha/annum have been recorded (Akinola
et al., 1975). Grain yield varies from 1500 to 3000 kg/ha/yr of shelled grain (Alemayehu et
al., 2017).

The most cattle fattening practices in Ethiopia use diverse feed resources, including
concentrates, it seems a costly venture and requires alternative feed supplements that can be
locally available. Hence, in the past, efforts have been made to identify adaptable and high
yielding forage crops for the mid and low land areas of Eastern Amhara. Among these,
Pigeon pea has been founded to be promising and recommended for the area. According to
the information obtained from the farmers, they utilize Pigeon pea as animal fattening to
some extent but farmers lack knowledge on how to use the fodder particularly the stage of
harvest and related management. This indicated that the fodder produced from pigeon pea is
not efficiently utilized. On the other hand, it is important to efficiently utilize the available
fodder to replace purchased concentrates so as to make fattening a profitable business. Hence,
this experiment was conducted to evaluate the substitution of noug seed cake with dried
pigeon pea leaves in a concentration mixture on feed intake and weight gain of local old
oxen.

MATERIALS AND METHODS

Description of the Study Area

The experiment was conducted at Haik Agricultural Research Sub-Center, South Wollo Zone
of Amhara region. The area is situated at 11°21’00’’N latitude and 39°38’00’’E longitude
located at about 435 kms North of Addis Ababa Capital City of Ethiopia. It lies within an
altitude of 1680 meter above sea level. The area receives an average annual rainfall of 1204.6
mm and a mean range temperature of 11.2 - 25.6oC.

122
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Experimental Animal’s Management and Treatments

The feeding trial was carried out for 105 days including 15 days of acclimatization period,
from April 13/2018 to July 26/2018. A total of 25 local old oxen (Bos indicus) with the
approximate age of six to seven years old (based on dentition) and mean initial live weight of
300.2±19.25 kg (mean±SD) were borrowed from the surrounding farmers. The borrower
farmers were benefited as they had got knowledge and extra income. The animals were
quarantined for twenty one days and during this period they were vaccinated for common
disease (Pasteurellosis and Black leg) in the area and treated for internal and external
parasites. The experimental design used in the study was randomized complete block design
(RCBD) with five replications.

The experimental animals were stratified based on their initial body weight and randomly
assigned into 5 treatment groups. The experimental feeds offered for oxen were:

T1 = Teff straw ad libitum + Wheat bran + Pigeon pea dried leaves (25%) + Noug seed cake (75%)
T2 = Teff straw ad libitum + Wheat bran + Pigeon pea dried leaves (50%) + Noug seed cake (50%)
T3 = Teff straw ad libitum + Wheat bran + Pigeon pea dried leaves (75%) + Noug seed cake (25%)
T4 = Teff straw ad libitum + Wheat bran + Pigeon pea dried leaves (100%)
T5 = Teff straw ad libitum + Wheat bran + Noug seed cake (100%)

Feed Preparation and Feeding

Pigeon pea was planted in the rainy season in an area where experiment was done and used as
hay for study on its optimum stage. It was harvested at 1m height. After air dried the leaf part
was lopped and packed in the sack for later use. Wheat bran and Noug seed cake for this
study were purchased from the market. The proportion of wheat bran, noug seed cake and
pigeon pea and the amount provided to the animals was calculated on a dry matter basis. All
experimental animals were allowed to feed Teff straw on ad-libitum. Supplement feeds were
provided two times a day at morning 8:00 AM and afternoon 4:00 PM. The experimental
animals were drunken water twice a day based on farmers practice and simple observation of
oxen interested in drunken water.

123
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Collection and Analysis

Initial body weight was measured at the beginning of the experiment and overnight fasted
body weight was recorded fortnightly then after. Total weight gain was calculated by
subtracting final body weight from initial body weight and average daily weight gains by
dividing total body weight gains by experimental periods. Daily feed offer and left over was
measured and recorded. The initial costs of the animals were recorded and at the end of the
experiment the selling price was estimated using three local live animal dealers. The data was
analyzed using Genstat software (18th edition) and a significant mean difference was
separated using Duncan’s multiple range test at P<0.05.

RESULTS AND DISCUSSION

Chemical Composition of Experimental Feeds

The chemical composition of experimental feeds are shown in table 1.The crude protein
content of Teff straw (5.02%) was below the maintenance requirement (7%) of adult
ruminants. The straw contains high NDF (84.2%), ADF (61.36%) and ADL (30.71%) which
affects DM intake and digestibility. The crude protein content of Pigeon pea dried leaves
(24.47%), Noug seed cake (31.92%) and wheat bran (15.98%) can satisfy adult cattle
production requirements. The NDF and ADL content of pigeon pea dried leaves, noug seed
cake and wheat bran were (53.33%, 29.29%), (64.44%, 15.15%) and (30.14%, 23.64%),
respectively.

The result indicated that Teff straw based diet required supplemented feed like browse trees
and agro industrial by products to improve the performance of mature oxen. The CP and NDF
content of Noug seed cake was higher compared with Pigeon pea dried leaves .This shows
that Pigeon pea had low fiber content than Noug seed cake and can improve intake and
digestibility. The CP content of Pigeon pea in the current study was higher as compared with
reports of 17.01%, Solomon et al. (2016), 23 % reported by Ajebu et al. (2013).

124
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Chemical composition of experimental feeds

Parameters
Ingredients DM% Ash% CP% NDF% ADF% ADL%
Teff straw 88 9.09 5.02 84.20 61.36 30.71
Pigeon pea 89 10.11 24.47 53.33 29.29 24.4
Noug seed cake 89 10.11 31.92 64.44 15.15 6.6
Wheat bran 89 12.36 15.98 30.14 23.64 13.04
DM = dry matter; CP = crude protein; NDF = neutral detergent fiber; ADF = acid
detergent fiber; ADL = acid detergent lignin

Feed Intake

The total feed and nutrient intake showed in table 2. Total dry matter, basal diet (Teff straw)
and nutrient intake had highly significant differences (P<0.01, P<0.001 and P<0.001)
respectively among treatments but supplement intake had not significant difference (P>0.05)
among treatments. The total dry matter intake ranged from 7.29 to 8.09 kg/d/head. This report
was highly disagreed with the work of Tesfaye et al. (2005) in who reported supplementation
of Oxen with Leucaena pallida had a total dry matter intake of 2.4 kg/d/head. Total dry matter
intake of treatment 1, 2, 4 and 5 were higher as compared to treatment 3. In all levels,
supplementation of pigeon pea was not significantly different among treatment in the intakes
of crude protein. The combination of ingredients teff straw + wheat bran + 50% pigeon pea +
50% noug cake was given best dry matter intake 8.09 kg/d/head than other combinations.

Table 2: Feed and nutrient intake of oxen

Treatments
Parameters T1 T2 T3 T4 T5 G mean SEM CV SL
DMI(kg/d/head)
Teff straw 4.47a 4.62a 4.00b 4.65a 4.54a 4.46 0.06 3.8 ***
Supplement 3.55 3.47 3.29 3.43 3.26 3.4 0.05 6.3 NS
Total 8.02a 8.09a 7.29b 8.08a 7.80a 7.86 0.08 3.8 **
Nutrient intake (kg/d/head)
CPI 0.99a 0.98a 1.02a 0.98a 0.79b 0.95 0.02 5 ***
NDFI 3.87d 4.35bc 4.50b 4.95a 4.12c 4.36 0.08 4.3 ***
ADFI 2.44c 2.89b 3.03b 3.50a 2.83b 2.94 0.08 5.2 ***
ADLI 0.78bc 0.69c 0.85ab 0.92a 0.76bc 0.8 0.02 8.3 ***
T = treatment; DMI = dry matter intake; d = day; CPI = crud protein intake; NDFI =
neutral detergent fiber intake; ADFI = acid detergent fiber intake; ADLI = acid detergent
lignin intake; G = grand; CV = coefficient of variance; SEM = standard error of mean; SL =
significance level; NS = non significant; ** = significant P<0.01; *** = Significant P<0.001

125
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Live Weight Gain

Initial and final body weight changes, average daily weight gain and feed conversion
efficiency Pigeon pea and concentrate mixture supplementation on Teff straw based fattening
of oxen is presented in table 3. The results showed that there were no significant differences
(P>0.05) in initial body weight gain, final body weight gain, total body weight gain and
average daily body weight gain among the different combinations of treatment groups. It is
unlikely that animals fed different levels of Pigeon pea dried leaves performed similarly. This
effect might be because the nutritional value of the pigeon pea dried leaves were better than
other supplements. The similarity in body weight gain among treatments reflects that the
supplements are comparable in their nutrient supply. Sole noug seed cake supplementation
(T5) resulted in a gain of 588.89 g/day which is almost statistically similar to those
supplemented with sole pigeon pea leaves (T4) (655.56 g/day) which indicates the
significance of pigeon pea leaves as a Crud protein supplement has a potential substitute noug
seed cake supplementation fully to fatten local oxen in the study area.

The numerical value of average daily weight gain of treatment two was the highest (675.56g)
as compared with T1, T3, T4 and T5; which had average daily weight gain values of 662.22g,
575.5g, 655.56g and 588.89g, respectively. The average daily body weight gain of this study
was significantly higher than reported by Wondimagegne and Mesfin (2012), who
supplemented oxen with different levels of Leucaena pallida was 355.6 g/head/day and
reported on Tesfaye et al. (2005) 400 g/head/day.

Table 3: Body weight change of oxen supplemented with different feeding level of pigeon
pea and noug seed cake
Treatments
Body weight change T1 T2 T3 T4 T5 OAM SEM CV SL
Initial BW(kg) 304 303.6 286.6 306.2 300.8 300.2 3.85 6.5 NS
Final BW(kg) 363.6 364.4 338.4 365.2 353.8 357.1 5.09 7.1 NS
Total BW gain (kg) 59.6 60.8 51.8 59 53 56.8 2.55 23.5 NS
ADWG (g/d/head) 662.22 675.56 575.56 655.56 588.89 631.56 0.03 23.5 NS
FCE 0.082 0.083 0.079 0.081 0.076 0.08 0.003 21.5 NS
T = treatment; OAM = over all mean; CV = coefficient of variance; SL = significance level;
SEM = standard errors of mean; BW = body weight; ADWG = average daily weight gain;
FCE = feed conversion efficiency; NS = non significance

126
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Partial Budget Analysis of Fattened Oxen

The partial budget analysis result of the current fattening experiment is shown in table 4.The
gross benefit of T2 (2159.96 ETB) was higher than others T1 (1323.18 ETB), T3 (1247.15 ETB),
T4 (1666.39 ETB), T5 (2081.04 ETB).The marginal revenue of T2 was also higher 1.11 ETB
than T1, T3, T4 and T5 with the estimated values; 1.06 ETB, 1.06 ETB, 1.08 ETB and 1.10 ETB,
respectively. The experiment indicated that fattening of an ox with combined supplementation of
50 % Pigeon pea dried leaf and 50% noug seed cake earned 2159.96 ETB which was the best cost
benefit or economically feasible feeding combination as to others supplementation levels. The
marginal revenue of T2 also indicated that one ETB investment on input for fattening of ox
provided gross profit of 1.11 ETB.

Table 4: Partial budget analysis

Treatments
Items T1 T2 T3 T4 T5
Gross benefits (ETB) 22200 22400 20600 22200 22600
Total cost that vary (ETB) 20876.82 20240.04 19352.85 20533.61 20518.96
Total feed cost(90 days) (ETB) 4554.58 4297.8 3930.61 4191.37 4636.72
Veterinary service cost (ETB) 157.92 157.92 157.92 157.92 157.92
labor cost (ETB) 764.32 764.32 764.32 764.32 764.32
Initial ox price (ETB) 15400 15020 14500 15420 14960
Net benefits (ETB) 1323.18 2159.96 1247.15 1666.39 2081.04
Marginal revenue(ETB) 1.06 1.11 1.06 1.08 1.10
T = treatment; ETB = Ethiopian birr

CONCLUSION AND RECOMMENDATION

The result of this experiment showed that Pigeon pea dried leaves can be used as the best
supplementation of crop residues. Supplementation of Pigeon pea dried leaves improves the dry
matter intake and performance of oxen in terms of body weight gain. In all treatment
combinations, Pigeon pea dried leaves were given the same daily weight gain. The result
indicated that feeding Pigeon pea dried leaves as a supplement for old oxen fully substituted noug
seed cake supplementation. Of the treatment combinations, fattening of oxen with
supplementation of 50% Pigeon pea dried leaves and 50% noug seed cake was more profitable
than other treatment combination levels. Supplementation of pigeon pea dried leaves feel oneself
revenue in ox fattening. Therefore, supplementation of old zebu oxen using feed combination
levels of teff straw (ad libitum) + wheat bran + 50% pigeon pea dried leaf + 50% noug seed cake
gave better fattening performance and profitable business which should be demonstrated for
smallholder farmers in the study area and other similar farming systems.

127
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENT

The authors would like to thank SARC and ARARI for financial and overall support. Our
special thanks go to SARC livestock research directorate researchers and Debre Birhan
Agricultural Research Center, livestock research directorate for their unreserved research
equipment collaboration.

REFERENCES

Agmas B. and Adugna M. 2018. Antimicrobial residue occurrence and its public health risk
of beef meat in Debre Tabor and Bahir Dar, Northwest Ethiopia.Veterinary World, 11
(7): 902 – 908.
Ahmed M.M.M. and Nour H.S. 1997. Legume hays as a supplement for dairy goats during
the dry season. Small Rumin. Res.26: 189-192.
Ajebu Nurfetaa, Asdesach Churfob and Aster Abebe. 2013. Substitution of pigeon pea leaves
for noug(Guizotia abyssinica) seed cake as a protein supplement to sheep fed low quality
tropical grass hay. Ethio. J. Appl. Sci. Techno. Vol. 4(1):1-13.
Akinola J.O. and Whiteman P.C. 1975. Agronomic studies on pigeon pea (Cajanus cajan
L.Millsp.) Field responses to sowing time. Aus. J. Agri. Res.26:43-79.
Alemayehu Mengistu. 2002. Forage production in Ethiopia a case study with implications for
Livestock production. Ethiopian Society of Animal Production. Addis Ababa, Ethiopia.
Alemayehu Mengistu, Gezahagn Kebede, Getnet Assefa and Fekede Feyissa. 2017.
Descriptions and characteristics of cultivated forage crops growing under different Agro
Ecological Zones in Ethiopia. Inter J Agri Biosci, 6(5): 238-247.
Damaris A.O. 2007. The potential of pigeon pea (Cajanus Cajan (L.) Millsp), In Africa,
Natural Resources Forum 31, 297–305.
Habtamu A., Mulugeta K., Chala M., Arse G. and Ulfina G. 2008. Proceedings of the 16th
Annual Conference of the Ethiopian Society of Animal Production (ESAP) held in Addis
Ababa, Ethiopia.
Habtamu Y. 2012. Risk analysis in cattle fattening in North West Ethiopia: Empirical
evidence form two liemittobit model. International Journal of Development and
Sustainability 1: 240-254.
Kerstin K., Zander A. and Ellengowan D. 2011. Attitudes of Livestock Keepers to Breeding
Strategies – Threats and Opportunities for on-farm Conservation of the Borana Cattle
Breed. Journal of Agricultural Science: 3.

128
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Matawork Milkias. 2017. Beef Cattle Production Systems, Marketing and Constraints in
Ethiopia. Journal of Marketing and Consumer Research. ISSN 2422-8451 An
International Peer-reviewed Journal Vol.32.
Abd El-Hack M.E., Swelum A.A., Abdel-Latif M.A., Más Toro D. And Arif M. 2018. Pigeon
Pea (Cajanus cajan) as an alternative protein source in broiler feed. World’s Poultry
Science Journal, Vol. 74.
Sarma P.K. and Ahmed J.U. 2011. An economic study of small state cattle fattening
enterprises of Rajbari district. Journal of Bangladeshian Agriculture Univ 9: 141-146.
Shumuye Belay, Guesh Godifey and Minister Birhanea. 2016. Effect of graded level of
pigeon pea (Cajanus cajan) leaves on growth performance of sheep. International Journal
of Agricultural Science Research Vol. 5(2), pp. 035-039.
Solomon T., Wondimagegne B., Desalegn A., Asress Z., Niguse B. and Misganaw W. 2016.
Evaluation of the effect of Pigeon pea dried leave supplementations on feed intake and
live weight change on local sheep breed in Eastern Amhara, Ethiopia. Int. J. Adv. Res.
Biol. Sci. (2016). 3(12): 81-85.
Tesfaye Lemma, Gebregziabher Gebreyohannes, Chala Merera, Jiregna Desalegn, Girma
Aboma and Diriba Geleti. 2005. Comparative Feeding Values of Leucaena Pallida and
Noug Cake (G. Abyssinica) for Fattening Horro Steers. Eth. J. Anim. Prod. 5(1) 2005:
67-76.
Tsegay T. and Mengistu U. 2013. Assessment of commercial feedlot finishing practices at
eastern Shoa, Ethiopia. Open Journal of Animal Sciences 4: 273-280.
Wondimagegne Bekele and Mesfin Dejene. 2012. Feed Intake and Body Weight Change of
Highland Zebu (Bos Indicus) Oxen Fed Teff (Eragostis Teff) Straw Supplemented with
Different Levels of Leucaena Pallida and Wheat Bran. In: Indigenous peoples and
Animal Agriculture in Ethiopia: Exploiting the potential and Reducing Limitations.
Proceedings of the 19th Annual Conference of the Ethiopian Society of Animal
Production (ESAP) held in Addis Ababa, Ethiopia, December 15 to 17.
Yeshitila A., Tessema Z. and Azage T. 2009. Utilization of feeds livestock unit versus dry
matter Requirement in Alaba, Southern Ethiopia. Pp.21-32. proceeding of the 16th
annual conference of the Ethiopian society of animal production (ESAP). Held from 8-
10 October 2008. Addis Ababa, Ethiopia. ESAP, Addis Ababa, Ethiopia.

129
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Feed Resources, Nutritive Value and Alternative Balanced Formulated Diets for
Lactating Dairy Cows at Andassa Livestock Research Center, Ethiopia

Shigdaf Mekuriaw1, 2, Addisu Bitew1, Biadiglign Hunegnaw1 and Firew Tegegne3

1
Andassa Livestock Research Center, P.O. Box 27, Bahir Dar, Ethiopia
2
United Graduate School of Agricultural Sciences, Arid Land Research Center, Tottori
University, Japan
3
Bahir Dar University, P.O. Box 5501, Bahir Dar, Ethiopia

ABSTRACT

The study aimed to evaluate the nutritive value of the available feed resources, milk yield,
and estimated methane emission of lactating dairy cows at the Andassa Livestock Research
center dairy farm. The feed samples were collected from the available feed resources on the
farm. The chemical composition and in-vitro analysis were conducted at Shimane University
(Japan) following the standard procedures of animal nutrition laboratory. The mineral profile
of the feedstuffs was analyzed using inductively coupled plasma mass spectrometry (ICP-
MS). The milk yield was recovered from the mid lactating dairy cows of Fogera and Holstein
Friesian crossbred dairy cows. The major feed resources in the farm were grazing pasture and
natural hay used as roughage feed while the concentrate mix used as supplementary feed. The
chemical composition of the feeds has wide variation such as the crude protein (CP) value of
natural pasture was less than the critical value (70 g/kg DM) for ruminant animals for
optimum activity of rumen microorganisms whereas the concentrate and Napier grass feeds
have above the minimum CP level required for lactating cows. The concentrations of macro
and micro minerals in all feedstuffs were above the requirements of dairy animals except for
sodium, copper, and cobalt. Based on the milk performance evaluation of the current feeding
practice for local breeds in the farm (2.0 kg/day/cow), it is performing below the estimated
potential milk yield of the dairy cows (4.8 kg/day/cow). Thus, to improve the nutritional
status of the lactating dairy animals, protein requirements have to be fulfilled (from 93.92 to
126.7 g/kg CP). This is supported by the ration formulation done based on the protein level
requirement of lactating dairy cows according to national research council NRC (2001).
Based on this, alternative modified formulated diets (MFD1, 2, and 3) were prepared for the
dairy cows in the assumptions to increase the potential milk yield. Based on these new diets,
the maximum daily protein corrected milk yield was more than 10 kg with less estimated

130
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

methane emission per dry matter intake (13.39 g/Kg DMI) for Holstein Friesian crossbred
while 5 kg for local breed lactating dairy cows with 18.55 g/Kg DMI methane emission.
Moreover, soya bean cake and/or alfalfa can be used as a protein corrector to achieve the
maximum milk yield on the farm. Regarding the least cost feed formulation that optimized
milk production, MFD 2 was recommended followed by MFD 3 and MFD 4. In conclusion,
formulated diets indicate that in the combination of high-quality feed ingredients it can be
possible to increase the milk yield as well as reduce methane emission. Further research is
recommended to validate and evaluate the newly formulated diets through an in-vivo feeding
trial using lactating dairy cows.

Keywords: Dairy Farm, Formulated diet, Methane emission, Milk yield, Ration Formulation

INTRODUCTION

The Ethiopian livestock master plan has been designed as a strategy to alleviate the feed
shortage situations by evaluating the local feeds resource (Shapiro et al., 2015). In this regard,
participatory identification and evaluation of local feed resources are essential (Talore, 2015).
It is therefore imperative to examine for cheaper feed resources that can improve the intake
and digestibility of low-quality forages (Negesse et al., 2009). However, their chemical
composition is variable due to variations in soil type, topography, and other environmental
factors that affect the forage yield and quality (Gemiyo et al., 2013). The nutritive value of a
ruminant feed is determined by the concentrations of its chemical components, as well as
their rate and extent of digestion (Getachew et al., 2004). Information on the nutritional
characterization of locally available feed resources at the country level is inadequate, and
where available the values are variably documented (Seyoum and Zinash, 1998). High
diversity, variability, and nutritional values of feeds in the northern region, in particular, have
not yet been investigated and their feeding value is unknown (Gemiyo et al., 2013).

Poor quality feeds and feeding systems could lead to reduced milk production and increase
methane emission (Niu et al., 2018). In the predominant small scale, subsistence farming
systems in Ethiopia, most of the farmers cannot provide a balanced diet supplementation for
their animals (Ayalew et al., 2003). Hence, knowledge regarding the nutrient composition of
different feedstuffs helps in the preparation of balanced rations for animals (Talore, 2015).

131
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In developing countries, farmers have a wide number of locally available cheap ingredients,
but there is a lack of experience on balanced ration formulation to incorporate these
potentially nutritious feedstuffs. Economic analysis indicated that formulating diets more
frequently based on a least-cost feed that generally increased profitability because any
increase in feed cost was mitigated by an increase in profitability through improved milk
yield (White and Capper, 2014). The linear programming model is certainly the most popular
method for diet formulation that can solve for a complicated set of nutrient requirements to
give a relatively well-balanced ration (Chakeredza et al., 2008). Furthermore, based on the
chemical composition of available feed resources and as per the nutrient requirement of the
animal, the software computes the least-cost ration within the given constraints (Garg and
Makkar, 2012).

Therefore, the feed formulation can allow livestock producers to utilize economically and
environmentally effective by-products as local feed resources (Oishi et al., 2011). Database
on nutritive values, access, and use of such information will have a significant impact on
improved animal performance and productivity (Waghorn and Clark, 2004). The ultimate
goal of feed analysis is to predict the productive response of animals when they are fed
rations of a given composition (NRC, 2001). This is the real reason for the necessity of
information on feedstuff composition. This paper aimed to evaluate and characterize the feed
resource and nutritional evaluation perspective toward ration formulation in the ALRC dairy
farm.

MATERIAL AND METHOD

Description of the Study Area

The study area is found about 22 km south of Bahir Dar city on the way to Tis Abay; located
at 11°29’N latitude and 37°29’E longitude with an elevation of 1730 m above sea level in
Amhara Region, Ethiopia (Kebede et al., 2013). The area has dark clay soil, which is
seasonally waterlogged and cracked when dry (Sewalem and Banjaw, 1992). The area
receives about 1434 mm of rainfall annually. The mean annual temperature varies from a
maximum of 29.5oC in March to a minimum of 8.8oC in the wet season. The dominant grass
of the area includes Cynodon spp., Hyparrhenia spp., Paspalum spp., Pennisetum spp., Setaria
spp., Eleusine spp., Eragrostis spp., Sporobolus spp., and Andropogon spp. and Trifolium
species (Yihalem, 2004).

132
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Dairy Cattle Herd Management

The center mainly conserves and improves the Fogera cattle breed and also crossbred with
Holstein Friesian. Currently, there are 611 animals on the farm (Table 1). The Fogera breed is
known for its tolerance to parasitic and disease infestations, fly burden, wet soils or swampy
areas, and other unfavorable environmental conditions (Alberro and Haile-Mariam, 1982;
Tesfa et al., 2016). In the breeding program, both natural mating with Fogera bulls and
artificial insemination with Friesian semen was used. Health management practice had
prevention and control schemes. The prevention scheme focused on vaccination against
anthrax, blackleg, and pasteurellosis once in every six to eight months and once per year for
Contagious Bovine Pleuropneumonia (CBPP). Control measures were also taken against
internal and external parasites (ALRC Progress report, 2018). All animals of the farm are
subjected to graze during the day for eight hours in their herd and kept by individual day
watchmen. During night time, the herds are yarded separately in their respective barn
designed as a loose housing system and watched by guards. In addition to grazing, all the
herds also feed conserved hay collected from the pastureland; milking cows are supplemented
with concentrate feed, which consists mainly of a mixture of maize (Zea mays) and noug
(Guitozia abyssinica) seed cake. The animals watered from Andassa river and spring water
during the wet season for young and sick animals which stayed at the barn (Tesfa et al.,
2017).

Table 1: Dairy cattle herd size in the farm

Young Lactating Pregnant Dry Adult
Breed type Calf Heifer Bull Cow Cow cow Bull Total
Fogera 145 63 65 48 80 110 30 541
HF x Fogera 39 14 - 14 - 3 - 70
Total 184 77 65 62 80 113 30 611
Source (ALRC, progress report, 2018).

Feed Sample Collection and Chemical Composition Analysis

The feed samples were collected from the available feed resources in the farm in wet and dry
seasons. For commonly available feeds, composite samples were taken for further chemical
analysis. The feed samples were analyzed for DM (method no. 930.15) according to the
procedures of AOAC (2000).

133
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin
(ADL) were determined according to the method of Van Soest et al. (1991). After
determining the total nitrogen and carbon of the feedstuffs by dry combustion in carbon and
nitrogen (CN-coder, Yanagimoto Co.), the Crude Protein (CP) content was calculated as N x
6.25. The concentration of Gross Energy (GE) was determined with a bomb calorimeter (CA-
4AJ, Shimadzu, and Kyoto, Japan). Ash content was determined by incineration at 600°C for
2 hours in a muffle furnace (AOAC, 1990). The contents of minerals were analyzed by
Inductively Coupled Plasma Mass Spectrometer (ICPMS) after digestion of the feed samples
with a mixture of nitric acids according to the procedures described by AOAC (1995). All the
chemical analysis was conducted at Shimane University Animal Nutrition Laboratory, Japan.

In Vitro Gas Production and Metabolizable Energy Content Analysis

In vitro gas production (IVGP) was measured with syringes according to the method
described by Menke and Steingass (1988). Rumen fluid was obtained from two healthy
mature Japanese Corriedale female sheep fitted with a permanent 70 mm rumen cannula that
was fed a daily ration of 800 g timothy hay and 200 g concentrates divided into two equal
portions at 8:00 and 16:00 daily at Shimane University, Japan. The rumen fluid was collected
in the morning 3 h after feeding, flushed with CO2, filtered through three layers of cheese-
cloth and mixed (1:2, v/v) with an anaerobic mineral buffer solution as described by Makkar
et al. (2000). Preparation of an in vitro mineral buffer media for the gas test was carried out
as described by Menke and Steingass (1988). Briefly, 200±0.43 mg of 1 mm milled feed
samples were weighed into 100 ml calibrated glass syringes in duplicate. About 30±0.5 ml of
rumen-buffer mixture was added into each syringe and then all the syringes were incubated in
a water bath maintained at 39±0.1°C. The syringes were gently shaken every hour during the
first 8 hr and reading gas was recorded after 3, 6, 12, 24, 48, 72, and 96 hrs. Total gas values
were corrected by blank incubation. The in vitro gas production data were fitted with the
‘Neway’ curve-fitting program of Macaulay Land Use Research Institute (2004) to estimate
the rumen degradation parameters based on the model of Orskov and McDonald (1979);

y = a + b (1 – e–ct) …………. (1)

Where: a is the gas production from the immediately soluble fraction (ml), b is the gas
production from the insoluble fraction (ml), c is the gas production rate constant, t is the
incubation time and y is the potential gas production (ml) at the time t.

134
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The metabolizable energy content (MJ/kg DM) and OMD (g/kg) of feedstuffs were
calculated using equations suggested by Menke et al (1979) equation (2) and (3):

ME (MJ/kg DM) = 2.20 + 0.136 GP + 0.057 CP + 0.0029 CP ….......... (2)

24
2

IVOMD (g/kg DM) = 148.8 + 8.89 GP + 4.5 CP + 0.651XA ………… (3)

24

Where: GP is 24 h net gas production (ml/200 mg DM), CP is Crude protein (%), and XA is
Ash content (g/kg).

Net energy maintenance (NEm) and Net energy lactation (NEl) was calculated as per the
following equation (NRC 1989):

NEm = ME (0.554 + 0.287ME/GE) …………. (4)

NEl = ME (0.4632 + 0.24ME/GE) …………… (5)

Where: Metabolizable Energy (ME) was calculated from in vitro gas fermentation

Effective dry matter rumen degradability (EDMD) of the feedstuffs was estimated from the
equation developed by (Kamalak et al., 2005):

Y = 24.7 + 0.358 (a + b) gas– 60.3cgas (R = 98.3; RSD = 0.780; Sig. ***) …. (6)
2

Where: c is the rate of dry matter degradation, a is the quickly soluble fraction, b is the
insoluble but fermentable fraction and Sig. is significance level.

Estimation of Methane Emission

Estimated methane production was calculated with the developed intercontinental CH4 yield
prediction (g/kg DMI) prediction equations which can be used in a wide range of diets and
production systems within regions and globally since there is no any specific equation
developed for this breed;

Methane (CH4/DMI (g/kg) day per cow) = 13.8 (0.63) + 0.185 (0.0133) x NDF... (7)

Where: DMI is dry matter intake and NDF is neutral detergent fiber (Garg et al., 2012).

135
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Volatile Fatty Acid Production

At the end of the 96 hr incubation of the last run, 5 ml of supernatant was collected and stored
for analysis of volatile fatty acid (VFA) production. Samples were centrifuged at 25000 g for
15 min at 4°C and a part of the supernatant was transferred to a microcentrifuge tube
(capacity 1.5 ml) and stored at -20°C until analysis by gas chromatography. Our standard
solution of VFA mixture consisted of acetic, propionic, butyric and isobutyric acids, and was
treated in the same manner as the sample. The final VFA concentrations in samples were
reported after deducting the corresponding blank values.

Monitoring Data Collection

Data were collected from the center through the year including herd size, milk production per
cow, body weight, milking and feeding system. The types and amounts of feed offered to
animals were recorded for each dairy cow daily. Fresh matter feed intake of lactating cows
was taken from the farm records. The major feed resources of the farm and feeding practice
were assessed in terms of their utilization as a basal diet and supplements. Moreover, the
major problems in the feeding management of the farm were identified and possible
suggestions for the improvement of the dairy farm were proposed.

Least Cost and Alternative Balanced Diet Formulation for Dairy Cows

Feed formulation was done with the linear programming (LP) tool “Solver” (Saxena and
Khanna, 2014). The LP model includes constraints the requirements for dry cow NE, CP,
least cost of the ration and dry matter intake (Table 2) since adding those constrained
parameters (Table 2) improve the result of formulation (Oishi et al., 2011). The energy,
protein and mineral requirement of dairy cows was referred from NRC (2001) (Table 3) and
the chemical composition of the feed ingredients from this study result (Table 4). The dry
matter intake estimated on the metabolic body weight of the cow based on NRC (2001).

The calculation and data analyzed done using Excel “Solver” software, 2016.

136
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Equations of constraints for formulated ration in Solver Linear Programming

Constraints Equation
Dry matter intake (DMI) ≤DMIm+DMIl
Net energy requirement (NE) = NEm+Mel
Crude protein requirement (CP) ≥CPm+CPl
Calcium requirement (Ca) =Cam+Cal
Phosphate requirement (P) =Pm+Pl
Roughage proportion (%) < 70%
For maximum intake Rumen capacity = 15% of the Body weight
Least cost of the ration ≤ 6 (ETB/kg DM)
DMI, dry matter intake; NEm, net energy maintenance; NEl, net energy for lactation; CPm,
crude protein for maintenance; CPl, crude protein for lactation; Cam, calcium for
maintenance; Cal, calcium for lactation; Pm, phosphorus for maintenance; Pl, phosphorus
for lactation.

Table 3: Nutrient Requirement Tables for Lactating Dairy Cows (NRC, 2001)
Energy and protein requirements for dairy cattle
Maintenance:
Energy 0.080 Mcal NE / kg W0.75
Protein 7.45 g CP/ kg W0.75
Lactation: (0.0929 * fat% + 0.0547 * protein% + 0.0395 * lactose %) Mcal
Energy NE/kg milk
Protein 83 g CP/kg milk
NE, net energy; CP, crude protein

RESULTS AND DISCUSSION

Available Feed Resources in the Dairy Farm

Roughage Feed Sources

The main roughage feed sources were natural pasture (grazing and grass hay) as a basal diet.
In addition, improved forage (elephant grass) was also used in some cases when available
especially in the dry season. The farm has ample natural pasture grazing land which is
accessible throughout the year through rotational grazing. Out of the total 360 ha of land
holding of the center, nearly 320 ha is under natural pasture. Of these, 170 ha is used for
continuous grazing whereas 150 ha is seasonally grazed or partially used for hay making
since it is a water logged area during the rainy season. Moreover, about 5 - 10 ha area is
cultivated and over-sown for forage production and natural pasture improvement (Yihalem,
2004). The native grass hay is the major dry roughage source harvested annually on average
10,000 bales (1 baled hay equivalent to 25 Kg). In addition, elephant grass is cultivated in
around 2.5 ha of land for improved forage development by irrigation. Moreover, Desmodium,
Alfalfa, Vetch, Oat, Sesbania and Stylo are sources of feed to the animals in the center rarely.

137
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Concentrate Feed Sources

The main concentrate feed ingredients used in the dairy farm were a mixture of maize grain,
noug seed cake, wheat bran and salt with the proportion of 40%, 30%, 28% and 2%,
respectively (ALRC progress report, 2017). Similarly, Tewodros (2015) also reported that
69% of maize with 29% of noug seed cake was supplemented for animals after mixing with
2% salts. Recently, the center installed a big grinding and mixer Feed mill machine which
was bought from Ottevanger Company, Holland. The feed mill machine has a proven
capacity grinding and mixing of 3 ton/hour. This encourages and enhances the utilization of
formulated concentrate feed by purchasing the ingredients.

Feeding Practice

The feeding practice mainly depends on grazing from natural pasture land and hay made from
the natural pasture land. Cattle are allowed to graze throughout the year. Mostly the animals
graze for eight hours daily. Feeding of hay mostly starts from March when the grazing
pasture deteriorates until the pasture land recovers. Recently, it is encouraged that there is
feed trough construction for the hay and fresh grass feeding the dairy cows in groups.
Improved forage, mainly Napier grass is provided to milking cows, sick animals and growing
calves during the dry season.

Water Supply

Water mainly from pipe is available at all times in the farm. Crossbred dairy animals were
watered on a watering trough, which is constructed out of their house (Barn) to serve the
animals divided in groups at a time. The animals can drink on both sides of the trough
without any difficulty. The trough is cleaned daily before the crossbred dairy animals begin
drinking. In contrast, there is no provision of clean water supply for Fogera cattle because of
their large number in the herds and difficult to manage; instead, they were watered at
Andassa River throughout the year (Personal observation and communication with the
herdsmen of the dairy cattle). Due to this poor water quality, animals were frequently infested
with GIT parasites (Tewodros, 2015). The water supply provision for both breeds is ad
libitum.

138
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Supplementation

There is concentrate supplementation for individual animals according to stage of lactation

and pregnancy mostly in dry season and sometimes in wet season. In addition, concentrate
feed is also provided for sick animals that stayed around the barn and selected under
conditioned animals to maintain their body condition. However, concentrate feeding depends
on availability of budget. Besides, green forage supplementation for the animals are kept
outdoors and in groups on the locally constructed feed troughs.

Feeding of Different Group of Cattle Herds

Calf Feeding Practices: the Fogera breed calves had free access to suckle their dams for the
first five days to ensure that they consume enough colostrum. They were then separated from
their dams and allowed to partially suckle (two teats) at milking times until weaning (Bitew
et al., 2010). While cross breed calves have no access to suckle their dams but they were fed
1.3 - 2 liters of milk individually per day through a bucket feeding system. After two weeks
of age calves were provided with succulent forages to adapt them to rumen fermentation and
gradually they were also provided other feeds such as hay and concentrates. Calves stayed
around the barn until weaning age (three months) and allowed for grazing thereafter by
joining the herd group of the young calves. For this study, it is difficult to evaluate the
feeding of calves because there is no any feeding regime developed for them. So, it is highly
advisable to prepare a standard feeding package for calves.

Feeding Heifers, Bulls and Dry Cows: feeding mainly depends on grazing but they were
supplemented with hay in the dry season and improved forage when available once per day.
Breeding bulls were offered 2 kg of concentrate per head per day at mating time but it was
not based on the protein and energy requirement for reproductive performance.

Pregnant and Lactating Cows: the pregnant (after second trimester stage) cows are usually
separated from their herd and supplemented with hay, 2 kg of concentrate per day to increase
their milk yield and improved forage when available.

139
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Chemical Composition and In Vitro Digestibility of the Feedstuffs in the Dairy Farm

The chemical composition such as the CP value of natural pasture was less than the critical
value (70 g/kg DM) for ruminant animals for optimum activity of rumen microorganisms
(Alkhatib et al., 2016; NRC, 2001). This might be due to the nutritional quality of natural
pasture has deteriorated due to seasonal effect (dry season, continuous (overgrazing) and lack
of range improvement interventions (Yihalem, 2004). However, the concentrate feed and
Napier grass feedstuffs in the farm were above the minimum CP level required for adequate
rumen function in ruminants (Van Soest et al., 1991). In addition, the minimum CP content
required for lactation and growth of cattle is 150 g/kg DM (Norton, 1981) suggesting that the
concentrate feed was almost satisfying the recommended value. While the Napier grass could
provide a minimum requirement for lactation (120 g CP/ kg DM) and growth (113 g CP /kg
DM) in cattle (NRC, 1981). This implies that the major feed resource grazing pasture and hay
should be supplemented to fulfill the requirement of dairy cows.

The threshold level of NDF in tropical grass beyond which DM intake of cattle is affected is
600 g/kg (Meissner et al., 1991). So most of the feed types have acceptable NDF values
except for the grazing pasture and hay (Table 4). In this study the digestibility of natural
pasture is low as it might be with their high fiber content to limit feed intake. Thus, the hay
residues require some degree of supplementation or treatment to support optimum animal
performance (Bogale et al., 2008). Oilseed cakes such as soya bean cake can be used as
sources of rapidly degradable nitrogen source (Salem et al., 2005).

Table 4: Chemical composition of feed ingredients (Mean)

Feed Grazing Natural grass Green fodder (Napier
characteristics Pasture Hay Concentrate grass)
CP (g/kg) 69.7 66.5 152.2 127.8
NDF (g/kg) 700.1 694 412 423.12
ADF (g/kg) 326.1 398.4 177.7 340.5
Ash (g/kg) 110.4 140 104.4 146
GE (MJ/kg) 15.6 14.9 16.2 15.2
ME (MJ/kg) 8.51 8.49 9.54 9.84
NEM (Mcal /kg) 6.11 5.95 6.95 7.22
NEL (Mcal /kg) 5.11 4.97 5.81 6.04
CP, crude protein; NDF, Neutral detergent factor; ADF, acid detergent factor; GE, gross
energy; NEM, net energy for maintenance; NEL, net energy for lactation.

140
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In Vitro Organic Matter Digestibility (IVOMD) and Metabolizable Energy (ME)

The in vitro gas production and IVOMD result of the feed ingredients (Figure 1) showed that
concentrate and Napier grass has a higher digestive value than natural pasture. The high
IVOMD values of concentrates might be due to differences in the chemical composition of
the feeds or in methods of removing attached microbial biomass (Getachew et al., 2004). The
grazing pasture has higher organic matter digestibility than crop residues such as Faba bean
straw (46.9 g/kg) (Asar et al., 2010). The ME values of the feeds were within the ranges
reported by Seyoum and Zinash (1989) (Figure 2) and it satisfies the requirement of lactating
dairy cows.

Figure 1: Gas production of the feed resources

Figure 2: IVOMD and ME of the feed ingredients

141
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Macro and Micro Minerals Concentration

The concentration of macro minerals has high variation among the feedstuffs in the farm
(Table 5). The highest concentration of Ca (5.01 g/kg) in concentrate feed while the lowest is
in hay (2.82g/kg). Grazing pasture has the highest potassium and sulfur concentration while
pasture hay has the lowest sodium and potassium concentration. Most of the micro minerals
(Fe, Mn and Zn) were found in highest concentration in the feed ingredients except Cu and
Co. The deficiency of Cu might be due to the low concentration in the soil (Gowda et al.,
2004).

The concentration of calcium, magnesium, potassium and phosphorus was sufficiently high
to meet the requirements of dairy animals in the study farm, while sodium concentration was
low. In agreement with this study, Zafar et al. (2009) reported that pasture grasses/ forages
have sufficient levels of K, Ca, Mg, Mn, Fe and Zn to meet the requirements of ruminants.
Almost similar results were obtained in other parts of the world where Ca contents of the
pasture ranged from 1.8 to 9.8g/kg (Aregheore et al., 2007; Reshi et al., 2013).

The agro-industrial byproducts, hays or crop residues are low in calcium. On the other hand,
the agro-industrial byproducts contain more phosphorus than calcium (Adugna, 2008). This
implies cows might lead to calcium-deficiency. Feeding a calcium-deficient diet may delay
uterine involution. High phosphorus intakes along with low calcium intakes also depress
fertility (Funston, 2007). While dietary Ca concentrations of 2 - 6 g/kg, with higher
requirements for lactation have been variously recommended for cattle (NRC, 1985:
(Shewangizaw and Bishoftu, 2014). The Ca contents were very high compared to the
recommended requirement of 4.3 g/kg in DM for cattle, whereas the P contents were equal
and/or almost above the minimum requirement of 1.7 g/kg in DM for grazing ruminants
(McDowell, 1985).

Table 5: Macro and micro mineral composition of feed ingredients (mean)

Macro Mineral (g/kg DM) Micro Mineral (mg/kg)
Feed types Ca K Na Mg P S Fe Mn Zn Co Cu
Grazing Pasture 4.06 48.43 1.11 7.08 9.71 1.94 766 29 36 5.0 2.0
Hay 2.82 32.14 0.27 5.63 3.74 3.45 65 45 24 1.2 0.7
Concentrate 5.01 27.71 20.57 7.35 12.98 6.09 60 17 77 1.0 1.0
Napier grass 2.8 24.08 0.46 10.14 4.48 4.16 62 9 14 1.4 1.1

142
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Volatile Fatty Acid Production

The acetate, propionate, Iso-butyric acid, n-butyric acid and total VFAs production of feed
resource (mmol VFA/g dry matter incubated) was demonstrated in (Figure 3). The lowest
VFA production was with natural pasture while the concentrate has the highest acetic acid
and total VFA production among the feed types.

Figure 3: Concentration of volatile fatty acid (VFA)

Alternative Formulated Balanced Diets for Optimum Milk Production

The alternative modified feed formulations have increment in estimated milk yield and
reduction in estimated methane emission as compared to the control diet (Table 6). The
increase in NDF levels in the control diet reduced the maximum intake and estimated milk
production than the formulated diets. Differences in calculated daily CP supply is less in the
control diet (93.92g/kg DM) than the recommended level for lactating cows (120g/kg DM)
due to the low CP content of the roughage in the diet. This implies that the control diet for the
high yielding cow was less adequate for the expected milk yield in the farm. The estimated
methane emission per DMI in the formulated diets has decreased by 40% than the control
diet. This might be due to the high content of the CP and nutrient digestibility that enhances
fermentation in the rumen. The major constraint in existing (control) diet is that the less dry
matter intake because of the high fiber value of the feed in which high yielding cows received
less adequate diets.

143
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A dairy cow producing 10.95 kg milk per day required 928.5 g/day of CP (NRC, 2001). This
implies that the protein supply by the control diet is less sufficient for high yielding cows but
through the supplementation of improved forage (Napier grass and alfalfa) or concentrate
mixture in the modified formulated diets (MDF1, 2, 3 and 4), the cow’s protein requirement
for optimum milk production could be fulfilled. The balanced diet formulation also enhance
the utilization of poor quality feedstuffs by their incorporation in complete diet using total
mixed ration (TMR) rather than feeding separately along with concentrate mixture to meet
the maintenance, growth and lactation requirements of the dairy cow (Beigh et al., 2017).
Therefore, under production and nutritional aspects, fiber levels around 500g/kg NDF are
recommended for dairy cows.

The calcium supply in the modified formulated diet was in agreement with the
recommendation of NRC (2001) that 10 kg milk /day producing 500 kg weighing standard
cow requires 30 g Ca per day (ARC, 1994). Similarly, the recommended phosphorus feed
supply is 1.65 g/kg of milk (ARC, 1994) in which the maximum 12.5 kg milk producing cow
in this formulated diet might be supplied up to 23 g/day. In brief, this feed formulation clearly
indicated that locally available feeds resources can significantly contribute to the nutrients
requirement of dairy animals. It was depicted from this formulation that mixing two or more
feed resources with different protein and energy content can yield a good balanced diet for
high yielding dairy cows (Ja’afar-Furo, 2011).

144
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 6: Feed ingredient composition of the alternative balanced formulated diet (mean)
Item Local Fogera breed HF Crossbred
Ingredients Control diet MDF 1 MDF 2 MDF 3 MDF 4 Cont.diet MDF 1 MDF 2 MDF 3 MDF 4
Hay (kg/d) 3.05 2.27 2.01 2.02 2.01 4.07 3.03 3.01 2.97 2.63
Concentrate Mix (kg/d) 1.43 2.09 1.95 2.08 1.95 1.92 2.79 2.91 2.98 2.71
Napier grass (kg/d) - 1.78 1.03 0.98 1.03 - 2.38 1.55 1.58 1.28
Noug seed cake (kg/d) - - 1.16 - - - - 1.73 - -
Alfalfa (kg/d) - - - - 1.16 - - - 2.37 -
Soybean Cake (kg/d) - - - 1.04 - - - - - 1.36
Ca (g/day) - 19.00 20.00 21.00 23.00 21.00 24.00 26.00 27.00 28.00
P (g/day) - 12.5 16.5 17.5 19.5 19.00 20.00 21.00 22.00 23.00
Chemical composition of diet
NDF (g/kg) 603.76 519.5 502.5 502.9 502.5 603.76 519.5 502.5 502.9 502.9
ADF (g/kg) 243.86 260.9 246.9 233. 246.9 243.86 260.9 246.9 233.1 233.1
NEm (Mcal/Kg DM) 1.47 1.38 1.40 1.47 1.40 1.47 1.38 1.40 1.47 1.47
NEl (Mcal/Kg DM) 1.41 1.38 1.42 1.45 1.42 1.41 1.38 1.42 1.45 1.45
CP (g/kg) 93.92 113.4 126.7 141.0 126.7 93.92 113.4 126.7 141.0 141.0
Calculated Estimated value
DMI (kg DM/d DM) 6.62 6.15 6.78 7.03 7.68 10.0 8.2 9.1 11.5 13.6
Milk yield (Kg FPCM) 2.09 3.00 3.89 4.5 4.8 4.1 5.8 8.5 11.2 12.5
CH4 (g/Kg DMI) 28.1 20.91 19.38 19.22 18.55 20.95 14.44 13.22 13.73 13.39
Feed cost (ETB/kg) 3.58 3.02 2.81 2.94 2.91 3.98 3.67 3.21 2.5 2.61
Income (ETB/kg Milk) 8.66 13.81 15.10 15.41 15.34 10.29 14.81 16.56 17.43 17.16
MFD, Modified Formulated Diet; FPCM, Fat Protein Corrected Milk; CH4, Methane

145
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The Way Forward

Based on the evaluation and characterization of the farm in the perspective of nutrition, the
following issues should be considered in order to improve the feeding management of the
farm and action should be taken soon.

Identified Problems Related to Feed Characterization

• The farm has a lack of proper feeding recording system and ingredients proportion during
supplementation in particular for fresh and green feeds.
• The immense problem of the dairy farm is that the feeding strategy is not following the
proper guidelines for ration formulation of the dairy cows according to NRC (2001). Due
to this; there is much loss of feed lying on hindering the potential production of dairy
cows.
• Supplementation of improved forages, green legume forages mixing hay in dry season:
since the farm feed the hay only in the dry season; it is better to feed by mixing the
legume with hay to increase the intake as well as digestibility in order to supply sufficient
nutrient for the dairy cow. It is also highly encouraged to enhance the production of
improved forage using irrigation schemes to make betterment of roughage feed utilization
to get quality grass hay especially in the dry season to improve milk production.

Established Computerized and Organized Data Collection System

The farm data recording should be well organized and it must include the average body
weight of each cows at different lactation stages and average milk yield of cows at different
lactation phases (0 - 60 days post calving, 60 - 120 days post calving, 120 - 200 days post
calving, 200 - 305 days post calving and more than 305 days post calving) with the daily feed
intake. In addition, the average milk yield composition should be measured at different
phases based on the feeding regimen.

Capacity Building

Training on practical skills based on the technique of feed formulation and feed management
should be given for the farm workers.

146
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Balanced Feed Formulation on the Basis of Targeted Milk Yield

Based on the actual milk production of the cows and the feed requirement, balanced ration
formulation should be done using the locally available feed resource in the farm. This is one
of the best nutritional strategies to enhance the milk production of dairy cows at the farm. In
this study, we suggested the alternative formulated balanced diets (Table 5) and evaluation of
the formulated diets on dairy cows should be done. It is also more preferable to prepare the
feed calendar for the feed resources and feeding practice in the farm throughout the year.

CONCLUSIONS

The major sources of feed to animals in the research center are grazing pasture, hay,
concentrate and improved forage. Natural pasture has poor quality due to low protein and
high fiber components hence their digestibility is low to fulfill the maintenance requirement
of the dairy animal. On the other hand, the concentrate feed and Napier grass have high CP
value greater than the maintenance requirement (>70 g/kg) with better digestibility that can
be used as feed supplementation to increase levels of dairy production. The concentrate feeds
supplemented for sick animals, milking cows and mating bulls were recorded. The feedstuffs
are rich in Ca, Mg, K, P and Fe content but poor in Na, Co and Cu. Based on the nutritional
evaluation of the dairy farm, it is performing below the potential. This is supported by the
calculations done based on energy and protein level. This implies that the feed resources
being utilised and the feed formulation process should be improved. Based on the given feed
stuffs, it could be difficult to foresee a target milk production level of more than 10 kg for
high yielding cows and 5kg for low yielding cows. Thus, in the future additional feed
ingredients should be included in the ration like Alfalfa, changing the proportion of the
ingredients; managing cows as per their production level and having a target more than the
current value of 3.2 kg for local breed (Fogera). The diet formulation indicates that in
combination of high quality feed ingredients it can be possible to increase the milk
production as well as reduce methane emission. This composition information can be used to
optimize the sustainable utilization of feed resources through balanced ration and increase
economic profitability of dairy farmers in the region. Further research is recommended to
validate and evaluate the new formulated diets through in vivo feeding trial using lactating
dairy cows.

147
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENTS

The first author would like to thank Prof. Dr. Ir. Veerle Fievez, who taught the corresponding
author about Dairy nutrition at Ghent University. We also thank Tottori and Shimane
Universities in Japan for laboratory analysis of the feed ingredients. The authors also have
special thanks for Prof. Ichinohe for giving the opportunity to undertake all the chemical
analysis of the feed samples in his animal nutrition laboratory, Shimane University Japan.
Moreover, we also thank ALRC Ruminant feed and nutrition researcher team, Ato Misganaw
Wale, Ato Eyasu lakew and other staff of the center for their facilitating and giving
appropriate information about the dairy farm.

REFERENCES

Alberro, M. & Haile-Mariam, S. 1982. The indigenous cattle of Ethiopia. Part I.

Alkhatib, A., Wamatu, J., Wegi, T. & Rischkowsky, B. A. 2016. Variation in the straw traits
of morphological fractions of faba bean (Vicia faba L.) and implications for selecting for
food-feed varieties. Animal Feed Science and Technology, 222, 122-131.
AOAC 1990. Official Methods of Analysis: Changes in Official Methods of Analysis Made
at the Annual Meeting. Supplement, Association of Official Analytical Chemists.
AOAC 1995. Official methods of analysis of AOAC International (16th edn): edited by
Patricia A. Cunniff, AOAC International, 1995. $359.00 (North America)/$399.00
(elsewhere)(xxvi 1899 pages) ISBN 0 935 584 54 4. Elsevier.
AOAC 2000. Official methods of analysis of AOAC international 17th edition. AOAC
international, Gaithersburg, USA, 43-44.
Ayalew, W., King, J., Bruns, E. & Rischkowsky, B. 2003. Economic evaluation of
smallholder subsistence livestock production: lessons from an Ethiopian goat
development program. Ecological Economics, 45, 473-485.
Beigh, Y. A., Ganai, A. M. & Ahmad, H. A. 2017. Prospects of complete feed system in
ruminant feeding: A review. Veterinary world, 10, 424.
Bitew, A., Taye, M., Kebede, A., Mekuriaw, G., Tassew, A., Mulugeta, T. & Goshu, G.
2010. Milk yield and calf growth performance of cattle under partial suckling system at
Andassa Livestock Research Centre, North West Ethiopia. Livestock Research for Rural
Development, 22, 2010.
Bogale, S., Melaku, S. & Yami, A. 2008. Potential use of crop residues as livestock feed
resources under smallholder farmers conditions in Bale highlands of Ethiopia. Tropical
and Subtropical Agroecosystems, 8.
Chakeredza, S., Akinnifesi, F. K., Ajayi, O. C., Sileshi, G., Mngomba, S. & Gondwe, F. M.
2008. A simple method of formulating least-cost diets for smallholder dairy production
in sub-Saharan Africa. African Journal of Biotechnology, 7, 2925-2933.

148
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Garg, M., Kannan, A., Shelke, S., Phondba, B. & Sherasia, P. 2012. Nutritional evaluation of
some ruminant feedstuffs by in vitro gas production technique. Indian Journal of Animal
Sciences, 82, 898.
Garg, M. R. & Makkar, H. P. 2012. Balanced feeding for improving livestock productivity.
Gemiyo, D., Hassen, A., Kocho, T., Birhanu, T., Bassa, Z. & Jimma, A. 2013. Chemical
composition and digestibility of major feed resources in the mixed farming system of
southern Ethiopia. World Applied Sciences Journal, 26, 267-275.
Getachew, G., Robinson, P., DePeters, E. & Taylor, S. 2004. Relationships between chemical
composition, dry matter degradation and in vitro gas production of several ruminant
feeds. Animal Feed Science and Technology, 111, 57-71.
Ja’afar-Furo, M. 2011. Nutrients evaluation of some common feed resources for cost
effective feeding of ruminants in Mubi, Nigeria. Agriculture and Biology.
Kamalak, A., Canbolat, O., Gurbuz, Y., Erol, A. & Ozay, O. 2005. Effect of maturity stage
on chemical composition, in vitro and in situ dry matter degradation of tumbleweed hay
(Gundelia tournefortii L.). Small Ruminant Research, 58, 149-156.
Kebede, A., Bitew, A., Bimrew, T., Yitayew, A., Denekew, Y., Ferede, Y. & Zeleke, G.
2013. Comparative Evaluation of the Fattening Performance of Fogera and Adet Old
Oxen at Andassa Livestock Research Center, Ethiopia. Value Ch.
Makkar, R., Lambadaris, I., Salim, J. H., Seddigh, N., Nandy, B. & Babiarz, J. Empirical
study of buffer management scheme for DiffServ Assured Forwarding PHB.
Proceedings Ninth International Conference on Computer Communications and
Networks (Cat. No. 00EX440), 2000. IEEE, 632-637.
Menke, K. & Steingass, H. 1988. Estimation of the energetic feed value obtained from
chemical analysis and. vitro.
Negesse, T., Makkar, H. & Becker, K. 2009. Nutritive value of some non-conventional feed
resources of Ethiopia determined by chemical analyses and an in vitro gas method.
Animal Feed Science and Technology, 154, 204-217.
Niu, M., Kebreab, E., Hristov, A. N., Oh, J., Arndt, C., Bannink, A., Bayat, A. R., Brito, A.
F., Boland, T. & Casper, D. 2018. Prediction of enteric methane production, yield, and
intensity in dairy cattle using an intercontinental database. Global change biology, 24,
3368-3389.
NRC 2001. Nutrient requirements of dairy cattle: 2001, National Academies Press.
Oishi, K., Kumagai, H. & Hirooka, H. 2011. Application of the modified feed formulation to
optimize economic and environmental criteria in beef cattle fattening systems with food
by-products. Animal feed science and technology, 165, 38-50.
Saxena, P. & Khanna, N. 2014. Animal feed formulation: mathematical programming
techniques. CAB Reviews: Perspectives in Agriculture, Veterinary Science Nutrition and
Natural Resources, 9, 1-12.
Sewalem, A. & Banjaw, K. 1992. Evaluation of the Reproductive and Pre-Weaning Growth
Performance of Fogera Cattle and Their F1 Friesian Crosses at Andassa Cattle Breeding
Station, Ethiopia. Haramaya University.

149
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Seyoum, B. & Zinash, S. Utilization of teff straw as livestock feed: research review.
Proceedings, 5th National Conference of the Ethiopian Society of Animal Production,
Addis Ababa, 1998.
Shapiro, B., Gebru, G., Desta, S., Negassa, A., Nigussie, K., Aboset, G. & Mechal, H. 2015.
Ethiopia livestock master plan. ILRI Project Report. Nairobi, Kenya: International
Livestock Research Institute (ILRI).
Shewangizaw, A. & Bishoftu, E. 2014. Assessment of feed formulation and feeding practices
for urban and peri urban dairy cows around Holetta, Ethiopia. MSc Thesis, Addis Ababa
University.
Talore, D. G. 2015. Evaluation of major feed resources in crop-livestock mixed farming
systems, southern Ethiopia: Indigenous knowledge versus laboratory analysis results.
Journal of Agriculture and Rural Development in the Tropics and Subtropics (JARTS),
116, 157-166.
Tesfa, A., Kumar, D., Abegaz, S. & Mekuriaw, G. 2017. Conservation and Improvement
Strategy for Fogera Cattle: A Lesson for Ethiopia Ingenious Cattle Breed Resource.
Advances in Agriculture, 2017.
Tesfa, A., Kumar, D., Abegaz, S., Mekuriaw, G., Bimerew, T., Kebede, A., Bitew, A.,
Ferede, Y., Mazengia, H. & Tilahun, M. 2016. Growth and reproductive performance of
Fogera cattle breed at Andassa Livestock Research Center. Development, 28, 1.
Tessema, Z. & Baars, R. 2004. Chemical composition, in vitro dry matter digestibility and
ruminal degradation of Napier grass (Pennisetum purpureum (L.) Schumach.) mixed
with different levels of Sesbania sesban (L.) Merr. Animal feed science and technology,
117, 29-41.
Van Soest, P. V., Robertson, J. & Lewis, B. 1991. Methods for dietary fiber, neutral detergent
fiber, and nonstarch polysaccharides in relation to animal nutrition. Journal of dairy
science, 74, 3583-3597.
Waghorn, G. & Clark, D. 2004. Feeding value of pastures for ruminants. New Zealand
Veterinary Journal, 52, 320-331.
White, R. R. & Capper, J. L. 2014. Precision diet formulation to improve performance and
profitability across various climates: Modeling the implications of increasing the
formulation frequency of dairy cattle diets. Journal of dairy science, 97, 1563-1577.
Yihalem, D. 2004. Effect of stage of harvesting on botanical composition of selected natural
pasture for optimum hay production at Andassa, North Western Ethiopia. M. Sc. thesis.
Alemaya University, Ethiopia.

150
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Yield and Yield Related Performances of Cowpea (Vigna Unguiculata L.) Varieties
Tested with and without Fertilizer under Irrigation in Central Gondar, Ethiopia

Alemu Tarekegn, Dessalegn Amsalu and Kifetew Adane

Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia

ABSTRACT

The study aimed to evaluate the yield and yield related performance of cowpea (Vigna
unguiculata L.) varieties at Western Dembia under irrigation in the year 2017 and 2018 for
two consecutive years. The experiment was laid down in a split plot design with randomized
complete block design arrangement in four replications. There were two main plots of which
one was planted with fertilizer and the other without fertilizer and the sub plot treatments
were eight Cowpea varieties (ILRI-9333, ILRI-9334, ILRI-11114, ILRI-12688, ILRI-12713,
Kenkety, Back eye bean and IT92KD258-9). The collected data were plant height (cm),
number of branches per plant, forage biomass yield (tha-1), dry matter yield (tha-1), leaf to
stem ratio, number of pods per plant, number of seeds per pod, grain yield (kgha-1), thousand
seed weight (g) and haulm yield (tha-1). The results revealed that the interaction between
fertilizer use and variety on plant height (cm), number of branches per plant, forage biomass
yield (tha-1), dry matter yield (tha-1) and haulm yield (tha-1) was significantly higher (P
<0.001). Highest dry biomass yield (tha-1) was recorded when ILRI-9334, ILRI-11114, ILRI-
12688 and IT92KD258-9 planted with fertilizer. The highest haulm yield was obtained from
ILRI-9334, ILRI-11114 and ILRI-12713 when planted with fertilizer and when ILRI-9334,
ILRI-12688, ILRI-12713 and IT92KD258-9 planted without fertilizer. Kenkety and
IT92KD258-9 produced higher grain yield when compared to other varieties. Back eye bean
recorded significantly (P <0.001) higher thousand seed weight than other genotypes. Cowpea
varieties ILRI-9333, ILRI-11114, ILRI-12688, ILRI-12713 and IT92KD258-9 when planted
with and without fertilizer; when variety ILRI- 9334 and Kenkety planted only with fertilizer
and also Back eye bean when planted without fertilizer was found economically profitable.
Therefore, planting Cowpea varieties ILRI-9334, ILRI-11114, ILRI-12688 and IT92KD258-
9 with 100 kgha-1 NPS fertilizer appears to be more preferable regarding biomass yield and
were found potentially profitable.

Keywords: Cowpea, dry matter yield, grain yield, variety

151
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

The farming system in Central Gondar zone predominantly mixed crop- livestock production.
However, there is repeated crop failure due to short rainy season, erratic rainfall, erosion, and
poor soil fertility, among other factors. There is also limited land availability for crop
production in the area. Thus livestock production plays a vital role in the livelihood of the
farmers not only during bad production years, but also during normal years. However,
livestock production in the area is constrained by feed shortage both in terms of quality and
quantity. A reason for feed shortage is that livestock production, particularly feed
development, is given a second priority to crop production. This is reflected in the basis
towards food crop production by the extension system and farmers’ reluctance to devote land
and labor for forage production (Yayneshet et al., 2009).

Because of the limited land available for food crop production, sole forage production may
not be feasible in the mixed crop- livestock production system of the area where food crop
production is given high priority. Therefore it is very important to seek multipurpose type
forage crop varieties which are adaptable and high yielding for the study area and similar
agro-ecologies. Cowpea is one of the candidate forage which is characterized by
multipurpose and short-lived annual legume that can best be inter-cropped with cereals.
Besides its suitability to inter cropping, crop residue (haulm) from cowpea is a very important
fodder resource which contains higher crude protein (21%) in the dry haulm (Singh and
Tarawali, 1997).

In addition to its food and forage value cowpea is very important in improving soil fertility
through the process of nitrogen fixation (Sanginga et al., 2003; Abayomi, 2008). Having all
the above mentioned advantages cowpea can be a potential feed and food crop in the midland
areas of Central Gondar. Considering the above advantages the development sector is trying
to introduce cowpea though the adoption is very low due to low productive varieties used. To
tackle this challenge, searching for adaptive and productive cow pea variety is becoming a
great deal. However, there were no recommended varieties in the potential midland areas of
North Gondar zone and the use of chemical fertilizer on this crop remained under deviation.
The current study was, therefore, conducted with the objective to estimate the yield potential
of different varieties of cowpea with and without fertilizer for future sustainable forage
production under irrigation in the study area.

152
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of Study Area

The study was conducted at West Dembia district, Melkam wuha village; 48 km south of the
Zonal capital, Gondar, which is located between latitude 1217’42.18’’N and longitude
3713’25.39’’E at an elevation of 1856.4 m.a.s.l. The soil texture of the study area is sandy
loam, good in water- holding capacity and fertility. The area has a moist tropical climate and
has the mean monthly temperature of 18.34C. Based on 10 years (2008 - 2017) data, the
total annual rainfall ranges between 665mm and 1524mm with a mean value of 1095mm.

Treatments and Experimental Design

The experiment was laid down in a split plot with RCBD arrangement in four replications.
The main plot factor was fertilizer (one was planted with fertilizer and the other one without
fertilizer) and the sub plot factor was variety (eight accessions of Cowpea (Vigna
unguiculata) ILRI-9333, ILRI-9334, ILRI-11114, ILRI-12688, ILRI-12713, Kenkety, Back
eye bean and IT92KD258-9) adaptive to the low land areas of Ethiopia. The varieties were
collected from International Livestock Research Institute (ILRI) forage seed unit. The plot
size was 2.4*2.8m with spacing of 40cm between rows and 20cm between plants. Spacing for
both between plots and replications was 1m. For the main plot which was planted with
fertilizer NPS at the rate of 100 kg ha-1 was applied at the time of planting.

Sampling, Data Collection and Processing

In this experiment, after planting in the field the performances of varieties were evaluated
with respect to plant height at forage harvest, number of branches per plant, forage biomass
yield (tha-1), dry matter percent (%), dry matter yield (tha-1), leaf to stem ratio, number of
pods per plant, number of seeds per pod, grain yield (kgha-1), thousand seed weight (g) and
haulm yield (tha-1). During sampling each plot was divided into two halves crosswise with an
effective plot size of 2.4*1.4m. One half was used for forage sampling and the other half for
seed sampling. Forage sampling was done at 50% blooming stage while grain yield of
Cowpea was determined at 12.5% moisture content. In each case, sampling was done from
the middle two rows excluding the guard rows. Immediately after forage harvested fresh yield
was measured by using Salter balance having a sensitivity of 0.1kg for green biomass yield
estimation.

153
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Individual samples of Cowpea forage were taken for DM% analysis, which was oven dried at
65oC for 72 hours until constant weight was obtained. Plant height was measured by
averaging the natural standing height of ten plants per plot. The main branch number was an
average of primary branches on the stem of ten plants per plot. To determine leaf to stem
ratio, five plants per plot were taken at random, dried in a paper bag and the leaves then were
carefully stripped from the stems. The ratio was obtained by dividing the percentage of leaf
by the percentage of stem. The number of pods per plant was calculated as the average of five
plants. Number of seeds per pod was calculated as the average of thirty pods. Before
measuring grain yield the seed samples were air dried until constant weight obtained.
Thousand seed weight was measured from a weight of thousand seeds. Haulm yield was
calculated after air dried until constant weight obtained.

Economic Analysis

Economic analysis was performed following the CIMMYT partial budget analysis
methodology (CIMMYT, 1988). Average fertilizer, Cowpea biomass and labour costs for
agronomic works were considered for the economic analysis. Following CIMMYT’s partial
budget analysis methodology, total variable costs (TVC), gross benefit and net benefit were
calculated. Total variable cost was calculated as the sum of cost of NPS fertilizer and labour
cost of harvesting and transporting of Cowpea. Net benefit was calculated as the difference
between gross benefit and the TVC. Cowpea biomass yield was adjusted downwards by 10%
assuming that farmers will obtain yields 10% lower than obtained by researchers. Then
treatments were listed in order of increasing total costs that vary and dominance analysis was
performed where dominated treatments were eliminated and the marginal rate of return
(MRR) calculated for the remaining treatments. A treatment that has net benefits that were
less than or equal to those of a treatment with lower costs that vary is dominated. A treatment
which was non-dominated and having a MRR of greater or equal to 100% and the highest net
benefit was considered as economically profitable.

Data Analysis

The data collected was subjected to analysis of variance (ANOVA) by using the general
linear model (GLM) procedure of Statistical Analysis System (SAS) (2003) version 9.1.
When the difference was significant among treatments, LSD (least significant difference) test
at 5% probability level was used to locate differences between the treatment means.

154
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Plant Height (cm)

There was a highly significant (P<0.001) interaction effect observed between Cowpea
varieties and fertilizer usage on plant height (Table 1). The tallest plant height was recorded
when Back eye bean Cowpea varieties planted with NPS fertilizer. On the contrary, the
Cowpea variety Kenkety planted without fertilizer showed the shortest plant height. This
could be attributed to the reaction of different cultivars for fertilizer usage. In contrast with
the current result relatively the shortest plant height was recorded from all the cultivars at
Western Belesa testing site that ranges from 54.75 to 82.62cm (Surafel et al., 2008). Tesfay et
al. (2018) also reported plant height of 62.75cm for other cowpea varieties at Southern
lowlands of Tigray which is relatively lower than the current result. Variations in the plant
height of Cowpea cultivars could be attributed to differences in the level of soil fertility and
PH, climatic zones of the different study areas and the amount and type of fertilizer utilized.

Number of Branches per Plant

There was highly significant (P<0.001) interaction among the treatment groups. The highest
number of branches was obtained when Cowpea Varity ILRI-9334 planted with NPS
fertilizer and Cowpea varieties ILRI-12688 and IT92KD258-9 planted without fertilizer. This
might be due to differences in response of different Cowpea varieties for fertilizer. Mean
number of branches per plant in this study was 9.17. Similarly Tesfaye et al. (2018) reported
a mean number of branches per plant of 8.52 for the same genotypes at Southern lowlands of
Tigray. In contrary to the current result, Tessema (2018) reported the number of branches per
plant that ranges from 10 - 17 at Salayish 1 in Lowlands of Southern Ethiopia. The
discrepancy of current result from previous findings could be due to the differences in
edaphic factors, environmental temperature and precipitation in different locations. In line
with the current result, Tesfay et al. (2018) achieved 8.52 plant heights (cm) for other cowpea
varieties at Southern lowlands of Tigray.

Dry Matter Percent

The two year combined analysis results of the experiment at West Dembia indicated that
there was no significant (P>0.05) interaction observed with the cowpea varieties and fertilizer
usage (Table 1). Mean dry matter percent was 16.49 for all the cowpea varieties.

155
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: The effect of fertilizer, Variety and their interaction on biomass yield and yield components of Cowpea in 2017 and 2018 irrigation
seasons at west Dembia

Cowpea Varieties
Fertilizer Usage ILRI-9333 ILRI 9334 ILRI 11114 ILRI 12688 ILRI 12713 Kenkety Back eye bean IT92KD258-9 Mean
Plant height (cm)
With fertilizer 61.43fg 81.45cd 73.83cdef 58.95fg 77.50cde 70.08defg 119.98a 65.55efg 76.10
Without fertilizer 58.85fg 60.22fg 56.07g 81.65cd 63.43efg 41.00h 100.83b 88.59bc 68.83
Mean 60.14 70.84 64.95 70.30 70.47 55.54 110.41 77.07 72.46
CV (%) 12.92
LSD (5%) 15.03
LS ***
Number of branches per plant
With fertilizer 9.28def 12.80a 10.30bcde 9.60def 6.70h 9.40def 10.80bcd 9.93cdef 9.85
Without fertilizer 8.73efg 8.50fg 7.11gh 11.65abc 2.43h 8.78efg 8.65efg 12.03ab 8.49
Mean 9.01 10.65 8.705 10.63 4.57 9.09 9.73 10.98 9.17
CV (%) 9.27
LSD (5%) 1.80
LS ***
Forage biomass yield (t ha-1)
With fertilizer 30.39abc 34.82ab 31.21abc 29.09bcd 33.36ab 25.05cde 23.22de 34.15ab 30.16
Without fertilizer 16.94ef 17.59ef 11.84f 37.01ab 20.12f 12.04f 12.49f 39.60a 20.95
Mean 23.67 26.21 21.53 33.05 26.74 18.55 17.86 36.88 25.56
CV (%) 17.69
LSD (5%) 7.02
LS ***
Dry matter percent (%)
With fertilizer 15.56 15.65 16.36 15.77 15.59 17.90 17.02 15.07 16.12
Without fertilizer 16.36 15.83 17.29 17.25 16.61 16.77 17.92 16.91 16.87
Mean 15.96 15.74 16.83 16.51 16.1 17.34 17.47 15.99 16.49
CV (%) 14.01
LSD (5%) Ns
LS Ns

156
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Cowpea Varieties
Fertilizer Usage ILRI-9333 ILRI 9334 ILRI 11114 ILRI 12688 ILRI 12713 Kenkety Back eye bean IT92KD258-9 Mean
Dry matter yield (tha-1)
With fertilizer 4.68cde 5.22abcd 5.42abc 6.32ab 5.07bcd 4.32de 3.80ef 6.61a 5.18
Without fertilizer 2.76fg 2.73fg 2.09g 4.56cde 3.22ef 1.98g 2.19g 5.02bcd 3.07
Mean 3.72 3.98 3.76 4.56 4.15 3.15 3.00 5.02 3.92
CV (%) 19.31
LSD (5%) 0.99
LS ***
Leaf to stem ratio
With fertilizer 1.02 0.99 0.94 0.89 1.03 1.03 1.06 1.01 0.99
Without fertilizer 0.98 0.99 0.97 0.87 1.05 1.00 1.02 1.03 0.99
Mean 1.00 0.99 0.96 0.88 1.04 1.02 1.04 1.02 0.99
CV (%) 11.23
LSD (5%) Ns
LS Ns
Haulm yield (tha-1)
With fertilizer 2.30bcde 3.17ab 3.00abc 2.18cdef 2.99abc 2.17ef 1.53ef 2.23bcdef 2.45
Without fertilizer 2.29bcde 2.57abcd 2.32bcde 3.37a 2.95abc 1.25f 1.78def 3.20ab 2.47
Mean 2.30 2.87 2.66 3.37 2.97 1.71 1.66 3.20 2.59
CV (%) 24.98
LSD (5%) 0.99
LS ***
*a, b, c, d, e, f, g = means within column having different superscript letters are significantly different at *** = P<0.001; ns = non-significant at
(P>0.05); CV = coefficient of variation; ILRI = international livestock research institute; LS = level of significant; LSD = least significant
difference.

157
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Green Biomass and Dry Matter Yield (t ha-1)

There was highly significant (P<0.001) interaction observed among Cowpea varieties and
fertilizer usage in green biomass and dry matter yield (tha-1) (Table 1). Cowpea varieties
ILRI-9334, ILRI-11114 and IT92KD258-9 when planted with NPS fertilizer produced
significantly (P<0.001) highest green biomass and dry matter yield (t ha-1). This is due to
positive response of the varieties for the NPS fertilizer in producing green biomass and dry
matter yield as well. Even though variety ILRI-12688 with NPS fertilizer produced
significantly lower green biomass due its higher dry matter percentage it also gave
significantly higher (P<0.001) dry matter yield (t ha-1). Compared to the current result
obtained, relatively lower dry matter yield (1.75 to 3.06 t/ha) was observed by different
cultivars at West Belessa (Surafel et al., 2008). The mean dry matter yield of Cowpea
varieties in this study was relatively lower than the reports of Tesfay et al. (2018) who
reported 6.54 t ha-1yr-1 for Cowpea varieties. This could be attributed to the differences in
varieties used, environmental temperature and precipitation in different locations.

Leaf to Stem Ratio

There was no significant (P>0.05) interaction observed with the Cowpea varieties and
fertilizer usage in leaf to stem ratio across years (Table 2). The mean leaf to stem ratio was
0.99 which agrees with the finding of Tesfaye et al. (2018) who reported leaf to stem ratio
that ranges from 0.75 to 1.10 for black eye bean, ILRI-9333, ILRI-9334 and Temesgen
cowpea varieties at Southern lowlands of Tigray.

Haulm Yield (t ha-1)

The result revealed that there was highly significant (p<0.001) interaction among Cowpea
varieties and fertilizer usages on cowpea haulm production (Table 1). The highest haulm
yield was recorded from cowpea variety ILRI-9334, ILRI-11114 and ILRI-12713 when
planted with fertilizer and when ILRI-9334, ILRI-12688, ILRI-12713 and IT92KD258-9
cowpea varieties planted without fertilizer. This indicates that different cowpea varieties react
differently for fertilizer usage in producing haulm yield (Tesfay et al., 2018).

158
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Mean grain yield and yield components of Cowpea varieties tested in 2017 and 2018
irrigation seasons at west Dembia
Number of pods Number of seeds Grain yield Thousand seed
Treatment per plant per pod (kgha-1) weight (g)
cde d b
ILRI- 9333 37.98 11.22 2261.10 156.64d
de cd d
ILRI- 9334 32.15 12.05 898.60 88.63g
ILRI- 11114 53.75b 13.14bc 2289.00b 136.36f
bc b b
ILRI- 12688 47.35 13.54 2157.90 177.99b
bcd cd c
ILRI- 12713 42.50 11.91 1458.10 159.61cd
Kenkety 23.33e 12.90bc 2335.10ab 144.71e
cde e b
Back eye bean 33.20 9.30 2168.10 211.75a
IT92KD258-9 69.60a 15.27a 2796.30a 165.98e
Mean 42.48 12.42 2045.53 155.21
CV (%) 24.68 11.29 22.71 4.41
LSD (5%) 14.87 1.41 468.83 6.91
LS *** *** *** ***
*a, b, c, d, e, f, g = means within column having different superscript letters are significantly
different at *** = P<0.001; CV = coefficient of variation; ILRI = international livestock
research institute; LS = level of significant; LSD = least significant difference.

Number of Pods per Plant and Number of Seeds per Pod

There was no significant (P>0.05) interaction effect observed between Cowpea varieties and
fertilizer usage in producing number of pods per plant and number of seeds per pod (Table 2).
The result revealed that there was a highly significant difference between cowpea varieties on
number of pods per plant and number of seeds per pod production. The highest number of
pods per plant and number of seeds per pod was obtained from Cowpea variety IT92KD258-
9. This could be attributed to the genetic differences by the different cultivars and variation in
the production environment and management practices applied on the cultivars. The mean
number of pods per plant and number of seeds per pod was relatively higher than the result
achieved by Tekle (2014) at Alduba, Southern Ethiopia. This might be due to the genetic
differences by the different cultivars and suitability of the production environment for the
cultivars.

Grain Yield (kg ha-1)

The combined analysis of variance results showed that there was a significant (P<0.001)
difference between Cowpea genotypes tested in grain yield (Table 2). Among the genotypes
tested Kenkety and IT92KD258-9 recorded significantly (P<0.001) higher grain yield than
the rest of varieties. This might be due to the genetic differences of the different cultivars.

159
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The mean grain yield of the current study was in contrary with the previous results of Surafel
et al. (2008) for all the cultivars tested at Western Belesa, which ranges from 2900 to 3900
Kg ha-1, respectively. This could be due to the agro ecological and edaphic differences and
Cowpea varieties used within the study areas.

Thousand Seed Weight (g)

The mean combined analysis result of thousand seed weight for the different Cowpea
varieties tested is given in Table 2. Back eye bean cowpea variety recorded significantly
(P<0.001) higher thousand seed weight compared to other cultivars. This was due to the seed
size and amount of endosperm contained related to the genetic difference among the
genotypes.

Partial Budget Analysis

The result of partial budget analysis was shown in Table 3. Financial profitability is the
ultimate measure to recommend a technology. Any technology that is agronomically feasible
and is beneficial for soil improvement would not be attractive to farmers unless it is
financially profitable. The partial budget analysis showed that variety ILRI-9333, ILRI-
11114, ILRI-12688, ILRI-12713 and IT92KD258-9 when planted with and without fertilizer;
when variety ILRI-9334 and Kenkety planted only with fertilizer and also variety back eye
bean when planted without fertilizer was found economically profitable as it gives a rate of
return above the 100% acceptable rate of return. In the current study, by the use of these
varieties either with or without fertilizer farmers in the study area will be able to gain from
7.76 to 9.00 ETB for each 1.00 ETB investment on inputs, which implies a very high increase
in farmers’ income with a simple improvement in soil fertility through nitrogen fixation.

In most cases the economic analysis result disagrees with the agronomic result. This financial
benefit is in addition to the benefit in terms of soil fertility improvement which could not be
directly quantified in terms of monetary value.

160
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: Economic analysis for the Cowpea biomass yield produced at west Dembia in 2017 and 2018 irrigation seasons
Cowpea dry matter Adjusted Cowpea Total gross field benefit Total cost that Net benefit MRR
Treatment yield (tha-1) biomass yield (tha-1) from Cowpea hay (ETB) vary (ETB ha-1) (ETB ha-1) Dominance (%)
Kenkety without fertilizer 1.98 1.78 2,670.00 304.84 2,365.16
ILRI- 11114 without fertilizer 2.09 1.88 2,820.00 321.97 2,498.03 775.66
Back eye bean without 2.19 1.97 2,955.00 337.38 2,617.62 776.05
ILRI- 9334 without fertilizer 2.73 2.46 3,690.00 421.30 3,268.70 D
ILRI- 9333 without fertilizer
fertilizer 2.76 2.48 3,720.00 424.30 3,295.70 900.00
ILRI- 12713 without fertilizer 3.22 2.90 4,350.00 496.65 3,853.35 770.77
ILRI- 12688 without fertilizer 4.56 4.10 6,150.00 702.16 5,447.84 775.87
IT92KD258-9 without 5.02 4.52 6,780.00 774.09 6,005.91 775.85
Back eye bean with fertilizer 3.80 3.42 5,130.00 2,025.71 3,104.29 D
Kenkety
fertilizer with fertilizer 4.32 3.89 5,835.00 2,106.20 3,728.80 775.89
ILRI- 9333 with fertilizer 4.68 4.21 6,315.00 2,161.00 4,154.00 775.91
ILRI- 12713 with fertilizer 5.07 4.56 6,840.00 2,220.94 4,619.06 775.88
ILRI- 9334 with fertilizer 5.22 4.70 7,050.00 2,244.92 4,805.08 775.73
ILRI- 11114 with fertilizer 5.42 4.88 7,320.00 2,275.74 5,044.26 776.05
ILRI- 12688 with fertilizer 6.32 5.69 8,535.00 2,414.46 6,120.54 775.87
IT92KD258-9 with fertilizer 6.61 5.95 8,925.00 2,458.99 6,466.01 775.81
*D = dominance; ETB = Ethiopian birr; MRR = marginal rate of return.

161
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

The green biomass and dry matter yield (t ha-1) of Cowpea varieties ILRI-9334, ILRI-11114 and
IT92KD258-9 when planted with 100 kgha-1 NPS fertilizer were highest compared to the other
varieties when planted with and without fertilizer. Among the genotypes tested, Kenkety and
IT92KD258-9 varieties produced relatively higher grain yield than the rest of genotypes
consistently when planted with and without fertilizer. With respect to haulm yield, from the
varieties used ILRI-9334, ILRI-11114 and ILRI-12713 when planted with fertilizer and when
ILRI-9334, ILRI-12688, ILRI-12713 and IT92KD258-9 Cowpea varieties planted without
fertilizer can produce better yield after grain harvest. From this study result it can be concluded
that Cowpea varieties ILRI-9334, ILRI-11114, ILRI-12688 and IT92KD258-9 with 100 kgha-1
NPS fertilizer should be used for better dry matter yield production. Moreover, it was concluded
that Cowpea varieties ILRI-9333, ILRI-11114, ILRI-12688, ILRI-12713 and IT92KD258-9
when planted with and without fertilizer and when variety ILRI-9334 and Kenkety planted only
with fertilizer was found economically profitable. Hence, according to the results of this study
use of Cowpea varieties ILRI-9334, ILRI-11114, ILRI-12688 and IT92KD258-9 were found
biologically efficient and potentially profitable for a better biomass production. Therefore, for
sustainable cowpea production these varieties with their production package should be
demonstrated in the study area and other areas having similar agro-ecologies.

REFERENCES

Abayomi, Y. A., Ajibade, T. V., Sammuel, O. F. and Saadudeen, B. F. 2008. Growth and yield
responses of cowpea (Vigna unguiculata (L.) Walp) genotypes to nitrogen fertilizer (NPK)
application in the Southern Guinea Savanna zone of Nigeria. Asian J. Plant Sci, 7(2), 170-
176.
CIMMYT (International Maize and Wheat Improvement Centre). 1988. From agronomic data to
farmer recommendation: an economics training manual completely revised edition.
CIMMYT, Mexico.

162
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sanginga, N., Dashiell, K. E., Diels, J., Vanlauwe, B., Lyasse, O., Carsky, R. J. and Singh, B. B.
2003. Sustainable resource management coupled to resilient germplasm to provide new
intensive cereal–grain–legume–livestock systems in the dry savanna. Agriculture,
Ecosystems and Environment, 100(2-3), 305-314.
SAS. 2003. SAS User’s Guide: Statistics. Version 9.1. Carry, North Carolina. USA. Statistical
Analysis System Inc.
Singh, B. B. and Tararwali, S. A. 1997. Cowpea and its improvement: Key to sustainable mixed
crop/livestock farming systems in West Africa. Crop residues in sustainable mixed
crop/livestock farming systems, 1997. Cab international, UK.
Surafel Melaku, Tesfaye Tsegaye, Yeshambel Tesfa and Solomon Abegaz. 2008. Adaptation of
cowpea varieties as livestock feed in North Gondar of Amhara region. Proceedings of the 3rd
Annual Regional Conference on Completed Livestock Research Activities. Bahir Dar,
Ethiopia. 1- 4 September, 2008.
Tekle Yoseph. 2014. Performance Evaluation of Cowpea (Vigna unguiculata L.) Varieties under
Moisture Conservation Practices for Yield and Yield Components at Alduba, Southern
Ethiopia. Journal of Natural Sciences Research. 4: 7.
Tesfaye Atsbha, Solomon wayu, Temesgen Tesfay, Adhanom Baraki and Neguse
Geberetsadkan. 2018. Evaluation of cowpea genotypes for yield and yield components in
the southern lowlands of Tigray, Northern Ethiopia. Inter J agri Biosci, 7(4): 186-191.
www.ijagbio.com.
Tessema Tesfaye. 2018. Evaluation of Forage Type Cowpea (Vigna unguiculata L.Walp.)
Accessions for Dry Matter Yield in Lowlands of Southern Ethiopia. Forage Res., 44 (2): pp.
74-80. http://forageresearch.
Yayneshet Tesfay., Eik, L. O. and Moe, S. R. 2009. The effects of enclosures in restoring
degraded semi-arid vegetation in communal grazing lands in northern Ethiopia. Journal of
Arid Environments, 73: 542-549.

163
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Evaluation of Cowpea (Vigna Unguiculata L.) Varieties for Yield and Yield Related
Attributes with and without Fertilizer under Rainfed Conditions at Central Gondar Zone,
Ethiopia

Alemu Tarekegn, Dessalegn Amsalu, Kifetew Adane and Eyaya Gashaw

Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia

ABSTRACT

The study aimed to evaluate the yield potential of Cowpea (Vigna unguiculata L.) varieties with
and without fertilizer at West Dembia and Gondar Zuria districts under rainfed conditions in the
year 2017 and 2018. The study design was split-plot with a Randomized Complete Block Design
arrangement in four replications. There were two main plots planted with fertilizer and without
fertilizer, and the subplot treatments were six cowpea varieties (ILRI-9333, ILRI-9334, ILRI-
12688, ILRI-12713, Kenkety and IT92KD258-9). To see there production performance plant
height (cm), number of branches per plant, forage biomass yield (t ha-1), dry matter yield (t ha-1),
leaf to stem ratio, number of pods per plant, number of seeds per pod, grain yield (kg ha-1),
thousand seed weight (g) and haulm yield (t ha-1) were collected and analyzed. The present study
results revealed that the interaction between fertilizer and variety was not significant (P>0.05) in
all parameters. Applying fertilizer did not improve any of the studied parameters. All the tested
varieties performed in the same manner in their biomass yield and its components. From the
varieties tested Kenkety, ILRI-9333, and IT92KD258 had produced much higher grain yield. A
significant thousand seed weight was recorded by ILRI-9333 and ILRI-12688. Therefore, all the
varieties are found promising for the study areas and can be planted without fertilizer to avoid
the extra cost. So to address several farmers at a time with quality seed and to produce better
biomass yield priority should be given for the variety Kenkety, ILRI-9333, and IT92KD258-9 in
the study areas. Further research needs to be conducted on their agronomic practices and
utilization as a protein supplement to animals and also testing of promising varieties in different
locations should be done.

Keywords: Cowpea, dry matter yield, grain yield, variety

164
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Livestock production is constrained by feed shortage both in quantity and quality; this remains
the leading constraint to a better animal performance in Ethiopia (Yayneshet et al., 2009). One of
the reasons for the prevailing feed shortage in the country is that livestock production
particularly feeds development is given a second priority to crop production. This is reflected in
the basis of food crop production by the extension system and farmer's reluctance to devote land
and required inputs for forage production.

Similar to other parts of the country, natural pasture and crop residues are the main feed sources
in the area. Feedstuffs of such composition are insufficient to provide year-round supply of
nutrients beyond maintenance (Hindrichsen et al., 2001). This suggests the need for searching
alternative feed sources and feeding strategies that would help to enhance the quality of these
low-quality feeds. Among the options that can improve the nutritional quality of poor roughages
are legume forage species. Legumes are the most important forage plants that substantially
improve the feed available for livestock (Akinlade et al., 2005). In this regard, the integration of
appropriate leguminous forage species into the existing feeding systems has continually been
advocated to be one of the options that have to be exploited.

Because of the limited land available for food crop production, sole forage production is not
feasible in the mixed crop-livestock production system of the area where food crop production is
given high priority. Therefore, it is very important to seek multipurpose type forage crops that
can adapt easily and provide required feed for livestock in the area. Cowpea is an important
multipurpose and short-lived annual legume that can best be intercropped with cereals. Besides
its suitability to intercropping, crop residue (haulm) from cowpea is a very important fodder
resource that contains higher crude protein (21%) in the dry haulm (Singh and Tarawali, 1997).
Both the grain and the haulm are valuable dietary proteins for the African human population and
their livestock (Fatokun, 2002).

165
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In addition to its food and forage value, cowpea is very important in improving soil fertility
through the process of nitrogen fixation. The crop cowpea is a major component of the tropical
cropping system because of its ability to improve marginal lands through nitrogen fixation and as
a cover crop (Sanginga et al., 2003; Abayomi, 2008). The crop can fix 240 kgha-1 of atmospheric
nitrogen and make available 60 - 70 kgha-1 nitrogen for succeeding crops grown in a rotation
with it (Amira and Oduwaye, 2007).

Having all the above- mentioned advantages, cowpea can be a potential feed and food crop in the
midland areas of Central Gondar zone and similar agroecological systems of the country.
Considering many advantages the crop has, the development sector is trying to introduce the
crop though the adoption is very low due to unknown and low productive variety used. To tackle
the challenge they face the demand for adaptive and productive cowpea variety is becoming a
great deal. However, there is no recommended variety in the potential midland areas of Central
Gondar and the use of chemical fertilizer on this crop has remained under deviation. The current
study was, therefore, conducted with the objective to estimate the yield potential of different
varieties of Cowpea in different fertilizer use for sustainable forage production under rainfed
conditions.

MATERIALS AND METHODS

Description of Study Areas

The study was conducted as a multi-site trial at Gondar Zuria and West Dembia districts during
the main cropping season of 2017 and 2018 under rainfed conditions. Geographically Gondar
Zuria district particularly Degola kebele is located between latitude 12°25’14.9’’N and longitude
37°36’18.5’’E at an elevation of 2104 meter above sea level. The area has a moist tropical
climate and the mean monthly maximum temperature ranges from 26.3 to 33°C with a mean
value of 29.5°C, while the mean monthly minimum temperature ranges from 12.6 to 18.1°C with
a mean of 15.6°C. Based on 10 years (2008-2017) data, the total annual rainfall of the area
ranges between 641 and 1678 mm with a mean value of 1052 mm and the temporal situation of
the rainfall of the district shows bimodal event. The main rainy season lies between June and
September.

166
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Mesekele kirestos Kebele of Western Dembia district is situated at 45 km south of the Zonal
capital, Gondar, which is located between latitude 12°17’42.18’’N and longitude 37°13’25.39’’E
at an elevation of 1856.4 m.a.s.l. The soil texture of the study area is sandy loam, good in water-
holding capacity and fertility. The area has a moist tropical climate and the monthly maximum
temperature ranges from 24.8C to 35.9C with a mean value of 30.5C, while the monthly
minimum temperature ranges from 13.1C to 19.5C with a mean value of 16.69C. Based on 10
years (2008 - 2017) data, the total annual rainfall ranges between 918mm and 2104mm with a
mean value of 1299.12mm.

Experimental Design and Treatments

The experiment was laid down in a split-plot design with RCBD arrangement in four
replications. There were two main plots of which one was planted with fertilizer and the other
one planted without fertilizer for comparison and the subplot treatments were six varieties of
Cowpea varieties (ILRI-9333, ILRI-9334, ILRI-12688, ILRI-12713, Kenkety and IT92KD258-9)
adaptive to the low land areas of Ethiopia which were collected from International Livestock
Research Institute (ILRI) forage seed unit. The plot size was 2.4*2.8m with spacing of 40cm
between rows and 20cm between plants. Spacing between plots and replications was 1m. For the
main plot which was planted with fertilizer NPS at the rate of 100 kg ha-1 was applied at the time
of planting.

Sampling, Data Collection and Processing

In this experiment, after planting in the field their performance was evaluated with respect to
plant height at forage harvest, number of branches per plant, forage biomass yield (t ha-1), dry
matter percent (%), dry matter yield (t ha-1), leaf to stem ratio, number of pods per plant, number
of seeds per pod, grain yield (kg ha-1), thousand seed weight (g) and haulm yield (t ha-1).

During sampling, each plot was divided into two halves crosswise with an effective plot size of
2.4*1.4m. One half was used for forage sampling and the other half for seed sampling. Forage
sampling was done at a 50% blooming stage and the grain yield of Cowpea was determined at
12.5% moisture content.

167
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In each case, sampling was done from the middle two rows excluding the guard rows.
Immediately after forage sampling, the fresh samples were weighed by using Salter balance
having a sensitivity of 0.1Kg for green biomass yield estimation. A minimum of 500g individual
samples of Cowpea forage was taken for DM% analysis, which was oven-dried at 65oC for 72
hrs until a constant weight was obtained. Plant height was measured by averaging the natural
standing height of ten plants per plot. For leaf to stem ratio determination when any plot was
harvested five plants were taken at random, dried in a paper bag and the leaves then were
carefully stripped from the stems. The ratio was obtained by dividing the percentage of a leaf by
the percentage of the stem. The main branch number was an average of primary branches on the
stem of ten plants per plot.

The number of pods per plant was calculated as the average of five plants. The number of seeds
per pod was calculated as the average of thirty pods. Before measuring grain yield the seed
samples were air dried until constant weight obtained. Thousand seed weight was measured from
a weight of thousand seeds. Haulm yield was calculated after air-drying until constant weight
was obtained.

Data Analysis

The data collected were subjected to analysis of variance (ANOVA) using the general linear
model (GLM) procedure of the Statistical Analysis System (SAS version 9.1, 2003). When the
difference was significant, LSD (least significant difference) test at 5% probability level was
used to locate differences among the treatment means.

RESULTS AND DISCUSSION

The two years combined analysis results of the experiment at West Dembia and Gondar Zuria
districts indicated that there was no significant (P>0.05) interaction observed with the Cowpea
varieties and NPS fertilizer usage. Due to its absence of interaction we are presenting the results
of all the parameters of the subplot (Cowpea varieties).

168
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Plant Growth Related Performances

The combined analysis of variance results depicted that plant height and number of branches per
plant of Cowpea varieties during the evaluation period in 2017/18 growing seasons is presented
in Table 1. The result showed non-significant difference (P>0.05) between treatment groups. In
contrary to the current result, Ali et al. (2009) and Ichi et al. (2013) on cowpea genotypes
observed a significant difference in the number of branches per plant that could be as a result of
the difference in cowpea varieties in their respective findings. In line with the current result,
Tesfay et al. (2018) achieved 62.75 and 8.52 plant height (cm) and a number of branches per
plant, respectively for other cowpea varieties at Southern lowlands of Tigray.

Table 1: Mean combined biomass yield and yield components of Cowpea varieties tested over
years (2017 and 2018) and locations (West Dembia and Gondar Zuria)
PH FBMY DMY Leaf to stem
Treatment (cm) NBPP t ha-1 DMP (%) t ha-1 ratio HY t ha-1
ILRI- 9333 66.59 10.88 28.63 16.51 4.67 1.08 2.41
ILRI- 9334 60.05 11.57 39.57 16.00 3.78 0.99 2.65
ILRI- 12688 65.78 10.73 29.28 16.29 4.74 1.03 2.62
ILRI- 12713 70.37 10.63 26.96 16.72 5.54 0.88 3.05
Kenkety 68.57 9.70 25.00 17.48 4.34 1.04 1.83
IT92KD258-9 70.33 9.84 27.54 18.27 5.12 1.11 2.47
Mean 66.95 10.56 29.50 16.88 4.53 1.02 2.50
CV (%) 14.92 20.48 27.04 19.54 14.24 13.24 20.60
LSD (5%) 10.58 1.98 15.08 2.96 2.58 0.65 1.43
LS NS NS NS NS NS NS NS
*NS = non-significant at (P>0.05); CV = coefficient of variation; DMP = dry matter percent;
DMY = dry matter yield; FBMY = forage biomass yield; HY = haulm yield; ILRI = international
livestock research institute; LS = level of significant; LSD = least significant difference; NBPP
= number of branches per plant; PH = plant height.

Dry Matter Percentage

The two year combined analysis results of the experiment at West Dembia and Gondar Zuria
districts indicated that there was no significant (P>0.05) difference observed between Cowpea
varieties in mean dry matter percentage (Table 1). The mean value of dry matter percentage of
the Cowpea varieties was 16.88% which is in agreement with previous reports of Alemu et al.
(2018) which are 16.49% for the same varieties at the same agro-ecology under irrigation.

169
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Fresh Biomass, Dry Matter and Haulm Yield (t ha-1)

The two years combined analysis results of the experiment revealed that there was no significant
(P>0.05) difference observed in fresh biomass, dry matter, and haulm yield among the treatment
groups (Table 1). The data for green biomass yield ranged from 25.00 - 39.57 t ha-1 with the
mean value of 29.50 t ha-1. The mean dry matter yield of Cowpea varieties in this study was
relatively lower than the reports of Tesfay et al. (2018) who reported 6.54t ha-1 yr-1 for Cowpea
varieties. Compared to the previous results, we obtained relatively lower dry matter yield (1.75
to 3.06 t ha-1) which was observed by different cultivars at Western Belessa (Surafel et al., 2008).
Variations in the yields could be attributed to differences in the level of soil fertility, climatic
zones, seasons, and agronomic practices adopted and differences in varieties used in different
study areas. The mean haulm yield of Cowpea varieties was 2.50 t ha-1.

Leaf to Stem Ratio

The statistical analysis result revealed that there was no significant (P>0.05) difference observed
among the genotypes in the leaf to stem ratio across the years. Leaf to stem ratio varied from
0.88 to 1.11 with the mean value of 1.02 over genotypes.

Number of Pods per Plant and Number of Seeds per Pod

There was no significant (P>0.05) varietal difference observed in producing the number of pods
per plant and number of seeds per pod (Table 2). The mean number of pods per plant was 15.05
which agree with the finding of Tekle (2014) who reported the number of pods that range from 8
- 18.5 at Alduba, Southern Ethiopia, while the number of seeds per pod was relatively higher
than the result achieved by the same author. This could be attributed to the genetic differences by
the different cultivars and suitability of the production environment for the cultivars.

170
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Mean combined grain yield and yield components of Cowpea varieties tested over years
(2017 and 2018) and locations (West Dembia and Gondar Zuria)
Treatment NPPP NSPP GY kg ha-1 TSW (g)
ILRI- 9333 16.56 10.42 1821.0ab 141.97ab
ILRI- 9334 15.10 10.34 1123.2c 87.66c
bc
ILRI- 12688 12.78 9.88 1413.6 148.73a
ILRI- 12713 14.25 10.42 1221.8c 137.38b
a
Kenkety 15.39 10.98 1945.1 134.04b
IT92KD258-9 16.23 10.45 1744.8ab 137.48b
Mean 15.05 10.41 1544.9 131.21
CV (%) 9.78 11.81 17.63 13.35
LSD (5%) 4.04 1.35 246.30 5.10
LS ns ns ** ***
*a, b, c = means within column having different superscript letters are significantly different at
*** = P<0.001; **= P<0.01; ns = non-significant at (P>0.05); CV = coefficient of variation;
GY = grain yield; ILRI = international livestock research institute; LS = level of significant;
LSD = least significant difference; NPPP = number of pods per plant; NSPP = number of seeds
per pod; TSW = thousand seed weight.

Grain Yield (kg ha-1)

The combined analysis of variance result showed that there was significant (P<0.01) difference
between Cowpea varieties tested in grain yield (Table 2). Among the varieties tested Kenkety,
ILRI-9333 and IT92KD258-9 recorded significantly (P<0.01) higher grain yield than the rest
genotypes. This may be due to the genetic differences of the different cultivars. The mean grain
yield of the current study was in contrary with the previous results of Surafel et al. (2008) for all
the cultivars tested at Western Belessa. This could be due to the agro ecological and edaphic
difference and Cowpea varieties used within the study areas.

Thousand Seed Weight (g)

The mean combined analysis result of thousand seed weight for the different Cowpea varieties
tested is given in Table 2. ILRI-9333 and ILRI-12688 Cowpea varieties recorded significantly
(P<0.001) higher thousand seed weight compared to other cultivars. This was due to the seed
size and amount of endosperm contained related to the genetic difference among the varieties.

171
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

Fertilizer usage did not improve biomass and grain yield with their components in the study area.
All the Cowpea varieties tested performed in the same manner in most parameters like biomass
yield and yield components. However, Kenkety, ILRI-9333 and IT92KD258 varieties produced
much higher grain yield than the rest of varieties. Significantly higher thousand seed weight was
recorded by ILRI-9333 and ILRI-12688 Cowpea varieties. Hence all the varieties can be used as
an alternative feed source to minimize the burden of feed shortage in the study areas and other
similar areas. Since fertilizer did not improve biomass and grain yield of Cowpea it is
recommended to use all the varieties without fertilizer to avoid extra cost. To address a number
of farmers with quality seed and to produce better biomass yield Cowpea varieties Kenkety,
ILRI-9333 and IT92KD258-9 should be given more priority. Further research should be done on
the improvement of these varieties including their agronomy and also the possibility of their
utilization as protein supplement to animals. To exploit its potential under a range of livestock
production performances further research could be done and these cultivars should be widely
demonstrated to the farmer in the study areas and other similar areas.

ACKNOWLEDGMENT

The authors would like to acknowledge the Agricultural Growth Program for funding the
research project.

REFERENCES

Abayomi, Y. A., Ajibade, T. V., Sammuel, O. F. and Saadudeen, B. F. 2008. Growth and yield
responses of cowpea (Vigna unguiculata (L.) Walp) genotypes to nitrogen fertilizer (NPK)
application in the Southern Guinea Savanna zone of Nigeria. Asian J. Plant Sci, 7(2), 170-
176.
Akinlade, J. A., Smith, J.W., Raji, A. M., Busari, A. A., Adekunle, I. O. and Adewumi, M. K.
2005. Effect of Two Cowpea (Vigna unguiculata) fodder cultivars as supplements on
voluntary intake, milk yield and manure production of Bunaji Cows. Journal of Agriculture
and Rural Development in the Tropics and Subtropics, 106(2): 105– 112.
Ali B, A. U., Izge, P. E., Odo, D. and Aminu. 2009. Varietal performance of dual purpose dry
season cowpea (Vigna unguiculata L. Walp) under varying plant spacing in the Fadama in
North eastern Nigeria. Am-Eurasia J. Sustain. Agric. 3(1): 13-18.

172
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Amira, J. O. and Oduwaye, O. A. 2007. Genetic variability for biological nitrogen fixation traits
in tropical Soybeans (Glycine max (L) Merr). Nature and Science, 5(2), 11-15.
Fatokun, C. A. 2002. Challenges and opportunities for enhancing sustainable cowpea production:
proceedings of the World Cowpea Conference III held at International Institute of Tropical
Agriculture, 4-8 September 2000.
Hindrichsen, I. K., Osuji, P. O., Odenyo, A. A., Madsen, J. and Hvelplund, T. 2001. Effects of
supplementation with four multipurpose trees and Lablab purpureus on rumen microbial
population, rumen fermentation, digesta kinetics and microbial protein supply of sheep fed
maize stover ad libitum. TSAP Proceedings Vol. 28. pp 98-119.
Ichi, J. O., Igbadun, H. E., Miko, S. and Samndi, A. M. 2013. Growth and yield response of
selected cowpea (Vigna unguiculata (L.) Walp) varieties to irrigation interval and sowing
date. The Pacific J. Sci. and Technol. 14(1): 453-463.
Sanginga, N., Dashiell, K. E., Diels, J., Vanlauwe, B., Lyasse, O., Carsky, R. J. and Singh, B. B.
2003. Sustainable resource management coupled to resilient germplasm to provide new
intensive cereal–grain–legume–livestock systems in the dry savanna. Agriculture,
Ecosystems and Environment, 100(2-3), 305-314.
SAS. 2003. SAS User’s Guide: Statistics. Version 9.1. Carry, North Carolina. USA. Statistical
Analysis System Inc.
Singh, B. B. and Tararwali, S. A. 997. Cowpea and its improvement: Key to sustainable mixed
crop/livestock farming systems in West Africa. Crop residues in sustainable mixed
crop/livestock farming systems, 1997. Cab international, UK.
Surafel Melaku, Tesfaye Tsegaye, Yeshambel Tesfa and Solomon Abegaz. 2008. Adaptation of
cowpea varieties as livestock feed in North Gondar of Amhara region. Proceedings of the
3rd Annual Regional Conference on Completed Livestock Research Activities. Bahir Dar,
Ethiopia. 1- 4 September, 2008.
Tekle Yoseph. 2014. Performance Evaluation of Cowpea (Vigna unguiculata L.) Varieties under
Moisture Conservation Practices for Yield and Yield Components at Alduba, Southern
Ethiopia. Journal of Natural Sciences Research. 4: 7.
Tesfay Atsbha, Solomon Wayu, Temesgen Tesfay, Adhanom Baraki and Nguse Gebretsadkan.
2018. Evaluation of cowpea genotypes for yield and yield components in the southern
lowlands of Tigray, Northern Ethiopia. Inter J Agri Biosci, 7(4): 186-191.
www.ijagbio.com.
Yayneshet Tesfay., Eik, L. O. and Moe, S. R. 2009. The effects of enclosures in restoring
degraded semi-arid vegetation in communal grazing lands in northern Ethiopia. Journal of
Arid Environments, 73: 542-549.

173
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Effect of Variety and Cutting Interval on Biomass Yield and Quality of Alfalfa (Medicago
Sativa L.) under Irrigation in Central Gondar Zone, Ethiopia

Alemu Tarekegn, Dessalegn Amsalu and Kifetew Adane

Gondar Agricultural Research Center, P.O. Box 1337, Gondar, Ethiopia

ABSTRACT

Understanding the relationship between dry matter yield production and forage quality
throughout the growing season will help to optimize the cutting intervals between harvests in
different Alfalfa (Medicago sativa L.) varieties. An experiment was conducted to determine the
effect of variety and different cutting treatments on forage yield and quality of Alfalfa under the
ecological condition of West Dembia district in the year 2017 and 2018 for 18 months. Two
improved Alfalfa varieties (ILRI-6984 and Var. Dz F-552) and four harvest schedules (30, 40, 50
and 60 days) were used for the experiment. The experiment was laid down in a randomized
complete block design (RCBD) in factorial arrangement with four replications. To assess their
production potential plant height (cm), number of branches per plant, fresh biomass yield (t ha-1)
and dry matter yield (t ha-1) of the forge and its nutrient content were recorded. The data
collected was subjected to analysis of variance (ANOVA) by using the general linear model
procedure of Statistical Analysis System (SAS, 2003). The results revealed that the highest forge
dry matter yield (t ha-1) was obtained when Alfalfa forage variety ILRI-6984 harvested at 40
days interval and Alfalfa variety Var.DzF-552 when harvested at 60 days interval. Crude protein
production potential was significantly higher when Alfalfa variety ILRI-6984 harvested at 40
days interval and when variety Var.DzF-552 harvested at 30 and 50 days interval. Though crude
protein production of Alfalfa variety Var.DzF-552 when harvested at 30 and 50 days interval
was much higher, its dry matter yield was relatively lower. Therefore, use of Alfalfa variety
ILRI-6984 with 40 days interval cutting appears to be optimum when considered a better dry
matter yield and Crude protein production in Alfalfa.

Keywords: Alfalfa, cutting interval, dry matter yield, forage quality, variety

174
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Feed shortage both in terms of quantity and quality is a crucial constraint for livestock production in
the country particularly in the zone (Alemu et al., 2018). This is because of little attention given to
the livestock sector together with farmers’ priority for the production of food crops; their reluctance
to devote land and labor for forage production. Because of the severe feed shortage problem of the
area, farmers are efficient in utilizing crop residue to feed their livestock and are completely
dependent on the crop residue for the long dry season which is poor in protein and vitamin content
and digestibility.

Most ruminant livestock in the zone rely on the local grasses for their roughage and much of their
nutrition. Many of these species have low palatability, poor productivity and inadequate nutrients to
maintain animals, especially during the dry season. Improved forage legumes, have better
productivity, palatability and nutrient characteristics that make them desirable for inclusion in
improved forage production programs (Alemayehu, 1997).

Alfalfa is grown over a wide range of soil and climatic conditions and plays an important role in crop
rotation, throughout its positive effects on soil fertility, soil structure enhancement and reduces soil
erosion. It is a versatile crop which can be used as pasture, hay, green silage and cash crop. A dry hot
climate is most suitable for Alfalfa but the crop does not thrive in hot humid conditions. One of the
problems of the crop in most Alfalfa growing areas was the increasing cost of seed due to the
progressive increase in the cultivated area in response to the increased demand for dairy products.
However, Nayle and Khidir (1995) found that the seed rate of 40 kg ha -1 had given more fodder yield
than either 30 or 50 kg ha-1. Marble (1984) stated that Alfalfa is the chief irrigated fodder crop in
many parts of the world. Khair (1999) reported that the most important factors that should be taken
into consideration when cultivating Lucerne are adequate irrigation water to enable seedling growth
and right time of sowing.

Alfalfa is known in its higher CP and vitamin profile and considered as a very important forage to the
dairy, beef and poultry industry (USDA Census of Agriculture, 2012). Alfalfa can be harvested five
to seven times a year under irrigation. Besides to use high yielder variety proper harvest management
such as cutting interval between harvests is essential to profitable Alfalfa production, particularly in
the manipulation of forage quality and yield (Brink and Marten, 1989).

175
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Generally in Ethiopia alfalfa research, production and utilization is at infant stage. Although Alfalfa
is a very important forage crop, it is little known in the study area and no research was conducted in
evaluating different Alfalfa varieties and optimum cutting interval between harvests to maintain
forage yield and quality. Therefore, this work was done to determine the effect of variety and
different cutting treatments on forage yield and quality of Alfalfa.

MATERIALS AND METHODS

Description of the Study Area

The study was conducted at West Dembia district, Melkam wuha village; 48 km south of the Zonal
capital, Gondar, which is located between 1217’42.18’’N and longitude 3713’25.39’’E at an
elevation of 1856.4 m.a.s.l. The soil texture of the study area is sandy loam, good in water-holding
capacity and fertility. The area has a moist tropical climate and the monthly maximum temperature
ranges from 21.4 to 31C with a mean value of 26.2C, while the monthly minimum temperature
ranges from 8 to 17.3C with a mean value of 12.6C. Based on 10 years (2008 - 2017) data, the total
annual rainfall ranges between 665 mm and 1524 mm with a mean value of 1095mm.

Treatments and Experimental Design

Two improved Alfalfa (Medicago sativa L.) varieties (ILRI-6984 and Var. DzF-552) were used for
the experiment. Four harvest schedules (30, 40, 50 and 60 days) were applied to the two cultivars and
tested in factorial arrangement. The experimental plot was laid down in randomized complete block
design with four replications. Plot size was 2.4*2.8m with spacing of 40cm between rows and 20cm
between plants. Spacing between for both plots and replications was 1m. The seed rate was 5 kg ha -1
and NPS fertilizer at the rate of 100Kg/ha was applied at the time of planting. Sowing was done by
drilling along the row and irrigation was applied immediately after sowing. Second irrigation water
was applied after 7 days from the first irrigation to facilitate seedling emergence. Following seedling
emergence irrigation water was given within a ten days interval. Tinning was done after
establishment considering 20cm plant spacing. Weeds were effectively controlled by hand
throughout the growing season. Cutting was done following its cutting interval for each treatment. A
sickle was used for clipping the plants 5 - 7cm above the soil surface to give chance for further
regeneration. At each cut individual samples of Alfalfa forage were taken and dried in an air draft
oven at 65oC for 72 hrs to determine dry matter content and calculate dry matter yield (t ha-1).

176
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Collection

In this experiment, after planting in the field their performance was evaluated with respect to
plant height, number of branches per plant, fresh biomass yield (t ha-1), dry matter percent
(DM%) and Dry matter yield (t ha-1). Plant height was measured by averaging the natural
standing height of ten plants per plot. Alfalfa forage harvested for herbage and dry matter yield
were following each cutting interval. The main branch number was an average of primary
branches on the stem of ten plants per plot.

Forage Quality Analysis

Chemical analyses of feeds were carried out by taking representative samples from different
harvests. The feed samples were dried in a forced air draft oven at 60oC for 72 hrs for partial dry
matter (DM) determination. Dried samples of feed were milled using a laboratory mill to pass
through a 1 mm screen. Milled samples of feed were taken to Debere Birhan Agricultural
Research Centre and stored at room temperature pending chemical analysis. Dry matter (DM),
crude protein (CP), organic matter (OM), and ash in the feed samples were determined according
to the procedure of (AOAC, 2000). Nitrogen (N) was measured using the micro-Kjeldahl
procedure (AOAC, 2000). Crude protein was determined by multiplying nitrogen content with
6.25 factor while Organic matter (OM) was determined as 100 minus ash. Neutral detergent fiber
(NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) were analyzed according to
the procedures of (Van Soest et al., 1991).

Data Analysis

The data collected was subjected to analysis of variance (ANOVA) by using the general linear
model (GLM) procedure in Statistical Analysis System (SAS version 9.1, 2003). When the
difference was significant, LSD (least significant difference) test at 5% probability level was
used to locate differences between the treatment means.

177
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Biomass Yield and Related Components

The statistical analysis result revealed that there was a highly significant (p<0.001) interaction
effect between variety and cutting interval on green biomass yield (t ha-1) and dry biomass yield
(t ha-1). Maximum green biomass yield (t ha-1) was recorded when Alfalfa variety ILRI-6984
harvested within 40 days interval, while the lowest green biomass yield was recorded when
Alfalfa variety Var.DzF-552 harvested at 30 days interval. Significantly (p<0.001) higher dry
biomass yield (t ha-1) was obtained when Alfalfa variety ILRI-6984 harvested within 40 days
interval and Alfalfa variety Var.DzF-552 harvested within 60 days interval. This suggests that
alfalfa yield increased by approximately 60% by delaying one week for harvesting Alfalfa from
30 to 40 days interval, based on dry matter yields in the case of Alfalfa variety ILRI-6984.

This also implies that delaying Alfalfa harvest too long cutting interval might decrease alfalfa
yield possibly resulting from some dry matter loss of older shoots and leaves in the lower side of
Alfalfa. These data were similar to those reported by Kallenbach et al. (2002) in Missouri, who
found that harvesting Alfalfa five times per year yielded more per year than Alfalfa harvested
four times per year over a 4 year period. Sheaffer and Marten (1990) reported that the optimum
harvest interval for Alfalfa was about the 35th day in Minnesota. This result was in accordance
with the findings of Doohong Min (2016) who reported Alfalfa yield increased as cutting
intervals increased up to every 42 days. However there was no statistically significant (p<0.05)
interaction effect between variety and cutting interval on plant height (cm), number of branches
per plant and dry matter percentage (Table 1).

178
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: The interaction effect of variety and cutting interval on biomass yield and yield
components of Alfalfa in 2017 and 2018 growing seasons

Number of
Plant height branches per Green Biomass Dry matter Dry biomass
Treatment (cm) plant Yield (t ha-1) percentage yield (t ha-1)
bc
ILRI- 6984*30 DCI 55.19 41.54 72.59 23.12 16.73c
a
ILRI- 6984*40 DCI 56.24 43.94 106.80 25.09 26.76a
ILRI- 6984*50 DCI 58.16 48.63 75.32bc 25.82 19.42bc
bc
ILRI- 6984*60 DCI 53.66 52.11 74.30 27.27 20.20b
Var.DzF-552*30 DCI 42.76 34.27 69.84c 25.82 17.65bc
c
Var.DzF-552*40 DCI 45.22 34.12 71.44 26.55 18.97bc
Var.DzF-552*50 DCI 53.28 38.12 71.79c 27.67 19.83b
b
Var.DzF-552*60 DCI 56.35 40.02 83.19 29.01 24.14a
Mean 52.61 41.59 78.16 26.29 20.46
CV (%) 12.92 5.89 9.64 3.86 9.32
LSD (0.05) NS NS 11.08 NS 2.81
LS NS NS *** NS ***
a, b, c
= means within columns having different superscript letters are significantly different at ***
= P<0.001; NS = non significant at (P>0.05); CV = coefficient of variation; DCI = days cutting
interval; LS = level of significant; LSD = least significant difference.

Forage Quality

The mean response for nutritive values to variety and cutting interval is shown in (Table 2). The
interaction of Variety and cutting interval effect was not significant (P>0.05) for dry matter
(DM), neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL)
which is in agreement with the findings of Ji-Shan et al. (2012) and in disparity with that of
others (Katić et al., 2008; Diriba et al., 2014). Though not statically significant, all cultivars
when harvested at 30, 40, 50 and 60 days interval had greater than 91, 51, 31 and 6% dry matter
(DM), neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin
(ADL), respectively which agrees with the findings of (Mekuanint et al., 2015). Ash content was
significantly (p<0.001) higher when Alfalfa variety ILRI-6984 and variety Var.DzF-552
harvested at 30 and 40 days intervals. This might be due to much biomass produced and less
organic matter accumulated to this stage for the two varieties. This result was in accordance with
the findings of Mekuanint et al. (2015) who reported the ash content of Alfalfa variety FG10-09
(F) when harvested at 45 days interval may reach at 10.47%.

179
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: The interaction effect of variety and cutting interval on chemical composition of Alfalfa
in 2017 and 2018 growing seasons
Chemical Composition of Feeds (%)
Treatment DM Ash CP NDF ADF ADL
a bc
ILRI- 6984*30 DCI 91.00 12.01 23.32 57.76 39.58 9.82
ILRI- 6984*40 DCI 91.67 12.00a 24.18ab 53.09 33.47 6.89
de bc
ILRI- 6984*50 DCI 91.67 8.73 20.48 67.06 44.31 10.99
ILRI- 6984*60 DCI 91.33 9.47cd 21.38bc 63.32 41.49 9.58
ab a
Var.DzF-552*30 DCI 91.33 10.95 29.10 51.39 31.37 6.56
Var.DzF-552*40 DCI 91.67 10.91ab 18.87bc 65.06 43.65 10.64
bc a
Var.DzF-552*50 DCI 91.33 10.59 24.22 61.03 37.25 9.03
Var.DzF-552*60 DCI 91.33 8.00e 18.50c 72.08 49.47 13.25
Mean 91.46 1o.34 22.37 61.35 40.07 9.60
CV (%) 1.06 6.98 4.37 13.47 9.30 19.59
LSD (0.05) NS 0.79 2.64 NS NS NS
LS NS *** * NS NS NS
a,b,c,d,e
= means within columns having different superscript letters are significantly different at *** =
P<0.0001; * = P<0.01; NS = non significant at (P>0.05); ADF = acid detergent fiber; ADL = acid
detergent lignin; CP = crude protein; CV = coefficient of variation; DCI = days cutting interval; DM =
dry matter; LS = level of significant; LSD = least significant difference; NDF = neutral detergent fiber.

The interaction of Variety and cutting interval effect was significantly (p<0.001) higher in crude
protein production when Alfalfa variety ILRI-6984 harvested at 40 days interval and when variety
Var.DzF-552 harvested at 30 and 50 days interval. This might be due to the lower maturity stage for
the most frequent harvest than the rest of cutting treatment but different varieties react differently for
different cutting treatments in crude protein production. Overall, forage quality increased as cutting
frequency optimized, which was consistent with other findings (Kallenbach et al., 2002; Moyer et al.,
1999).

CONCLUSION AND RECOMMENDATION

From this study result it can be concluded that variety and harvesting frequency are critically
important in dry matter yield and nutritive value of Alfalfa in west Dembia district. Based on the 18
months experimental period result dry matter yield of Alfalfa was higher when Alfalfa variety ILRI-
6984 harvested within 40 days interval and Alfalfa variety Var.DzF-552 harvested within 60 days
interval. Moreover, in this study, the effect of Variety and cutting interval had no significant effect on
plant height (cm), number of branches per plant and Dry matter percentage of Alfalfa. When Alfalfa
varieties ILRI-6984 harvested at 40 days interval and variety Var.DzF-552 when harvested at 30 and
50 days interval recorded higher crude protein than other cuttings.

180
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Therefore, based on the study results, use of variety ILRI-6984 with 40 days interval cutting
appears to be optimum when a better dry matter yield and Crude protein production are
considered. Hence Alfalfa variety ILRI-6984 with 40 days cutting interval can be used to
produce better dry matter yield with the better nutrient content especially crude protein. This
practice should be widely demonstrated to the poultry and dairy producer farmers in the study
area and other similar areas. Further research is needed to assess the reaction of other Alfalfa
varieties to different cutting intervals in respect to produce better yield in a better nutritional
quality. Based on the perennial nature of the plant for sustainable Alfalfa based feeding system
use of irrigation water is advisable.

REFERENCES

Alemayehu Mengistu. 1997. Conservation- Based Forage Development for Ethiopia. Self Help
Development International and Institute for Sustainable Development. Addis Ababa,
Ethiopia.
Alemu Tarekegn, Yeshambel Mekuriaw and Firew Tegegne. 2018. Growth and Carcass
Characteristics of Gumuz Sheep Supplemented with Different Proportions of Noug Seed
(Guizotia abyssinica) Cake and Wheat Bran. Proceeding of the 10th Annual Regional
Conference on Completed Livestock Research Activities March 13 - 16, 2017, Bahir Dar,
Ethiopia. http://www.arari.gov.et.
AOAC (Association of Official Analytical Chemists). 2000. Official method of analysis. 15th
edition, AOAC, Washington DC, 1298 pp.
Brink, G. and Marten, G. 1989. Harvest Management of Alfalfa-Nutrient Yield vs Forage
Quality and Relationship to Persistence. Journal of Production Agriculture, 2, 32-36.
http://dx.doi.org/10.2134/jpa1989.0032.
Ji-shan, C., Fen-lan, T., Rui-fen, Z., Chao, G., Gui-li, D. and Yue-xue, Z. 2012. Effects of cutting
frequency on alfalfa yield and yield components in Songnen Plain, Northeast China. African
J. of Bio techno., 11(21), 4782-4790.
Kallenbach, R., Nelson, C. and Coutts, J. 2002. Yield, Quality, and Persistence of Grazing- and
Hay-Type Alfalfa under Three Harvest Frequencies. Agronomy Journal, 94, 1094-1103.
http://dx.doi.org/10.2134/agronj2002.1094.

181
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Katić, S., Mihailović, V., Milić, D., Karagić, Đ., Glamočić, D. and Jajić, I. 2008. Genetic and
seasonal variations of fibre content in lucerne. Proceedings of the XXVIIth EUCARPIA
Symposium on Improvement of Fodder Crops and Amenity Grasses, Copenhagen.
Khair, M. A. 1999. Principles of forage crops production. Training and publication
administration, ARC, Wad Medani, Sudan (in Arabic).
Marble, V. L. 1984. Report of consultancy on alfalfa Seed production Morocco, Sudia Arabia
Yemen and Sudan AGPC: MICS PP.82-85 Rome, FAO.
Mekuanint Gashaw, Ashenafi Mengistu and Diriba Geleti. 2015. Biomass yield dynamics and
nutritional quality of alfalfa (Medicago sativa) cultivars at Debre Zeit, Ethiopia. E3 Journal
of Agricultural Research and Development Vol. 5(2). pp. 0120-0127, June, 2015.
http://www.e3journals.org.
Min, D. 2016. Effects of Cutting Interval between Harvests on Dry Matter Yield and Nutritive
Value in Alfalfa. American Journal of Plant Sciences, 7, 1226-1231.
http://dx.doi.org/10.4236/ajps.2016.78118.
Moyer, J. R., Fraser, J., Rode, L. M. and Topinka, A. K. 1999. Effects of Growth Stage-Based
Alfalfa Harvest on Weed Encroachment and Resultant Quality. Canadian Journal of Plant
Science, 79, 243-247. http://dx.doi.org/10.4141/P98-059.
Nayel, B. A. and Khidir, M. O. 1995. Note in the effect of cutting management on seed
production in Lucerne (Medicago sativa L.). Univ. Khart. J. Agric. Sci., 3:163-166.
SAS. 2003. SAS User’s Guide: Statistics. Version 9. 1. Carry, North Carolina. USA. Statistical
Analysis System Inc.
Sheaffer, C. and Marten, G. 1990. Alfalfa Cutting Frequency and Date of Fall Cutting. Journal of
Production Agriculture, 3, 486-491. http://dx.doi.org/10.2134/jpa1990.0486.
USDA Census of Agriculture. 2012. http://www.agcensus.usda.gov/Publications/2012.
Van Soest, P. J., Robertson, J. B. and Lewis, B. A. 1991. Methods for dietary fiber, neutral
detergent fiber, and non-starch polysaccharides in relation to animal nutrition. J. Dairy Sci.
74 (10): 3583-3597.

182
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Evaluation of Biomass Yield and Nutritional Composition of Soybean (Glycine Max (L.)
Merrill) Varieties Grown in Lowland Areas of Eastern Amhara, Ethiopia

Abito Asres and Solomon Tiruneh

Sirinka Agricultural Research Center, P. O. Box 74, Woldia, Ethiopia

ABSTRACT

The experiment was conducted in three locations Jari, Chefa and Sirinka of Eastern Amhara to
select the best performing varieties, in terms of biomass yield, chemical composition, haulm
yield, seed yield and other agronomic characteristics of Glycine max (L.) Merrill grown under
the rain- fed condition of lowland areas of Eastern Amhara. Randomized complete block design
with three replications was used to set the experiment. Twelve released soybean varieties, Afgat,
Belesa-95, Boshe, Cheri, Dhidhessa, Gishama, Gizo, Korme, Pawe-03, katta, Wegayen and
Wollo were used as a treatment. The seeds were planted in 40 cm between rows and 10 cm
between plants on a plot size of 3.2m*4m. Spacing between blocks and plots were 1 and 0.5m,
respectively. The seed rate was 60 kg/ha and a fertilizer rate 100 kg/ha NPS was applied during
seed planting. The combined analysis of variance over two years at location Jari for dry matter
yield of varieties Afgat, Gizo, Pawe-03, Wogayen and Wollo were significantly higher as
compared to other soybean varieties. The combined analysis across locations at Jari and Sirinka
(2019 - 2nd year) showed that varieties Afgat, Gishama, Gizo, Pawe-03, Wogayen and Wollo had
higher dry matter yield and varieties Gishama (3.97 t/ha), Gizo (3.60 t/ha), Pawe-03 (4.04 t/ha)
and Wogayen (3.36 t/ha) had higher haulm yield as compared with other varieties. The combined
analysis across locations at Jari and Sirinka (2019) showed that varieties Pawe-03 (2951 kg/ha),
Gizo (2862 kg/ha), Afgat (2859 kg/ha), Gishama (2654 kg/ha), Wollo (2461 kg/ha) and
Wogayen (2404 kg/ha) had higher seed yield as compared with other varieties. The variety
Wollo gave higher crude protein content in two locations (Jari and Sirinka). Therefore, varieties
Afgat, Pawe-03 and Wollo were recommended for the given areas of Jari, Chefa, and Sirinka and
could be produced in similar environments for the best of produced optimal amount of dry
matter, haulm and seed yield and good protein supplement for production of ruminants. Thus,
further research will be needed to investigate the utilization of livestock.

Keywords: Crude protein, Dry matter yield, Soybean

183
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

The low productivity of Ethiopian livestock is a result of several limiting factors among which
feed is the major one. The existing feed does not meet the amount required by livestock. Natural
pastures are the most important livestock feed resource. In most parts of the country natural
grasslands are confined to degraded shallow upland/highland, fallow crop lands and soils that
cannot be successfully cropped due to physical constraints such as flooding and water logging.
They are low yielding and their production is insufficient and grazing conditions are only
favorable for four to five months per year. For several decades, grazing areas have been
shrinking and likely to continue to do so because of rapid expansion of cultivated land for crop
production. As a result, there is always likely to be limited feed resources for the existing
livestock population. Therefore, selection of high-yielding and better-quality forage varieties,
and development of improved forage production systems are critically important (Alemayehu
and Getnet, 2012; Alemayehu et al., 2016).

Soybean (Glycine max (L.) Merrill) is a major legume crop grown for its protein and oil rich
seeds but it also makes valuable forage for grazing, silage and hay. It is a fast growing
herbaceous annual native to Asia that is currently grown worldwide. It is an erect leguminous
plant, which grows up to a height of 1.3 m. Like other forage legumes, soybean forage has many
valuable traits as fodder. Soybean leaves and stems can be grazed, ensiled or dried to make hay.
The foliage is very palatable to cattle and has a high nutritive value and good digestibility
(Koivisto, 2006).

Ideally soybean forage is suited for winter cropping in Ethiopia; it is tolerant of drought and
thrives when other forage legumes like alfalfa are not available. Forage soybean can be sown
alone or in combination with other forage species such as sorghum (Fujita et al., 1990). It grows
quickly and can provide 5 to 10 DM t/ha within 3 - 4 months (Doo-Hong Min, 2012; Koivisto et
al., 2003). It can be first grazed when it reaches 60 cm in height (Lugin buhl, 2006). In three to
four months after seeding it produces cut hay equal in quality to alfalfa cut hay (Hintz et al.,
1992).

184
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The nutritive value of soybean can be comparable to early bloom alfalfa (that shows high protein
content and is very digestible to mature ruminant animals). Soybean forage is relatively rich in
protein, which varies between 11% and up to more than 22% of the DM. Fibre content is also
quite variable, with ADF content ranging from 20% to more than 45% of the DM. It is relatively
poor in lignin (about 6% DM). The main factor influencing soybean forage quality is the
maturity at harvest. Protein concentration decreases during flowering and increases during pod
formation, while fibre concentration changes inversely. The stem and leaf proportion of the plant
decreases as the pod and seed components increase. The lipid content of the mature forage can
reach 10% of the DM due to the presence of oil in the seeds (Hintz et al., 1992).

Haulm is plant material above the ground level, harvested, dried and used for feeding livestock.
Grain legume haulms are also playing a significant role in supplying fodders for ruminant
feeding in small scale mixed farming systems. The highlands of Ethiopia demonstrated
increasing trends in the use of grain legume haulms as livestock feed by smallholder farmers
(Alkhlib et al., 2014). Soybean haulms (residue) have already become important components of
ruminant diet in small scale mixed crop-livestock farming areas. Soybean haulms had also
figured out value of 89.18, 5.10, 2.85, 96.90, 80.80, 63.20 and 13.00% for DM, CP, EE, OM,
NDF, ADF and ADL contents, respectively (Maheri-Sis et al., 2011). The dry matter yield of
soybean haulm was 5.31 t/ha harvested with application of NPK followed by urea fertilizer, 4.31
t/ha with organic fertilizer (compost) and 4.29 t/ha without any fertilizer (Yagoub et al., 2012)
and 3.23 t/ha with fertilizer in (Sisay, 2017).

This study, therefore, was carried out to select the best performing varieties, in terms of biomass
yield, chemical composition, haulm yield, seed yield and other agronomic characteristics of
Glycine max (L.) Merrill grown under the rain- fed condition of low land areas of Eastern
Amhara.

185
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of the Study Area

The study was conducted in three locations, Jari, Chefa and Sirinka of Eastern Amhara. Jari is
situated at 11°21’00’’N latitude and 39°38’00’’E longitude located at about 435 km North of
Addis Ababa, Capital City of Ethiopia. It lies within an altitude of 1680 meter above sea level.
The area receives an average annual rainfall of 1204.6mm and a mean range temperature of 11.2
- 25.6°C. There is black soil type and clay soil texture. Chefa is located about 355 km North of
Addis Ababa, Capital City of Ethiopia. It lies within an altitude of 1450 meter above sea level.
The area receives an average annual rainfall of 850 mm and a mean range temperature of 21 -
36°C. There is black soil type and clay soil texture. Sirinka is situated at 11°45’00’’N latitude and
39°36’36’’E longitude located at about 508 km North of Addis Ababa, Capital City of Ethiopia. It
lies within an altitude of 1850 meter above sea level. The area receives an average annual rainfall
of 950 mm and a mean range temperature of 16 - 21°C. There is Eutric soil type and clay soil
texture (Adem et al., 2016).

Experimental Procedures and Design

The experimental fields were divided into three blocks and the twelve treatments randomly
assigned to the plots in each block using a randomized complete block design (RCBD). Twelve
released soybean varieties were used as a treatment. Planting was done the second week of July
2018 and 2019 right after the onset of the rainy season. The seeds were planted in 40 cm between
rows and 10 cm between plants on a plot size of 3.2m * 4m where each plot contains 8 rows and
the total area of the experiment was 620.6m2. Spacing between blocks and plots were 1m and
0.5m, respectively. The seed rate was 60 kg/ha. A fertilizer 100 kg/ha NPS was applied during
seed planting. For estimation of biomass yield only the entire three rows were harvested at 50%
of flowering stage and the remaining three rows used for seed yield and haulm yield production.
Soybean varieties Afgat, Belesa-95, Boshe, Cheri, Dhidhessa, Gishama, Gizo, Korme, Pawe-03,
katta, Wegayen and Wollo were described in Table 1.

186
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Backgrounds of soybean varieties used as a treatment

No. Variety Name Year of Release/Regester Releaser Organization/Institution
1 Afgat 2007 Awassa Agricultural Research Center (AwARC)
2 Belesa-95 2003 Areka Agricultural Research Center (ARARC)
3 Boshe 2008 Bako Agricultural Research Center (BARC)
4 Cheri 2003 Bako Agricultural Research Center(BARC)
5 Dhidhessa 2008 Bako Agricultural Research Center(BARC)
6 Gishama 2010 Pawe Agricultural Research Center (PARC)
7 Gizo 2010 Pawe Agricultural Research Center (PARC)
8 Korme 2011 Bako Agricultural Research Center(BARC)
9 Pawe-03 2016 Pawe Agricultural Research Center (PARC)
10 katta 2011 Bako Agricultural Research Center (BARC)
11 Wegayen 2010 Pawe Agricultural Research Center (PARC)
12 Wollo 2012 Sirinka Agricultural Research Center (SARC)
Source: MoANR, 2016

Data Collection and Statistical Analysis

The varieties were harvested at 50% flowering stage and evaluated for their biomass yield
(ton/ha) and chemical composition (CP, NDF, ADF, ADL and Ash). The harvested materials
were dried in a forced draft oven for 72 hours at 65oC till constant weight was obtained for each
variety. Seed yield, haulm yield and other agronomic characteristics were collected. The data
which were collected was estimated to the sampled three rows and then extrapolated to hectares
of land. Plot coverage, vigorsity and leafiness score were collected on given lebled scoring (1-9)
and biomass yield was taken when the plant reached optimum harvesting stage (50% flowering)
and measured fresh green weight of sample and extra pollet to hectare of land.

The dry matter content was taken on the fresh green weight dried in a forced draft oven for 72
hours at 65oC till constant weight and converted to sample yield to hectare yield multiplied by
dry matter percentage. Ten plants were taken randomly and measured for plant height, plant
height at maturity and number of pod per plant parameters. The DM, ash, nitrogen analysis,
NDF, ADF and ADL were determined according to the procedure of Van Soest (1970). The data
were subjected to analysis of variance (ANOVA) using the Genstat computer software program
and significant mean differences were separated using Duncan’s multiple range test (DMRT) at
5%.

187
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Dry Matter Yield and other Morphological Characteristics of Soybean Varieties

The combined mean (Table 2) showed that there was highly significant variation (P<0.001)
between varieties in all parameters. There was significant difference (P<0.05) among locations in
plot coverage, days to 50% flowering and biomass yield, but no significant difference (P>0.05)
between locations under vigorsity, leafiness score, dry matter yield and plant height at 50%
flowering stage. However, there was no significant difference (P>0.05) in variety by location
interactions except days to 50% flowering which showed significant difference (P<0.05) on
interaction effect.

The combined analysis across locations (Jari and Chefa) for year 2018 showed that varieties
Afgat, Gizo, Pawe-03 and Wollo had higher dry matter yield as compared to other soybean
varieties. The Afgat, Gishama, Gizo, Pawe-03 and Wollo varieties had higher biomass yield as
compared to other soybean varieties. The dry matter yield of Afgat was 6.54 t/ha lower than 7.34
t/ha reported by Suzan et al. (2004) and 6.72 t/ha reported by Assaeed et al. (2000). However, the
current result was sufficient for supporting livestock production as the existing environment.

The combined analysis of variance over two years at Jari for vigorsity, leafiness score, days to
50% flowering, biomass yield, dry matter yield and plant height at 50% flowering (Table 3)
showed that there were highly significance difference (P<0.01) between varieties but there was
no significant difference (P>0.05) on plot coverage. There were significant differences (P<0.05)
among years on leafiness score, days to 50% flowering, biomass yield, dry matter yield and plant
height at 50% flowering except plot coverage and vigorsity were no significance difference
(P>0.05) between years. Plot coverage, vigorsity, days to 50% flowering and plant height at 50%
flowering showed that significance difference (P<0.05) on varieties by year interactions effect.
However, leafiness score, biomass yield and dry matter yield had no significant difference
(P>0.05) on varieties by year interaction effect.

188
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The presence of significance over varieties by year interactions indicated that there was
inconsistent performance of the varieties for this character over years. However, the presence of
non significant interaction indicates the consistent performance of the variation over years. The
mean biomass yield of varieties Afgat, Gizo, Pawe-03 and Wollo were significantly higher 17.2
t/ha, 16.1 t/ha, 16.7 t/ha, 16.71 t/ha, respectively as compared to other varieties. And also, the
mean dry matter yield of varieties Afgat, Gizo, Pawe-03, Wogayen and Wollo were significantly
higher 4.99 t/ha, 4.34 t/ha, 4.64 t/ha, 4.19 t/ha and 4.79 t/ha, respectively as compared to other
varieties. The dry matter yield of Afgat was 4.99 t/ha lower than 7.34 t/ha reported by Suzan et
al. (2004) and 6.72 t/ha reported by Assaeed et al. (2000).

Table 2: The combined mean of dry matter yield and other related characteristics of soybean
varieties over locations (Jari and Chefa) in 2018
Variable N PC Vg LS DF 50% BY (t/ha) DMY (t/ha) PH50% (cm)
Variety 72
Afgat 6 8.833a 8.5a 8.833a 65.83c 14.68a 6.541a 81.13ab
Belessa-95 6 7.667bcde 6.833d 7.167c 68.67b 9.76de 4.525de 71.52cd
Boshe 6 7.5cde 7d 7c 62d 8.43e 4.646cde 64.9e
Cheri 6 8abcd 8abc 7c 68.33b 10.59cd 5.323bcde 77.16bc
Dhedessaa 6 7.167de 7.167cd 6.833c 69.67b 9.07de 4.29e 70.28de
Gishama 6 7.667bcde 8.333a 8.167ab 69.5b 12.62ab 5.315bcde 79.25ab
Gizo 6 8.5abc 8.5a 8.333ab 69.33b 13.25ab 5.879ab 78.28ab
Korme 6 7.833abcd 7.333bcd 7.167c 69.33b 9.03de 4.5de 70.91de
Pawe-03 6 8.5abc 8.167ab 8b 72.33a 13.05ab 5.939ab 77.2bc
Kata 6 6.833e 7d 7.167c 69.33b 9.46de 4.266e 70.61de
Wogayen 6 8abcd 8.5a 8b 65.5c 12.11bc 5.467bcd 83.87a
Wollo 6 8.667ab 8.5a 8.833a 65.83c 13.24ab 5.698abc 81.16ab
Gmean 7.931 7.819 7.708 67.97 11.27 5.199 75.52
SEM 0.111 0.112 0.105 0.382 0.306 0.123 0.862
CV 9.6 9.4 7.8 2 14.3 15.7 6.5
SL V *** *** *** *** *** *** ***
SL LO ** NS NS *** *** NS NS
SL VX LO NS NS NS **NS NS NS
PC (%) = plot coverage; Vg = Vigorsity of plant; LS = leafiness score of plant; DF (50%) =
date at 50% flowering stage; PH (50%) = plant height; BY = biomass yield; DM = dry matter
yield; N = number of observation; SEM = standard error of mean; CV = coefficient of variation;
SLV = significant level of variety; SL LO= significant level of location; SL VX LO = Significant
level of variety by location interaction; NS = non significant at P>0.05; ** = significant at
P<0.01; *** = significant at P<0.001; means with the same superscript in columns are not
significant difference

189
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: The combined mean of dry matter yield and other related characteristics of soybean
varieties over years (2018 and 2019) at Jari
Variable N PC Vg LS DF 50% BY (t/ha) DMY (t/ha) PH50% (cm)
Variety 72
Afgat 6 8.5 8.667a 8.5a 63.33e 17.2a 4.993a 71.77a
Belessa-95 6 8.17 7.167cd 7c 66.83c 9.45c 3.159cd 58.97bc
Boshe 6 8 7.167cd 7.333bc 60.67f 8.14c 3.026cd 56.75bc
Cheri 6 8.17 7.833abc 7c 66.5c 10.41c 3.676bcd 61.99b
Dhedessaa 6 8 7.333cd 7c 66.83c 9.8c 3.038cd 59.63bc
Gishama 6 7.5 8.167ab 7.667abc 69.17b 13.63b 3.813bc 69.45a
Gizo 6 8.5 8.5a 8.167ab 67.67bc 16.1ab 4.335ab 70.27a
Korme 6 8.17 7.333bcd 7.5bc 67.33bc 9.56c 3.364cd 59.19bc
Pawe-03 6 8.5 7.5bcd 7.333bc 74.83a 16.7a 4.635a 71.7a
Kata 6 7.67 6.667d 7c 67c 8.63c 2.888d 56.11c
Wogayen 6 8.33 8.333a 7.667abc 65.5cd 13.36b 4.189ab 71.77a
Wollo 6 9 8.667a 8.167ab 64de 16.71a 4.787a 74.61a
Gmean 8.21 7.778 7.528 66.64 12.47 3.825 65.18
SEM 0.115 0.119 0.108 0.535 0.545 0.192 1.473
CV 10.7 8.4 9.6 2.5 18.6 16.8 6.8
SL V NS *** ** *** *** *** ***
SL Y NS NS * *** *** *** ***
SL VXY * * NS ** NS NS *
PC (%) = plot coverage; Vg = Vigorsity of plant; LS = leafiness score of plant; DF (50%) =
date at 50% flowering stage; PH(50%) = plant height; BY = biomass yield; DM = dry matter
yield; N = number of observation; SEM = standard error of mean; CV = coefficient of variation;
SLV = significant level of variety; SL Y = significant level of year; SL VXY = Significant level of
variety by year interaction; NS = non significant at P>0.05; * = significant at P<0.05; ** =
significant at P<0.01; *** = significant at P<0.001; means with the same superscript in
columns are not significant difference

The combined mean (Table 4) showed that there was highly significant variation (P<0.001)
between varieties in all parameters. There were highly significant differences (P<0.001) among
locations in day’s to 50% flowering, plant height at 50% flowering, biomass yield, dry matter
yield and haulm yield but no significance difference (P>0.05) between locations under vigorsity.
There was a significant difference (P<0.05) in variety by location interaction under vigorsity,
days to 50% flowering, plant height at 50% flowering and biomass yield but dry matter and
haulm yield were no significance difference (P>0.05) on interaction effect.

190
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The combined analysis across locations showed that varieties Afgat, Gishama, Gizo, Pawe-03,
Wogayen and Wollo had higher dry matter yield as compared to the other soybean varieties. But,
there was low yield in this cropping season (2019) as compared with first year (2018) because,
unfortunately, in this cropping season, there was heavy rain intensity with snow during early
stage of growing at Sirinka. The dry matter yield of Pawe-03 was 2.39t/ha lower than 7.34 t/ha
reported by Suzan (2004) and 6.72 t/ha reported by Assaeed et al. (2000). The combined analysis
across locations showed that variety Gishama (3.97t/ha), Gizo (3.60 t/ha), Pawe-03 (4.04 t/ha)
and Wogayen (3.36 t/ha) had higher haulm yield as compared to the other soybean varieties. The
haulm yield of Pawe-03 (4.04 t/ha) was higher than 3.23 t/ha reported by Sisay (2017) but lower
than 5.31 t/ha reported by Yagoub et al. (2012).

Table 4: The combined mean of dry matter yield and other related characteristics of soybean
varieties over locations (Jari and Sirinka) 2019
Variable N Vg DF (50%F) PH (50%F) BY (t/ha) DMY (t/ha) HY (t/ha)
Variety 72
Afgat 6 8.17a 59.33ef 52.07b 13.08a 2.19ab 3.28bc
Belessa-95 6 7.42b 62.00cd 45.73c 7.19b 1.41c 2.29d
Boshe 6 7.25b 58.17f 46.77c 7.12b 1.53c 1.85d
b cde c
Cheri 6 7.42 61.00 44.00 8.30b 1.67bc 2.38d
Dhedessaa 6 7.42b 60.50cdef 45.13c 7.81b 1.52c 2.09d
Gishama 6 8.42a 67.67b 57.73a 13.56a 2.38a 3.97ab
Gizo 6 8.25a 63.00c 52.93ab 13.34a 2.08ab 3.60abc
b cde c
Korme 6 7.25 61.17 44.90 7.78b 1.41c 1.94d
Pawe-03 6 7.08b 75.17a 52.58ab 13.88a 2.39a 4.04a
Kata 6 6.83b 61.50cde 42.20c 6.66b 1.31c 2.30d
Wogayen 6 8.17a 61.17cde 56.87ab 11.37a 2.15ab 3.36abc
a def ab
Wollo 6 8.25 60.00 55.53 13.37a 2.32a 3.09c
Gmean 7.66 62.56 49.70 10.29 1.86 2.85
SEM 0.10 0.6 1.12 0.69 0.11 0.13
CV 7.3 3.1 8.2 21.6 22.4 20.3
SL V *** *** *** *** *** ***
SL LO NS *** *** *** *** ***
SL VXLO * * *** *** NS NS
Vg = Vigorsity of plants; DF (50%) = date at 50% flowering stage; PH(50%) = plant height; BY =
biomass yield; DMY = dry matter yield; HY = haulm yield; N = number of observation; SEM = standard
error of mean; CV = coefficient of variation; SLV = significant level of variety; SL LO = significant level
of location; SL VX LO = Significant level of variety by location interaction; NS = non significant at
P>0.05; * significant at P<0.05; *** significant at P<0.001; means with the same superscript in columns
are not significant difference

191
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Seed Yield and Other Agronomic Characteristics of Soybean Varieties

The combined mean (Table 5) showed that there was highly significant variation (P<0.001) between
varieties in all parameters. There were significant differences (P<0.05) among locations in date of
maturity, plant height at maturity, seed yield and thousand seed weight but no significance difference
(P>0.05) between locations under number of pods per plant. There was significant difference
(P<0.05) in variety by location interaction under date of maturity, plant height at maturity and
thousand seed weight but number of pods per plant and seed yield had no significance difference
(P>0.05) on interaction effect. The combined analysis across locations showed that variety Pawe-03
(2951 kg/ha), Gizo (2862 kg/ha), Afgat (2859 kg/ha), Gishama (2654 kg/ha), Wollo (2461 kg/ha) and
Wogayen (2404 kg/ha) had higher seed yield as compared to the other soybean varieties. The seed
yield of Pawe-03 was 2951 kg/ha greater than 2609.1 kg/ha reported by Deresse et al. (2019).

Table 5: The combined mean of Seed yield and other agronomic characters of soybean varieties over
locations (Jari and Sirinka) 2019
Variable N DM PHM (cm) NPPP SY (kg/ha) 1000swt (gm)
Variety 72
Afgat 6 136de 59.6bcd 65.3b 2859a 134.3abc
abc b de bc
Belessa-95 6 147.17 69.5 38.9 1539 137.4abc
f cd de c
Boshe 6 129.33 54.6 37.3 1256 94.9g
de d de bc
Cheri 6 136.5 50.9 38.4 1337 114.9ef
cd b de bc
Dhedessaa 6 140.33 70.6 41.9 1406 118.3de
a a bc a
Gishama 6 152.5 89.1 62.1 2654 131.1abcd
Gizo 6 146.33abc 67.7bc 66.4b 2862a 139.9ab
bcd bcd de bc
Korme 6 142.5 59.9 36.6 1465 124cde
ab b a a
Pawe-03 6 148.83 71.1 88.8 2951 104.1fg
bc bc e b
Kata 6 143.5 65.4 35.7 1841 133.8abc
abc b cde a
Wogayen 6 147.17 71.3 49 2404 129.2bcd
ef bcd c a
Wollo 6 131.5 58.9 51 2461 145a
Gmean 141.81 65.7 50.9 2086 125.6
SEM 1.17 3.02 2.26 100.6 2.21
CV 3.8 16.8 22.3 20.9 8.9
SL V *** *** *** *** ***
SL LO *** *** NS *** **
SL VX LO * *** NS NS *
DM = date of maturity; PHM = plant height at maturity; NPPP = number of pod per plant; SY = seed yield;
swt = seed weight; N = number of observation; SEM = standard error of mean; CV = coefficient of variation;
SLV = significant level of variety; SL LO = significant level of location; SL VX LO = Significant level of
variety by location interaction; NS = non significant at P>0.05; * =significant at P<0.05; ** = significant at
P<0.01; *** = significant at P<0.001; means with the same superscript in columns are not significant
difference

192
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Chemical Composition of Soybean Varieties

The mean nutrient composition of soybean varieties is shown in Table 6. The mean CP content
range from Boshe 21.30% to Gizo 33.60%, NDF from Cheri 28.79% to Belessa-95 58.57%,
ADF from Cheri 19.78% to Afgat and Belessa-95 35.16% and ADL from Cheri 4.56% to Afgat
and Belessa-95 8.94% at Chefa district. The CP content of Gizo was higher as compared to
Boshe that means it is a good source of protein supplement for different production systems. The
NDF content of Belessa-95 was higher as compared to Cheri so it was more fibrous as compared
to Cheri and other varieties. The ADF content of Afgat and Belessa-95 were higher as compared
to Cheri that means Afgat and Belessa-95 had the highest lingo-cellulose content as compared to
Cheri and other soybean varieties. The ADL content of Afgat and Belessa-95 were higher as
compared to Cheri and other varieties. This indicated that Afgat and Belessa-95 had the highest
lignin content as compared to Cheri and other varieties. The crude protein content of Gizo
(33.60%) was comparatively higher than 24.5% (Craig et al., 1998) and 20.01% reported by
Assaeed et al. (2000) at Chefa.

Table 6: Chemical composition of soybean varieties (% DM basis) in 2018

Location Chefa
Variable N DM% Ash % CP % NDF% ADF% ADL%
Variety 36
Afgat 3 91 9.89 29.68 47.67 35.16 8.94
Belessa-95 3 91 10.99 24.89 58.57 35.16 8.94
Boshe 3 91 10.99 21.30 43.79 32.97 7.61
Cheri 3 91 9.89 23.20 28.79 19.78 4.56
Dhedessaa 3 90 11.11 22.23 44.44 31.11 7.77
Gishama 3 91 10.99 29.12 44.68 32.97 7.67
Gizo 3 91 9.89 33.60 44.68 32.97 8.64
Korme 3 89 11.24 24.92 46.71 33.71 8.74
Pawe-03 3 91 8.79 27.50 31.75 21.98 4.72
Kata 3 91 10.99 28.69 45.79 32.97 4.61
Wogayen 3 91 10.99 27.34 37.59 26.37 5.79
Wollo 3 94 13.83 31.60 46.82 34.04 8.64
Overall mean 91 10.80 27.01 43.44 30.77 7.47
DM = dry matter; CP = crude protein; NDF = neutral detergent fiber; ADF = acid detergent
fiber; ADL = acid detergent lignin

193
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The mean nutrient composition of soybean varieties is shown in Table 7. The mean CP content
range from korme 17.20% to Wollo 35.56%, NDF from Boshe 28.79% to Belessa-95 58.57%,
ADF from Wogayen 19.57% to kata 40%, ADL from Boshe 4.56% to Kata 10% at Jari. The
numerical values showed that the CP content of Wollo was highest as compared to others that
meant it was a good source of protein supplement for different production systems. The NDF
content of Belessa-95 was higher as compared to other varieties so it was more fibrous as
compared to other varieties. The ADF content of Kata was highest as compared to other varieties
that meant kata had highest lingo-cellulose content compared to other soybean varieties. The
ADL content of Kata has the highest as compared to other varieties, which indicates that it has
the highest lignin content as compared to other varieties. The variety Kata could not easily be
digested and absorbed by animals that were relatively high in crude fiber and low in total
digestible nutrients and protein.

The mean CP content range from korme 14.59% to Wollo 27.66%, NDF from Wollo 40.8% to
Belessa-95 56.14%, ADF from Wollo 24.74% to Gishama 35.05%, ADL from Gizo and Wollo
7.27% to Gishama and Korme 9.1% at Sirinka. The numerical value showed that the CP content
of Wollo was highest as compared to others that meant it was a good source of protein
supplement for different production system.This indicated that the variety Wollo had high crude
protein content and digestible nutrients that can support animal production over maintenance
requirement. It was used as supplementation of roughage feed sources used as animal feed. The
NDF content of Belessa-95 was higher as compared to other varieties so it was more fibrous than
others. The ADF content of Gishama was highest as compared to other varieties that mean
Gishama had highest lingo-cellulose content as compared to other soybean varieties. The ADL
content of Gishama and Korme were highest as compared to other varieties. This indicated that
Gishama and korme had the highest lignin contents as compared to other varieties.

The crude protein content of variety Wollo 35.56%, 27.66% at Jari and Sirinka, respectively was
comparatively higher than 24.5%, (Craig et al., 1998) and 20.01% (Assaeed et al., 2000).
Therefore, it is possible to generalize that all varieties in all three locations could be used as
protein supplements for growing and lactating animals where soybean varieties can be accessible
to producers.

194
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 7: Chemical composition of soybean varieties (% DM basis) in 2019

Location Jari Sirinka
Variable DM% Ash % CP % NDF% ADF% ADL% DM% Ash % CP % NDF% ADF% ADL%
Variety 36 36
Afgat 3 91 10.99 32.46 41.35 30.77 7.67 3 96 10.42 26.56 44.32 27.08 7.32
Belessa-95 3 91 10.99 20.98 58.57 35.16 8.94 3 96 6.25 18.09 56.14 31.25 8.74
Boshe 3 91 9.89 18.8 28.79 19.78 4.56 3 97 12.37 17.81 45.38 28.87 8.12
Cheri 3 90 10 20.12 33.33 22.22 5.55 3 97 12.37 17.88 45.38 28.87 8.15
Dhedessaa 3 91 9.89 17.68 44.68 32.97 8.64 3 95 12.63 20.31 45 29.47 8.64
Gishama 3 91 10.99 29.24 43.59 32.97 7.67 3 97 11.34 23.06 55.62 35.05 9.1
Gizo 3 91 9.89 23.4 47.39 35.16 8.94 3 97 10.93 26.03 44.29 26.80 7.27
Korme 3 91 9.89 17.2 42.15 30.77 7.67 3 96 12.50 14.59 50.2 33.33 9.1
Pawe-03 3 91 10.99 30.49 50 39.56 9.82 3 97 11.34 25.09 50.15 32.99 8.4
Kata 3 90 11.11 20.4 52.22 40 10 3 97 13.40 18.53 45.5 30.93 8.31
Wogayen 3 92 9.78 27.63 30.24 19.57 4.68 3 97 11.34 18.13 45.38 28.87 8.15
Wollo 3 91 9.89 35.56 45.75 32.97 7.61 3 97 11.34 27.66 40.8 24.74 7.27
Overall mean 91 10.36 24.5 43.17 30.99 7.65 96.58 13.02 21.15 47.35 29.85 8.21
DM = dry matter; CP = crude protein; NDF = neutral detergent fiber; ADF = acid detergent fiber; ADL = acid detergent lignin

195
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATION

The result indicated that there were significant differences among soybean varieties for dry
matter yield, haulm yield, seed yield and other agronomic traits. The varieties in their
combined analysis showed that Afgat, Pawe-03 and Wollo gave better in most important
parameters than the others. The CP content of all varieties had above the optimal level to use
as a protein supplement in livestock nutrition. Specifically, Wollo and Gizo varieties gave
higher crude protein content in three locations than others. The crude protein content of such
varieties indicated that these species can be used as supplementary feed for livestock
production. Therefore, Afgat, Pawe-03 and Wollo were recommended for the given areas of
Jari, Chefa, Sirinka and could be produced in similar environment for the beast of produced
optimal amount of dry matter, haulm and seed yield as forage of high quality and good
protein supplement for growth and production of ruminant. Thus, further research is needed
to investigate the utilization of livestock and varieties response to production.

ACKNOWLEDGEMENTS

The authors would like to thank SARC and ARARI for financial and valuable contribution in
the compilation of the study. Our special thanks go to SARC livestock research directorate
researchers for their unreserved over all support.

REFERENCES

Adem Mohammed, Tamado Tana, Piara Singh and Adamu Molla. 2016. Modeling climate
change impact on chickpea production and adaptation options in the semi-arid North-
Eastern Ethiopia. JAEID, 110 (2): 377-395.
Alemayehu Mengistu and Getnet Assefa. 2012. The Evolution of Forage Seed Production in
Ethiopia. Forage Seed Research and Development in Ethiopia. Ethiopian Institute of
Agricultural Research. Addis Ababa, Ethiopia. pp.15-32.
Alemayehu Mengistu, Getnet Assefa, Gezahagn Kebede and Fekede Feyissa. 2016. Review
on the Evolution of Forage Seed Production in Ethiopia: Experiences, Constraints and
Options. Academic Research Journal of Agricultural Science and Research. Vol. 4(6),
pp. 231-240.
Alkhlib Ashraf, Jane W., Barbara R., Yahya K. and Mohannad M. 2014. Status, challenges
and opportunities for sustainable utilization of pulse crop residues in Ethiopia. Africa
RISING in the Ethiopian Highlands.
Assaeed A. M., Saiady M. Y. and El-Shawaf I. I. 2000. Yield and quality of soybean forage
as affected by harvesting time and cultivar. Res. Bult., No (89), Agric. Res. Center, King
Saud Univ., PP. (5-13). (5-13). (5-13). (5-13).

196
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Craig Altemose, Jeff, Emile senesac and Daniel Baker. 1998. Forage soybean and corn
research trial a collaborative research project. Between UVM extension service and
senesac farm. Colchester,Vermont.
Deresse Hunde, Mola Malede, Asmamaw Amogne and Gezahegn Tefera. 2019. Genotype by
Environment Interaction and Yield Stability Studies on Medium Maturing Soybean
Genotypes Tested in Northwest, Southern and Western Parts of Ethiopia for two
Consecutive Years (2016-2017) Across Five Testing Locations. ACTA SCIENTIFIC
AGRICULTURE (ISSN: 2581-365X). Volume 3 Issue 9.
Doo-Hong Min. 2012. Forage soybean as an alternative forage crop. Michigan State Univ.
Extension.
Fujita k., Ogata S., Matsumoto K., Masuda T., Godfred k., Ofosu-Budu G.K. and Kuwata K.
1990. Nitrogen transfer and dry matter production in soybean and sorghum mixed
cropping systems at different population densities. Soil sci.plant Nutr., 36(2): 233-241.
Hintz R. W., Albrecht K. A. and Oplinger E. S. 1992. Yield and quality of soybean forage as
affected by cultivar and management practices. Agron. J., 84 (5): 795–798.
Koivisto J.M., Devine T.E., Lane G.P.F., Sawyer C.A. and Brown H.J. 2003. Forage
soybeans (Glycine max (L.) Merr.) In the United Kingdom: test of new cultivars.
Agronomie, 23(4):287-291.
Koivisto J. 2006. Glycine max L. Grassland Index. A searchable catalogue of grass and
forage legumes.
Lugin buhl J-M. 2006. Pastures for meat goats. In: Gipson T.A., Merkel R.C., Williams K.,
Sahlu T. Meat goat production hand books, Langston University.
Maheri-Sis N., B. Abdollahi-Ziveh, R. Salamatdoustnobar, A. Ahmadzadeh, A.
Aghajanzadeh-Golshani and M. Mohebbizadeh. 2011. Determining Nutritive Value of
Soybean Straw for Ruminants Using Nylon Bags Technique. Pakistan Journal of
Nutrition 10 (9): 838-841; ISSN 1680-5194.
MoANR. 2016. Plant Variety Release, Protection and Seed Quality Control Directorate. Crop
Variety Register Issue No. 19. Addis Ababa, Ethiopia.
Murphy A.M. and Colucci P.E. 1999. A tropical forage solution to poor quality ruminant
diets: a review of Lablab purpureus Livestock Research for Rural Development [online]
11 (2) 16 p .http: www.Cipav. Org. Co/lrrd/index. Htm/.
Sisay Belete. 2017. Effects of Phosphorus Fertilizer and Inoculation on Yield and Nutritive
Values of Grain and Haulm of Selected Grain Legumes in Mixed Crop-Livestock
Production System of Ethiopia. M.Sc.Thesis. Hawassa University, Ethiopia.
Suzan Altinok, Ilker Erdogdu and Istvan Rajcan. 2004. Morphology, forage and seed yield of
soybean Cultivars of different maturity grown as a forage crop inTurkey. In: plant
science 84: 181-186.
Yagoub S. O, Ahmed W. M. A. and Mariod A. A. 2012. Effect of Urea, NPK and Compost
on growth and yield of soybean (Glycinemax L.), in Semi-Arid Region of Sudan.
International Scholarly Research Network. ISRN Agronomy.

197
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Participatory On-Farm Evaluation of Different Forage Species in Bure and Guagusa

Shikudad Districts of Western Amhara Region, Ethiopia

Adebabay Adane, Beyadglign Hunegnaw, Misganaw Walie, Berhanu Demeke, Mola Haile,
Yohanes Amsalu, Desalegn Ayichew and Wondimeneh Mekonnen

Andassa Livestock Research Center, P.O. Box 27, Bahir Dar, Ethiopia

ABSTRACT

The trial was carried out during 2017 and 2018 cropping seasons in Bure zuria (mid altitude)
and Guagusa Shikudad (Highland) districts on five farmers' fields on one kebele in both
districts. The Kebeles were selected based on experience of smallholder dairy and improved
forage production potentials. Four forage species (Napier grass, Desho grass, Sweet lupine
and Cowpea at Bure zuria and at Guagusa Shikudad district; Napier grass, Desho grass,
Sweet lupine and Oat grass) and within each four varieties or accessions were evaluated. The
objective was to evaluate the on farm Yield performance and select best performing
improved forage species through farmers’ participation under farmers and researchers
management scope. The design of the experiment was randomized complete block design
(RCBD) with five farmers considered as replications and four forage treatments. Farmers’
perception at vegetative stage was taken by using both pair wise ranking and preference
ranking matrix methods. Data were analyzed using the General Linear Model procedures of
SAS (2002). The combined analysis of variance done over two years showed that there was
significant difference among treatments. The result of growth and yield performance of
forage accessions and farmers' preference selection among forage cultivars or varieties,
15743, Kindo kosha-DZF#591,bora and ILRI-12688A showed better value from Napier,
desho, sweet blue lupin and cowpea forage types at Burie zuria district (mid altitude)
respectively. On the other hand at Guagusa Shikudad district (high altitude) 15743, Areka
DZF#590, vitabor and CI-8237 from Napier, desho, sweet blue lupin and oat forage types
showed better production performance. Based on the findings each forage species can be
recommended for each agro-ecologies accordingly.

Keywords: participatory variety selection, dry matter yield, improved forage, on farm
evaluation

198
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Livestock production is the integral part of the farming system in Ethiopia. The sector has
considerable economic and social importance at household and national levels in Ethiopia
(Gelayenew et al., 2016; CSA, 2019). However, productivity of livestock is low mainly due
to inadequate supply of feed and poor feeding practice (Gelayenew et al., 2016). The major
feed resources for livestock in Ethiopia are natural pastures, crop residues and aftermath
grazing (Getnet, 2012; CSA, 2016). These types of feed resources cannot support the
expected animal productivity due to their nutritional limitations. To overcome these
problems, production of improved forage, in the backyards and farm lands can be sought
among available options. According to Getnet (2012), the integration of improved forage
crops in agricultural systems has many advantages including soil conservation; reduced
weeds, pests, and diseases, in addition to their primary use as high quality animal feeds.

The yield and nutritional qualities of forage are influenced by numerous factors representing
ecological conditions and management activities (Enoh et al., 2005). It is important to screen
forage varieties using empirical evidence and participatory assessment and selection
techniques to insure that not only yield is considered but also other forage traits preferred by
smallholder farmers for easy adoption (Gregory et al., 2015) The mean dry matter yield of
Napier grass accessions was reported 1.4t/ha under on farm participatory evaluation in north
Tanzania (Gregory et al., 2015). According to Wondmeneh et al. (2016 unpublished), the
mean dry matter yield per harvest and crude protein content of ten Napier accessions was
12.65 ton ha -1 and 8.25%, respectively, in northwestern Ethiopia. On the other hand highest
yield (12.34 t/ha) was obtained for the same grass from cutting at 1.0 m with 92 kg N kg/ha
(Tesema et al., 2003). The dry matter yield in ton per hectare desho grass under irrigation
condition at Wondo Genet was 28.35, 26.52, 23.37 and 21.95 for grass lines of Areka-
DZF#590, Kulumsa-DZF#592, Kindo kosha-DZF#589 and Kindo kosha-DZF#591,
respectively (Tekalign et al., 2017).

So far research has been conducted to identify potential forage species for different agro-
ecologies and as a result of that different forage species have been recommended. These
forage species have been tested on-farm and they were promising. However, development
improved forage without participatory evaluation of farmers seems meaningless for a
smallholder farmer.

199
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

There were no detailed participatory activities conducted in western Amhara. Unavailability

of forage seed, socio-economic status of the smallholder farmer and poor extension service
are among the major constraints for the increment of improved forage production under
farmers’ condition. Consequently, there is no formal documentation of farmers' production on
improved forage types and varieties preferences by smallholder farmers were not addressed
in western Amhara, Ethiopia. Therefore the objectives of this study was on farm evaluation of
the growth and yield performance of different forage species and to select best performed
improved forage species through farmers participation in the study districts.

MATERIALS AND METHODS

Description Study Area of District

The trial was carried out during 2017 and 2018 cropping seasons in Bure zuria (mid altitude)
and Guagusa Shikudad (Highland) district in Amhara National Regional State (ANRS),
Ethiopia. Bure zuria is situated between 10˚15’N and 10˚42’29”N latitude and between
36˚52’1’’E and 37˚7’9’’E longitude at an altitude of 2091 meter above sea level. On the other
hand Guagusa Shikudad district is found at 10°50'60'' N latitude and 37°1'0''E longitude at an
altitude of 2500 meter above sea level.

Selection of Participant Farmers and Districts

Among AGP-II supported districts of western Amhara Region, Bure zuria and Guagusa
Shikudad were selected with zonal livestock experts based on criteria like lack of improved
forage production and utilization under smallholder farmers. Ten farmers five from each
district were selected by giving equal chances based on consultation with agricultural experts
and key informants knowledgeable about the forages and small scale dairy farming practice.
Twenty five farmers in one kebele of one district are considered as a FREG (farmer research
group) from each district. Training was given to participant farmers and experts at the
beginning and mid of trial for refreshment

200
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Participatory Evaluation with Farmers

Five farmers were selected from each Bure zuria and Guagusa Shikudad districts were used
to evaluate Napier grass, Desho grass, Oats grass, cowpea and sweet blue lupin forages.
Selection of individual farmers was made on meeting with key informants that are familiar
with the forages to determine the adaptability and growth performance throughout the entire
growing period and experience about small scale dairy farming practice. Group discussions
were held to carefully build on and critically examine, the information derived from
individual farmers of different households. It was also intended to clear conflicting ideas on
issues like adaptability and growth performance of the forages. The group discussions
focused on i) preference and selection of forage ii) availability of feed during the dry season
iii) growth performance and biomass yield of each crop and other related parameters.

Land Preparation, Planting and Experimental Design

Table 1: Treatments and experimental district

Forage species Entries/ Treatments District
Napier (Pennisetum purpureum) 16815, 15743, 14984 and 16791 Bure zuria and
Guagusa Shikudad
Desho (Pennisetum graucifolium) Kindo kosha-DZF#589, Areka Bure zuria and
DZF#590,Kindo kosha-DZF#591 Guagusa shikudad
and Kulumsa-DZF#592
Sweet blue lupin (Lupinus Spp) Probor, sanabor, sanabor and Bure zuria and
vitabor Guagusa shikudad
Cow pea (Vigna unguiculata (L.) ILRI-9333A, ILRI-12688A, Bure zuria
Kenkety and blackeyebeen
Oat (Avena sativa) CI-8251, CI-8237, CI-2291 and Guagusa shikudad
lampton

A total area of 441m2 was prepared for each farmer in both districts .The experimental land
was plowed and harrowed before June 2017. The land was divided into four blocks each of
which comprised four plots (4*4 m each). The forages were planted using vegetative root
splits and seed in rows with distance between row and distance between plant for Napier,
desho, sweet lupin, cowpea and oat was (1 and 0.5m), (0.5 and 0.5m), (30 and 7cm), (30and
10cm) and 20cm and at rate of 80kg/ha) respectively and distance between block and plot
1m.

201
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The experiment was laid out in a randomized complete block design (RCBD) arrangement
for each forage species with five replications (farmers consider as replication) in each district.
Land preparation, planting, weeding, harvesting and related management practices were
applied according to standard practice for the forages by farmers and researchers. Artificial
fertilizers (NPS (19:38:7) compound and urea) were applied at a rate of 100 kg/ha and urea at
25 kg/ha during planting (for the grasses only).

Data Collection

Growth data like Plant height (cm) and tiller number were measured from 5 plants that were
randomly selected from middle rows of each plot. Yield parameters like Number of pods per
plant, number of seeds per pod, thousand seed weight (g), biomass yield (kg) and grain yield
(kg ha-1) were taken in standard procedure.

Farmers’ Perception and Variety Selection

Farmers’ perception at vegetative stage was taken by using both pair wise ranking and
preference ranking matrix methods in order to capture farmer’s perceptions and criteria in
variety selection. Pairwise ranking was used to weight each of the farmers’ variety selection
criteria as each of the criterions does not have equal weight. Preference ranking was used to
score each of the variety based on the weighted variety selection criteria. During the two
consecutive years the farmer’s perception was taken at vegetative stage based on different
parameters preferred by farmers’ like germination, biomass, height, disease, smoothness of
leaf and pest resistance

Data Analysis

Differences among accessions were tested using analysis of variance (ANOVA) general
linear model procedures of SAS to compare treatment means (SAS version 9.0, 2002). Least
significant difference (LSD) at 5% significance level was used for comparison of means.

202
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Growth and Yield Performance of Forages Species at Study Districts

Growth and Yield Performance of Napier Grass

Forage DM yield showed significant (P<0.05) variation among the tested Napier grass
accessions in Bure zuria district (Table 2). Accession 16791 gave the highest mean DM yield
(5.62 t/ha) per harvest followed by accession 15743 (5.33 t/ha) while 16815 (3.66 t/ha)
recorded the lowest value. On the other hand, at Guagusa Shikudad district there was no
significant difference (P>0.05) due to agro ecological differences. In both districts mean dry
matter yield ranged 2.6 - 4.77 t/ha/harvest was lower as compared to the result (7.05 - 13.06)
of Gezahagn et al. (2017). It is because of only for two years and there was no application of
urea fertilizer after harvesting of Napier in the current experiment.

Mean plant height at harvest was ranged from 101.4 to 108.5cm lower as compared to the
result (96.33 to 140.53) reported by Gezahagn et al. (2017), this was one factor that
contributed for higher dry matter yield in forages. Other findings also indicated that plant
height at cutting significantly affects the fodder yield of Napier grass in Kenya (Muinga et
al., 1992). Amongst the major agronomic practices required, harvesting of Napier grass at
appropriate cutting height and defoliation frequencies are very important to improve DM
yield and nutritive values of this plant (Butt et al., 1993; Tessema et al., 2003).

Table 2: Combined growth and yield performance of Napier grass at Burie zuria and Guagusa
Shikudad districts
Bure zuria Guagusa Shikudad
Accessions N PH (cm) TN DMY (t/ha) N PH (cm) TN DMY (t/ha)
Overall 19 108.5 51.02 4.77 8 101.44 25.78 2.6
b ab ab
15743 5 107.5 50.73 5.33 2 99.5 24.5 2.67
14984 5 113a 49.58ab 4.49bc 2 103.9 22.2 2.74
16791 5 110ab 57.28a 5.62a 2 104 29.2 2.99
c b c
16815 4 103.4 46.48 3.66 2 99.1 27.7 2
CV 2.4 12.2 15.9 4.3 28.4 27.9
SL *** * *** NS NS NS
***p<0.001, *<0.05, Means with different letters in a column are significantly different; N =
umber of harvests CV = coefficient of variation SL = significance level, NS = non significant
at (P>0.05); PH = plant height, TN = tiller number, DMY = dry matter yield

203
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Growth and Yield Performance of Desho Grass

Forage DM yield showed significant (P<0.05) variation among the tested desho grass
accessions in Guagusa Shikudad while non significance at Bure zuria testing sites (Table 3).
At Guagusa Shikudad accession Kindo kosha-DZF#589 gave the highest mean DM yield
(8.64 t/ha) per harvest while Kulumsa-DZF#592 (5.15 t/ha) recorded the lowest value. Even
if the dry matter yield per harvest was high in Guagusa Shikudad district only one harvest
during the experimental period but three harvesting stages/cycles recorded in Bure zuria
district. This is due to the presence of soil acidity and high rainfall in Guagusa Shikudad. In
both districts mean dry matter yield ranged from 6.57 to 6.98 t/ha/harvest higher as compared
to the results (6.26 t/ha) by Tekalegn et al. (2017) and similarly reported by (6.17 t/ha)
Birmaduma et al. (2019) in Daro Labu district of West Hararghe, Ethiopia.

The highest number of tillers over years was obtained from accessions Areka DZF#590,
Kindo kosha-DZF#591 and Kulumsa-DZF#592 while accession Kindo kosha-DZF#589 gave
the lowest. The difference in tillers produced per plant among the accessions of desho grass
could be attributed to genetic variations among the accessions of the environments where the
forages were managed. Tillering performance also varies with production years due to
variation in distribution and amount of rainfall.

Table 3: Combined growth and yield performance of desho grass at Burie zuria and Guagusa
Shikudad districts
Bure zuria Guagusa Shikudad
Accessions N PH (cm) TN DMY (t/ha) N PH (cm) TN DMY (t/ha)
Overall 12 76.6 114.14 6.57 4 68.08 121.76 6.98
Kindo kosha-DZF#589 3 72.68 87.02b 5.94 1 77.3a 102b 8.64a
Areka DZF#590 3 76.76 122.98a 7.09 1 69.3ab 124.6a 7.47ab
Kindo kosha-DZF#591 3 80.4 129.22a 6.45 1 69ab 137.3a 6.65ab
Kulumsa-DZF#592 3 76.57 117.34a 6.81 1 56.8b 123.2ab 5.15b
CV 10.3 18.2 17.6 14.1 11 21.1
SL NS * NS * * *
* = p<0.05, Means with different letters in a column are significantly different; N = umber of
harvests, CV = coefficient of variation SL = significance level, NS = non significant at
(P>0.05); PH = plant height, TN = tiller number, DMY = dry matter yield

204
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Growth and Yield Performance of Sweet Lupin Varieties

The sweet lupin cultivar Vitabor (17.21) and probor (19.5) had the highest seed yield while
the lowest recorded by sanabor (9.76 qt/ha) at Guagusa Shigudad district. On the other hand
at Bure zuria district there was no significant difference but values ranged from 24.56 to
27.23 qt/ha (Table 4). This result indicates that higher than Alemu et al. (2017), at Gondar
zuria (3.3 - 8.1), Sirinka (9.75 - 13.12) and Sekota (3.3 - 8.1qt/ha) the same cultivars. The
variation of genotypes in performance in number of seed yield at different locations might be
attributed to variation in agro-ecological conditions such as soil associated factors, moisture
and temperature and management practices followed. The current finding disagrees with the
previous report of Likawent et al. (2012), that (44 to 50qt/ha) seed yield at the highland of
Kosober in north-western Ethiopia.

205
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 4: Growth and yield performance of sweet lupin at Burie zuria and Guagusa Shikudad districts

Bure zuria Guagusa Shikudad

Year Cultivar PH(cm) NPP NSP PL 1000 SW (g) SY qt/ha PH (cm) NPP NPS PL 1000 SW (g) SY (qt/ha)
Sanabor 80.1a 89.5a 4.9bc 4.6a 142 27.73
Vitabor 82.6a 91.5a 5.4a 4.2b 128.7 25.7
a b c
Bora 87.8 51.6 4.8 4.5ab 142 33
b a ab ab
Probor 70.3 97.8 5.2 4.4 126.7 28.12
CV (%) 6.8 24.5 3.7 4.9 8.1 29.7
2017 SL ** * ** * NS NS
Sanabor 83.7 22.2 4.8c 4.5b 148.5a 21.4 101.8a 38.4 4.7 4.83 133.9ab 9.76b
a ab b ab ab
Vitabor 83.9 30.6 5.6 4.8 128.9 28.75 89.8 31.9 5.2 4.81 132.8 17.21a
Bora 89.4 23.3 5.1bc 4.8ab 157a 17.5 102.2a 38.6 5 4.43 145.5a 14.6ab
ab a b b b
Probor 79.1 30.4 5.4 5.1 127 22.9 85.1 28.7 5 4.77 123.7 19.5a
CV (%) 7.5 25.1 4.9 7.5 4.4 28.3 9.4 31.3 8.9 7.2 6.9 21
2018 SL NS NS ** * *** NS * NS NS NS * *
***p<0.00, **p<0.01, *<0.05, means with different letters are significantly different; N = umber of harvests, CV = coefficient of variation SL =
significance level, NS = non significant at (P>0.05); PH = plant height, NPP = number of pods per plant, NSP = number of seeds per pod, PL
= pod length, SW = seed weight, SY = seed yield

206
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The highest pod set per plant (89.5, 91.5 and 97.8) was obtained on sanabor, vitabor and probor,
respectively. While Bora (51.6) the lowest recorded at Bure zuria at year one (Table 4). There
was no significant difference among cultivars pod set at year two in both districts, while
numerically lower than first year in Bure zuria district. This might be due to rainfall distribution
in both agro-ecologies. This finding disagrees with the previous report of Likawent et al. (2012),
that ranged from (12.8 to 17.9) at mid altitude and (38.2 to 42.5) high altitude pod set in north-
western Ethiopia. First year growth data were not collected due to natural (Snow) hazard.

Growth and Yield Performance of Cowpea at Bure Zuria District

At Burie zuria district dry matter herbage yield per harvest of cowpea (Vigna unguiculata) was
significantly different (P<0.05), highest dry matter yield of 4.7 and 4.1 t/ha was obtained from
genotypes ILRI-12688A and ILRI-9333A respectively, while the lowest value recorded by
blackeyebeen (1.4 t/ha) (Table 5). In the present experiment, the mean values for herbage dry
matter recorded of cowpea (3.55 t/ha) is lower than values reported for five cowpea accessions
(6.7 t/ha) (Negasu et al., 2017) and in line with reported for six cowpea accessions (2.3 to 16
t/ha) with five location Tesema (2018). The variation between Negasu et al. (2017) and present
finding is due to environmental differences. Significant differences (P<0.05) were observed
among the genotypes in seed yield kenkety and ILRI-9333A were shown highest value while
blackeyebeen the lowest. Seed yield varied between 10.7 and 24.58 quintal per ha with a mean of
20.18 quintal per ha (Table 5). This result disagrees with a report on five cowpea accessions with
mean of 25.2 (19 to 33.2 quintal per ha) by Solomon and Kibrom (2014).

Table 5: Growth and yield performance of cowpea at Burie zuria district

Cowpea accessions PH (cm) NPP NSP PL 1,000 SW (g) SY qt/ha DMY (t/ha)
Overall 81.37 8.74 9.9 11.56 134.95 20.18 3.55
ILRI-9333A 92.6 8 9.6b 11.7 122.3c 24.38a 4.1a
ILRI-12688A 80.8 10.1 9.1b 12.5 136.5b 16.56ab 4.7a
Kenkety 75.6 8.8 11.8a 10.9 123.4c 24.58a 3.13b
Blackeyebeen 73.3 8.1 8.5b 11.2 173.3a 10.7b 1.4c
CV (%) 16 36.9 11.1 8.6 4.4 36.3 16.9
SL NS NS ** NS *** * ***
***p<0.001, **p<0.01, *<0.05, Means with different letters are significantly different; qt =
quintal, CV = coefficient of variation SL=significance level, NS = non significant at (P>0.05),
PH = plant height, NPP = number of pods per plant, NSP = number of seeds per pod, PL = pod
length, SW = seed weight, SY = seed yield, DMY = dry matter yield

207
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Growth and Yield Performance of Oat Grass at Guagusa Shikudad District

At Guagusa Shikudad district mean plant height of oat grass varieties was significantly (P<0.05)
varied in combined analysis (Table 6). In combined analysis, the plant height ranged from 84.3
to 107.7 cm with a mean of 98.85 cm. Generally, CI-8251 and CI-8237 gave the highest mean
plant height (107.7 and 103.5cm), respectively followed by lampton 99.9 cm) while CI-2291
gave the lowest plant height.

Dry matter herbage yield per harvest of oat grass (Avena sativa) was significantly different
(P<0.05), highest mean dry matter yield of 3.26 at year two than year one 2.94 t/ ha was obtained
and significant difference among varieties across years. Higher value recorded by variety CI-
8251 (3.85) followed by CI-8237 (3.27) and the lowest 2.37 t/ha was lampton (Table 6). The
difference was due to genotype performance of oat (Avena sativa) variety.

Table 6: Growth and yield performance of oat grass at Guagusa Shikudad district
Year
2017 2018 Mean
PH DMY SY PH DMY SY PH DMY SY
Variety (cm) (t/ha) (qt/ha) (cm) (t/ha) (qt/ha) (cm) (t/ha) (qt/ha)
Overall 88.38 2.94 12.32 109.38 3.26 10.74 98.85 3.1 12.9
CI-8251 94.5a 3.96a 1.89 121 3.74a 10.9 107.7a 3.85a 12.5
a b
CI-8237 96 2.9 10.97 111 3.63a 10.7 103.5a 3.27ab 12.6
CI-2291 71.6b 2.4b 13.42 97 3.45 a
11.9 84.3b 2.9bc 13.3
a b b
Lampton 91.4 2.5 14 108.4 2.23 9.4 99.9ab 2.37c 13.2
CV 15 24.2 26.8 15 21 33 12.6 19.3 23
SL * ** NS NS * NS * ** NS
, **p<0.01, *<0.05, Means with different letters are significantly different; N = umber of
harvests, CV = coefficient of variation SL = significance level, NS = non significant at (P>0.05),
qt = quintal, PH = plant height, DMY = dry matter yield, SY = seed yield,

Farmers Forage Evaluation and Selection

Farmers Selection

Germination (stand count), disease resistance, pod setting, number of branches, and plant height
were criteria that farmers used to select forage legume variety and leaf smoothness, number of
tiller, leaf width and plant height to select grass variety in decreasing order of their importance.

208
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The result showed that from Napier accessions 15743 ranked first due to high number of tiller
per plant, better leaf smoothness and leaf width while 16815 was the last choice due to low
values of all the criteria set by farmers in both districts (Table 7).

Table 7: Farmers' selection of Napier accessions

Farmers Burie zuria Guagusa Shikudad
Criteria’s 16815 15743 14984 16791 16815 15743 14984 16791
LS 76 332 190 162 52.5 57 20.5 67
TN 73.5 228 132 136.5 170 246 94 290
LW 38 164 90 88 61.5 225 124.5 189
PH 40 19 76 61.5 40 130 146 84
WS 227.5 743 488 448 324 658 385 630
Rank 4 1 2 3 4 1 3 2
LS = leaf smoothness, TN = tiller number, LW = leaf width, PH = plant height, WS = weighted
Score

Based on their selection criteria farmers were preferred first Kindo kosha-DZF#591 among the
tested desho accessions due to its high plant height and leaf width in Burie zuria district while at
Guagusa Shikudad Areka DZF#590 preferred first due to high tiller number, leaf smoothness and
plant height (Table 8).

Table 8: Farmers' selection of desho varieties

District
Burie zuria Guagusa Shikudad
Kindo Kindo Kindo Kindo
Farmers Kulumsa- Areka kosha- kosha- Kulumsa- Areka kosha- kosha-
Criteria’s DZF#592 DZF#590 DZF#591 DZF#589 DZF#592 DZF#590 DZF#591 DZF#589
TN 83 86 65 89 150 162 159 132
LW 34 32.5 112 46.5 - - - -
LS - - - - 42 65.5 53.5 39
PH 118.5 75 228 148.5 63 135 84 118
Score 235.5 193.5 405 284 255 362.5 296.5 289
Rank 3 4 1 2 4 1 2 3
TN = tiller number, LW = leaf width, LS = leaf smoothness, PH = plant height

Regarding lupin varieties preferences, Bora variety was selected first by farmers due to better
germination, number of pod per plant and plant height in Burie zuria district while vitabor
preferred first in Guagusa Shikudad district due to high disease resistance, number of pod per
plant and plant height (Table 9). This difference may be agro ecological and genotype
environment interaction of varieties.

209
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 9: Farmers selection of sweet lupine varieties

Farmers Burie zuria Guagusa Shikudad
Criteria’s Probor Sanabor Bora Vitabor Probor Sanabor Bora Vitabor
Germination 87.5 255 335 195 - - - -
DR 80 228 198 194 159 60 135 246
Np 64.5 173 192 96 120 40 100 140
Nb 57 133 107 53 - - - -
PH 24.5 42.5 70 45 26 49 60 65
Score 313.5 831 902 583 305 149 295 451
Rank 4 2 1 3 2 4 3 1
DR = disease resistance, NP = number of pods, NB = number of branch, PH = plant height

The finding indicated that the kenkety variety from cowpea was selected first followed by ILRI-
12688A by farmers due to better germination, disease resistance and leafiness in Burie zuria
district. From oat variety CI-8237 preferred first in Guagusa Shikudad district due to better leaf
smoothness and plant height (Table 10).

Table 10: Farmers’ selection of cowpea and oat varieties

Burie zuria Guagusa Shikudad
Farmers ILRI- ILRI- CI- CI- CI-
Criteria’s 9333A Blackeyebeen Kenkety 12688A 8237 8251 2291 Lampton
Germination 166 124 190 280
DR 145.5 57 228 139.5
PH 69 146 93 72 136 95 40 131
Leafiness 49 16.5 71.5 34.5
LS 52.5 50.5 45.5 51.5
NS 150 60 240 150
Score 429.5 343.5 582.5 526 338.5 205.5 325.5 332.5
Rank 3 4 1 2 1 4 3 2
DR = disease resistance, PH = plant height, LS = leaf smoothness, NS = number of spikelet

CONCLUSION AND RECOMMENDATION

From this study the result growth and yield performance of forage accessions and farmers
preference selection we can conclude that among forage cultivars or varieties, 15743, Kindo
kosha-DZF#591, bora and ILRI-12688A showed better value from Napier, desho, sweet blue
lupin and cowpea forage species at Burie zuria district (mid altitude), respectively. On the other
hand at Guagusa Shikudad district (high altitude) 15743, Areka DZF#590, vitabor and CI-8237
from Napier, desho, sweet blue lupin and oat forage types showed better performance/values. So
we recommended the selected forage varieties to each agro-ecology according to their
performance for development.

210
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENTS

The authors would like to thank Agricultural Growth Program II and Amhara Agricultural
Research Institute for their financial support and development agents and participant farmers for
their valuable contribution in the completion of the study.

REFERENCES

Alemu T, Likawent Y, Belete S, Yengusie D, Desalegn A, Solomon T, Menegesha A, Shambel

k, Molla B, Getachew W, Misganaw W, Mulatu G, Bewketu A, Bekahgn W. and Baye B.
2017. Study on the Adaptability of Different Sweet Lupin (Lupinus Spp.) Varieties for Feed
Production in Different Agro-Ecologies of Amhara Region, Ethiopia, In: Mesfin Lakew and
Likawent Yeheyis (Eds.). 2018. Proceeding of the 10th Annual Regional Conference on
Livestock Completed Research Activities 13-16 March, 2017, Amhara Agricultural
Research Institute, Bahir Dar, Ethiopia.
Butt MN, Donart GB, Southward MG, Pieper RD. and Mohammad N. 1993. Effect of defoliation
on plant growth of Napier grass. Trop. Sci. 33:111-120.
CSA (Central Statistical Agency of Ethiopia). 2016. Agricultural Sample Survey Livestock and
Livestock Characteristics (Private Peasant Holdings). Central Statistical Agency of Ethiopia,
Addis Ababa, Ethiopia.
CSA (Central Statistical Agency of Ethiopia). 2019. Agricultural Sample Survey Livestock and
Livestock Characteristics (Private Peasant Holdings). Central Statistical Agency of Ethiopia,
Addis Ababa, Ethiopia.
Enoh, M.B., Kijora, C., Peters, K.J. and Yonkeu, S. 2005. Effect of stage of harvest on DM
yield, nutrient content, in vitro and in situ parameters and their relationship of native and
Brachiaria grasses in the Adamawa plateau of Cameroon. Livest. Res. Rural Dev. 17, 1e14.
Gelayenew B, Nurfeta A, Assefa G. and Asebe G. 2016. Assessment of Livestock Feed
Resources in the Farming Systems of Mixed and Shifting Cultivation, Gambella Regional
State, Southwestern Ethiopia. Global J Sci Front Res 16 (5): 11-20.
Getnet A. 2012. Retrospects and prospects of forage and pasture crop research in Ethiopia. In:
Getnet A, Mesfin D, Jean H, Getachew A, Solomon M, Alemayehu M (eds.). Forage Seed
Research and Development in Ethiopia. Ethiopia Institute of Agricultural Research, Addis
Ababa, Ethiopia.
Gezahagn k, Fekede F, Getnet A, Mengistu A, Alemayehu M, Aemiro K, Kassahun M, Solomon
M, Estifanos T, Shewangizaw W. and Mergia A. 2017. Agronomic performance, dry matter
yield stability and herbage quality of Napier grass (Pennisetum purpureum (L.) Schumach)
accessions in different agro-eco, logical zones of Ethiopia. Journal of Agricultural and Crop
Research, Vol. 5(4), pp. 49-65.

211
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Gregory N Sikumba, Walter Mangesho, Ben Lukuyu, Festo Ngulu and Mateete Bekunda. 2015.
Participatory evaluation of farmer preferences and productivity of selected Napier grass
(Pennisetum purpureum) accessions in northern Tanzania.
Likawent Yeheyis, C. Kijora, E. van Santen and K. J. Peters. 2012. Sweet Annual Lupins
(Lupinus spp.); Their Adaptability and Productivity in Different Agro-ecological Zones of
Ethiopia. Journal of Animal Science Advances 2(2):201-215.
Muinga RW, Thorpe W. and Topps JH. 1992. Voluntary food intake, live-weight change and
lactation performance of crossbred dairy cows given ad libitum Pennisetum purpureum
(napier grass var. Bana) supplemented with leucaena forage in the lowland semi-humid
tropics. Anim. Prod. 55(3):331-337.
Negasu Gamachu, Gizahagn Mengistu and Gizahu Wekgari. 2017. Adaptation Trail of
Improved Legume Cowpea, (Vigena Angulucata) at Haro Sabu, Kelem Wollega Zone,
Oromia, Ethiopia. International Journal of Research Studies in Agricultural Sciences
(IJRSAS) Volume 3, Issue 6, 2017, PP 32-36 ISSN 2454-6224. http://
dx.doi.org/10.20431/2454-6224.0306005 www.arcjournals.org.
Solomon Gebreyowhans and Kibrom Gebremeskel. 2014. Forage production potential and
nutritive value of cowpea (Vigna unguiculata) genotypes in the northern lowlands of
Ethiopia E3 Journal of Agricultural Research and Development Vol. 5(4). pp. 066-071,
Available online http://www.e3journals.org ISSN: 2276-9897 © E3 Journals 2014.
Tekalegn Yirgu, Solomon Mengistu, Edao Shanku and Fromsa Ijara. 2017. Desho Grass
(Pennisetum pedicellatum) Lines Evaluation for Herbage Yield and Quality under Irrigation
at Wondogenet, Shashemene, Ethiopia. American-Eurasian J. Agric. & Environ. Sci., 17
(5): 427-431.
Tessema Tesfaye. 2018. Evaluation of forage type cow pea (Vigna unguiculata L.Walp.)
accessions for dry matter yield in lowland of southern Ethiopia. Forage Res., 44 (2): pp. 74-
80 (2018), http://forageresearch.in.
Tessema Z, Baars R, Alemu Y. and Dawit N. 2003. Effect of plant height at cutting and fertilizer
on growth of Napier grass (Pennisetum purpureum (L.) Schumach) Tropical Sci. 43:57-6.
Wondimeneh Mekonnen, Biadegelegn Hunegnaw, Meseganaw Walle, Tekaba Eshetie,
Mekonnen Tilahun and Adebabay Kebede. 2015. Selection of Productive Napier Grass
Varieties (Pennisetum Purpureum) at Andassa Livestock Research Center, North West
Ethiopia, In: Solomon Abegaz and Likawent Yeheyis (Eds). 2016. Proceeding of the 9th
Annual Regional Conference on Livestock Completed Research Activities 9 - 20 March,
2015, Amhara Region Agricultural Research Institute, Bahir Dar, Ethiopia.

212
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Evaluation of Anthelmintic Drug Efficacy and Risk Factors for Anthelmintic Drug
Resistance in Ruminants in Selected Districts of Western Amhara Region

Yeshambel Muluye, Lissanework Mola, Belayeh Lulie, Mehari Ayalew and Yeshiwas Ferede

Andassa Livestock Research Center, P.O. Box 27, Bahir Dar, Ethiopia

ABSTRACT

Anthelmintic resistance has become a global concern for the livestock industry in many parts of the
world. Under dosing, high drug frequency and use of expired drugs are considered as major factors
for anthelmintic resistance in ruminants. This study was conducted in three administrative zones
(Awi, East Gojjam and West Gojam) of Amhara region to evaluate the efficacy of commonly used
anthelmintics against gastrointestinal nematodes of ruminants and to assess the perception of farmers
on anthelmintic utilization practice. 83 animals (34 bovine, 44 ovine and 5 caprine) were selected
based on fecal egg count (≥200 eggs per gram of feces). The animals were allocated randomly into
three groups and treated with Albendazole, Tetramisole, and Ivermectin. Faecal samples were
collected on day 0 (before treatment) and day 14 (after treatment). The modified McMaster technique
was used for quantifying the eggs. Faecal egg count reduction test (FECRT) was applied to
determine the efficacy of anthelmintic at day 14 (post-treatment). Questionnaire survey was also
conducted on 387 randomly selected livestock owners. Faecal egg count reduction (FECR %) levels
for Albendazole, Tetramisole, and Ivermectin were 99.74, 99.4 and 99.3 respectively which is an
indication for their efficacy. The questionnaire survey revealed that Albendazole was the most
commonly (61.5%) used anthelmintic followed by Tetramisole (32%) and Ivermectin (6.5%).
Effectiveness (73.1%), easy availability (25.8%), acceptable cost (0.8%) and reduced side effects
(0.3%) to their animals were measures of these drugs to be chosen by farmers. Only 10% of the
farmers request the drugs dosage estimates. Farmers, who estimate dosage by themselves use age
(15.4%), age and weight (7.7%), body size (5.2%), bold guess (53.8%) and professional’s advice
(17.9%) to administer anthelmintics to their animals. Sources of the drugs were 61.5% private
veterinary clinics, 35.9% both private and governmental clinics and 2.6% governmental veterinary
clinics. Generally this study revealed that the tested anthelmintics are effective against nematode
parasites of ruminants but there was a gap in managing drugs in some farmers; they required to be
refreshed with training.

Keywords: Anthelmintic Resistance, Anthelmintic Drug, Efficacy, Fecal Egg Count Reduction Test

213
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Amhara national regional state has the largest livestock population in Ethiopia. It is estimated at
about 15.98 million heads of cattle, 10.74 million sheep and 6.44 million goats (CSA, 2017/18).
Generally, about 33 million of the livestock population of the country is found in this region.

Compared to its potential, the economic benefits remain marginal due to poor genetic potential;
feed shortage in quantity and quality, poor husbandry practices, high prevalence of livestock
disease and low technological support. Parasitic infections remain one of the most prevalent and
important diseases. They are the most limiting factors on the production and productivity of
livestock due to delay puberty (Increased age at first service), poor feed utilization, impaired
future performance (inferior weight gains, stunted growth), treatment cost, mortality and
morbidity and increase risk of being culled and condemnation of organs and carcasses during
meat inspection (Demessie et al., 2016).

The control of helminth parasites mostly relied on the use of broad spectrum anthelmintic drugs.
Widely available broad-spectrum anthelmintic groups commonly used by Ethiopian farmers
include benzimidazole group (Albendazole, Mebendazole, Fenbendazole, Triclabendazole),
imidazothiazole group (Levamisole, Tetramisole…) and microcyclic lactone group (Doramectin,
Eprinomectin, Moxidectin, Ivermectin (Terefe et al., 2014).

Anthelmintic resistance has become a global problem in the small ruminant industry during the
last three decades, with the first cases emerging from the southern hemisphere (Waller, 1994). A
recent study performed on anthelmintic worm control practices has indicated under dosing, lack
of anthelmintic class rotation, high drench frequency and mixed grazing, which alone or in
combination, are likely to increase the risk for anthelmintic resistance (Maroto et al., 2011).
Anthelmintic resistance in ruminants has been reported as a major problem in different countries,
such as New Zealand (Waghorn et al., 2006), Brazil (Soutello et al., 2007), Ethiopia (Sissay et
al., 2007). Thus, the objectives of this study was to assess the perception of livestock owners
about their anthelmintic utilization practice, to evaluate the efficacy of the most commonly used
anthelmintics against gastrointestinal nematodes in naturally infected animals, to identify
resistant helminth parasites against most commonly used anthelmintic drugs and to set options to
combat drug resistance.

214
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIAL AND METHODS

Study Area

The study was conducted from February to October, 2019 in three administrative zones (Awi,
East Gojjam and West Gojjam) of western Amhara region. Two districts from each
administrative zone were selected for an anthelmintic drug resistance test based on; high
parasitic disease prevalence report, grazing land favorability to parasitic disease occurrence,
extensive grazing practice, high Anthelmintic drug frequency administration and unregulated
drug usage practice. Two rural kebeles for each district were selected for the study.

Awi zone: is one of the administrative zones of the Amhara region. The center of Awi zone is
Injibara which is located at 10°57'N latitude and 36°56'E longitude. The mean altitude of the
town is 2560 m.a.s.l. Awi zone has 1,231,447 cattle, 676,509 sheep and 162,576 goat populations
(CSA, 2017).

East Gojjam Zone: is one of the administrative zones of the Amhara region. The center of East
Gojjam zone is Debre markos which is located at 10°20'N latitude and 37°43'E longitude and its
mean altitude is about 2446 m.a.s.l. The livestock population of the zone is about 2,071,364
cattle, 1,403,264 sheep and 451,290 goats (CSA, 2017).

West Gojjam Zone: is one of the administrative zones of the Amhara region. The center of West
Gojjam zone is Finote Selam which is located at 10°42' latitude and 37°16'E longitude and its
mean altitude is about 1917 m.a.s.l. The livestock population of the zone is about 2,438,956
cattle, 1,242,591 sheep and 227,189 goats (CSA, 2017).

Study Population

Cattle, Sheep and goats kept under extensive traditional management systems were included in
an anthelmintic efficacy trial. Farmers engaged in livestock production were also included for the
questionnaire survey.

Study Design

Cross-sectional study design was employed to collect fecal samples and to solicit information
from farmers through questionnaire surveys.

215
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Questionnaire Survey

Questionnaire survey on 387 livestock owners was conducted in the area aimed to gather
information on Anthelmintic utilization practice of the farmers. Participants were requested for
their parasitic disease control practices (conditions for deworming, frequency of deworming,
drug administration practices, dosage estimation basis, and the sources of the drugs). Farmers
also noted their most frequently used anthelmintic and effectiveness of those drugs.

Table 1: Selected zones, districts, rural kebeles and number of interviewed Farmers in Western
Amhara region
No of interviewed Farmers
Zone Woreda Kebele Village No Total
E/Gojjam D/Eleyas Yeqegat Yeshelel 32 65
Mundibra 33
Gozamen Wonka Abuyin 28 62
Dijil 34
Awi Banja Meselachait Tahilu 34 63
Drikira 29
G/Shikudad Ashifa Lako 35 60
Shalegna 25
W/Gojjam Jabitehinan Abasem Marko 31 50
Mehal Amba 19
Bahir Dar zuria D/Mariam Tachi Dehina 87 87
Total 6 6 11 387

Fecal Sample Collection and Fecal Egg Count

Fecal sample was collected directly from the rectum of 561 animals (bovine: 466, Ovine: 88 and
Caprine: 7) for screening test. Animals that should not have been drenched with any anthelmintic in
the last 8 - 10wks were included. Selected animals were subjected to coprological examination using
standard fecal examination techniques to identify infected animals with helminths parasites. Positive
samples for parasite eggs were subjected to fecal egg count determination using

Modified Mcmaster Technique Described by (MAFF, 1977)

Bovine, Ovine and Caprine with more than 200 EPG of feces were included in the anthelmintic
resistance test. Anthelmintic drugs were administered and fecal samples were taken directly from
rectum 14 days later to check anthelmintic efficacy (Chartier et al., 1998). Anthelmintic efficacy was
detected through fecal egg count reduction test (FECR) (Coles et al., 1992).

216
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Allocation of Treatments

Among 83 animals (34 bovine, 44 ovine and 5 caprine) passed the screening test for fecal egg
count. Test animals were selected according to their pretreatment fecal egg count results (above
200 eggs per gram of faeces) and randomly allocated into three treatment groups at each site.
Three anthelmintic drugs from different groups were used for the test (albendazole, tetramisole
and ivermectin). Treatments were given based on the manufacturer recommendation and body
weight of the animal. Unfortunately fecal samples collected from G/shikudad district did not
fulfill the criteria of >200 epg and included through questionnaire survey.

Table 2: Anthelmintic treatment allocated animals in districts

Treatments given
Districts Albendazole Ivermectin Tetramisole Total
B/dar zuria 6 7 6 19
Banja 5 3 5 13
D/Elias 9 7 9 25
Gozamen 2 3 3 8
Jabi 7 4 7 18
Total 29 24 30 83

Table 3: Description of anthelmintic used for fecal egg count reduction test
Group name Generic Trade name Manufacturer Dose Route
name (mg/kg
BW)
Benzimidazoles Albendazole Ashialben 2500 mg Ashish Life science pvt limited, India 12 Per os
Bezimidazoles Albendazole Albenat 300mg Hebie Haurun pharmacy CO.,LTD, 7.5 Per os
Chin
Imidazothiazole Tetramisole Ashitetra 600 mg Ashish Life science pvt limited, India 15 Per os
Macrocyclic Ivermectin H-IVER 1% Hebie hope harmony pharmaceutical 0.2 Subcut
lactones CO.,LTD, China aneous

Data Analysis

Questionnaire Survey

All questionnaire data collected from respondents were entered into Excel spreadsheet and
analyzed using Statistical Package for Social Sciences (SPSS) Version 20 statistical software.
Descriptive statistics (percentage, frequency) were used to measure the results describing the
respondents’ responses to the questionnaire.

217
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Fecal Egg Count Reduction Test

Among 83 animals (34 bovine, 44 ovine and 5 caprine) passed the screening test for fecal egg
count. Mean FEC, percentage reduction and 95% confidence interval (CI) was determined by
using the formulae recommended by the World Association for the Advancement of Veterinary
Parasitology guidelines for detecting anthelmintic resistant nematodes (Coles et al., 1992).

The formula, FECR = (1- (MEAT/MEBT)) *100

Where, MEAT is the mean fecal egg count after treatment and MEBT is mean fecal egg count
before treatment in the host where each animal served as its own control (Cabaret and Berrag,
2004). Resistance to an anthelmintic class was considered to be present if the percentage
reduction in FEC after treatment was less than 95%, and the lower limit of the 95% confidence
interval was less than 90% (Coles et al., 1992).

RESULTS AND DISCUSSION

Albendazole, tetramisole and ivermectin were anthelmintic mostly used by 61.5%, 32% and
6.5% farmers respectively. Effectiveness (73.1%), easy availability (25.8%), acceptable cost
(0.8%) and reduced side effects (0.3%) to their animals were measures of these drugs to be
chosen by farmers. The majority of farmers (44.7%) stated that drugs were administered to the
animals mostly based on recommendations and prescriptions of animal health professionals
(Table 4).

Table 4: Bases of farmers to estimate dosage of anthelmintics in selected districts

Dosage estimation basis Frequency Percentage (%)
Veterinarian prescription 173 44.7
Self-estimation 39 10
Self-estimation and veterinarian prescription 159 41.1
Para veterinarian prescription 1 0.3
Veterinarian and Para veterinarian prescription 15 15

Farmers estimate the dosage by themselves using bold guesses (53.8%), professional’s advice
(17.9%), age (15.4%), age and weight (7.7%) and body size (5.2%) to administer anthelmintics
to their animals. Sources of the drugs were 61.5% private veterinary clinics, 35.9% both private
and governmental clinics and only 2.6% governmental veterinary clinics as replied by the
farmers.

218
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The respondents also evaluate the curability of the drugs. 61.2% of the farmers recognized these
anthelmintics as providing remedy for their sick animals. However, 38.8% of farmers
complained that they had not seen satisfactory results on the animals treated with these drugs.
This might be due to the farmers’ limitation in identifying parasitic diseases from other microbial
diseases.

Screening Test

Screening test for nematode parasite identification was done for 561 animals (cattle, sheep and
goat) in 6 selected districts of western Amhara region (Jabi, Gozamen, Banja, Debre Elias, Bahir
Dar zuria and Guagusa). From those animals subjected to coprological examination 33% was
found to be infected with strongyle (nematode) parasites.

Faecal Egg Count Reduction and Anthelmintic Efficacy

Anthelmintic efficacy was interpreted based on WAAVP recommendations for anthelmintic drug
resistance tests (Coles et al., 1992). Reduction in fecal egg counts and their confidence limits for
albendazole, tetramisole and ivermectin revealed that there was no resistance for all of the three
anthelmintics. The FECR% for albendazole, ivermectin and tetramisole in cattle and small
ruminants was found to be 99.74, 99.4 and 99.38 respectively (Table 5).

Table 5: Fecal Egg Count Reduction Test of Albendazol, Ivermectin and Tetramisole in
ruminants of selected districts
MFEC MFEC Lower Upper
Treatments (day 0) ( day 14) FECRT% 95% CI 95% CI Interpretation
Albendazole 3920.69 10.34 99.74 93.8 100 Effective
Ivermectin 3483.33 20.83 99.4 96.6 100 Effective
Tetramisole 2163.33 13.33 99.38 98.2 100 Effective

The fecal egg count reduction for albendazole in small ruminants was found to be 99.81% which
agrees with the findings of (Ahmed, 2017) 97.59% in sheep of Haramaya district, Eastern
Ethiopia and 96.6% in weken, 96.6% in Dabat zuria and 98.3% Chila of western Ethiopia
(Seyoum et al., 2017). This result is also in accordance with the reports found in studies from
Sebeta (Bahiru et al., 2017) and Bedle district of Oromia region (Terefe et al., 2014).

219
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Ivermectin and tetramisole were also found highly effective against nematode parasites of small
ruminants in the study areas with the FECR of 99.47% and 99.25% respectively. This result
supports the findings in Weken (98.3% and 98.3%) of northern Gondar (Seyoum et al., 2017),
dale district (97% and 97.5%) of Southern Ethiopia (Sheferaw et al., 2013.) and Bedele district
(96.7% and 98.5%) of Oromia region (Terefe et al., 2014). However, our result for ivermectin
was not in agreement with the finding in sheep from sebeta with FECR 97% and lower
confidence limit of 84% (<90%) was suspected for the presence of resistance (Bahiru et al.,
2017).

Table 6: Fecal Egg Count Reduction Test of Albendazol, Ivermectin and Tetramisole in sheep
and goat
MFEC MFEC Lower Upper
Treatments (day 0) (day 14) FECT% 95% CI 95% CI Interpretation
Albendazole 7106.67 13.38 99.81 95.76 100 Effective
Ivermectin 5040 26.67 99.47 97 100 Effective
Tetramisole 2800 21.05 99.25 97 100 Effective

Nematode parasites of cattle also found susceptible to all the three anthelmintic drugs with FECR
of 98.59%, 100% and 100% for albendazole, Ivermectin and tetramisole respectively. The lower
and upper confidence limit in this study for albendazole was found to be 92% and 99.8%
respectively.

The study on Bangladesh farms revealed that there was resistance in cattle treated with
albendazole with FECR 79.7%, 95% CI (65.5 - 87.9) and suspected resistance with FECR of
95.5%, 95% CI (87.1 - 98.7). Ivermectin was not also found effective on those farms with FECR
94.8% and lower confidence limit of 85.5% (Rahman et al., 2018). This might be due to frequent
drug administration and uncontrolled drug usage practice in those farms.

Table 7: Fecal Egg Count Reduction Test of Albendazol, Ivermectin and Tetramisole in cattle
FECRT Lower Upper
Treatments MFEC (day 0) MFEC (day 14) (%) 95% CI 95% CI Interpretation
Albendazole 507.14 7.14 98.59 92 99.8 Effective
Ivermectin 888.89 0 100 - - Effective
Tetramisole 1063.64 0 100 - - Effective

220
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

Most worms are controlled by the use of anthelmintic drugs. Even these drugs have an
imperative benefit uncontrolled and indiscriminate use of anthelmintics increases the risk of
development of resistant parasite populations. The pretreatment screening test showed that there
is high prevalence of nematode parasites in western Amhara region. However, based on this
study there was no anthelmintic resistance for the three anthelmintic groups namely
Albendazole, Ivermectin and Tetramisole since all the FECRT revealed >95%. This result also
showed that farmers’ limitation on estimation of dosage since some livestock owners did not
follow manufacturer’s recommendations and professional advice is an alarm for development of
anthelmintic resistance in the forthcoming years. Based on this conclusion the following
recommendations are forwarded:

• Strategic deworming should be practiced using these drugs to minimize the burden of
gastrointestinal nematodes
• Since there are reports of resistance in other areas of the country, further studies should be
conducted in different agro ecologies and sites with different drug brands and manufacturer
recommendations.
• Trainings on effective drug usage and management should be given to the farmers to prevent
the occurrence of anthelmintic drug resistance

REFERENCES

Ahmed, J. 2017. Gastrointestinal Nematode Parasites of Small Ruminants and Anthelmintics

Efficacy Test in Sheep of Haramaya District, Eastern Ethiopia. Animal and Veterinary
Sciences 5, 39. https://doi.org/10.11648/j.avs.20170503.11
Bahiru, D., Dulo, F. and Mekonnen, N. 2017. Efficacy of Commonly Used Anthelmintics
Against Gastrointestinal Nematodes in Naturally Infected Sheep in Sebeta, Central Ethiopia.
Journal of natural sciences research 7(19), 2224-3186.
Cabaret, J. and Berrag, B. 2004. Faecal egg count reduction test for assessing anthelmintic
efficacy: average versus individually based estimations. Veterinary Parasitology 121, 105–
113. https://doi.org/10.1016/j.vetpar.2004.01.020

221
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Chartier, C., Pors, I., Hubert, J., Rocheteau, D., Benoit, C. and Bernard, N. 1998. Prevalence of
anthelmintic resistant nematodes in sheep and goats in western France. Small Ruminant
Research, 29, 33-41.
Coles, G.C., Bauer, C., Borgsteede, F.H.M., Geerts, S., Klei, T.R., Taylor, M.A. and Waller, P.J.
1992. World Association for the Advancement of Veterinary Parasitology (W.A.A.V.P.)
methods for the detection of anthelmintic resistance in nematodes of veterinary importance.
Veterinary Parasitology 44, 35–44. https://doi.org/10.1016/0304-4017(92)90141-U. date
retrieved 2/5/2019.
CSA. 2017. Central Statistical Agency of the Federal Democratic Republic Of Ethiopia.
Agricultural Sample Survey . Volume II. Report on Livestock and Livestock Characteristics
(Private Peasant Holdings), Central Statistical Agency, Addis Ababa, Ethiopia.
Demessie, Y., Seyoum, Z., Getnet, K. and Yitbarek, D. 2016. Anthelmintics Resistance against
Gastrointestinal Nematodes of Sheep: A Review. World journal of agricultural sciences,
12(4): 245-253
MAFF (Ministry of Agriculture, Fisheries and Food). 1977. Manual of Veterinary Parasitological
Techniques.Technical Bulletin No. 18. Her Majesty's Stationery Office, London, 129 pp.
Maroto, R., Jiménez, A.E., Romero, J.J., Alvarez, V., De Oliveira, J.B. and Hernández, J. 2011.
First Report of Anthelmintic Resistance in Gastrointestinal Nematodes of Sheep from Costa
Rica. Veterinary Medicine International 2011, 1–4. https:// doi.org/ 10.4061/ 2011/145312
Rahman, T.M., Dey, A.R., Islam, S., Hossain, M.S., Talukder, M.H. and Alam, M.Z. 2018.
Anthelmintic resistance to cattle gastrointestinal nematodes in selected dairy farms of
Mymensingh and Sirajganj districts of Bangladesh. Research in Agriculture Livestock and
Fisheries 5, 87–92. https://doi.org/10.3329/ralf.v5i1.36556. date retrieved 22/4/2019.
Seyoum, Z., Demessie, Y., Bogale, B. and Melaku, A. 2017. Field evaluation of the efficacy of
common anthelmintics used in the control of gastrointestinal nematodes of sheep in Dabat
district, Northwest Ethiopia. Irish Veterinary Journal 70. https:// doi.org/ 10.1186/ s13620-
017-0097-6. date retrieved 25/4/2019.
Sheferaw, D., Getachew, D., Bekele, J. and Denbarga, Y. 2013. Assessment of anthelmintic
resistance in gastrointestinal nematodes of small ruminants, Dale district, Southern Ethiopia.
Journal of Veterinary Medicine and Animal Health 5: 257-261.

222
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sissay, M.M., Uggla, A. and Waller, P.J. 2007. Epidemiology and seasonal dynamics of
gastrointestinal nematode infections of sheep in a semi-arid region of eastern Ethiopia.
Veterinary Parasitology 143, 311–321. https://doi.org/10.1016/j.vetpar.2006.08.026
Soutello, R.G.V., Seno, M.C.Z. and Amarante, A.F.T. 2007. Anthelmintic resistance in cattle
nematodes in northwestern São Paulo State, Brazil. Veterinary Parasitology 148, 360–364.
https://doi.org/10.1016/j.vetpar.2007.06.023
Terefe, G., Faji, U. and Tolossa, Y. 2014. Field investigation of anthelmintic efficacy and risk
factors for anthelmintic drug resistance in sheep at Bedelle District of Oromia Region,
Ethiopia. Ethiopian Veterinary Journal 17, 37. https://doi.org/10.4314/evj.v17i2.3
Waghorn, T., Leathwick, D., Rhodes, A., Jackson, R., Pomroy, W., West, D. and Moffat, J.
2006. Prevalence of anthelmintic resistance on 62 beef cattle farms in the North Island of
New Zealand. New Zealand Veterinary Journal 54, 278–282. https:// doi.org/10.1080/
00480169.2006.36711
Waller, P.J. 1994. The development of anthelmintic resistance in ruminant livestock. Acta
Tropica 56, 233–243. https://doi.org/10.1016/0001-706X(94) 90065-5. Date retrieved
5/4/2019.

223
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Seroprevalence of Border Disease in Recently Introduced Dorper Sheep Flocks at Debre

Birhan Agricultural Research Center, Ethiopia
Chekol Demis, Tadiwos Asfaw, Eridachew Yitagesu and Enyiew Alemnew
Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia

ABSTRACT

A study was conducted to investigate the serological prevalence of Border Disease Virus (BDV)
antibodies in newly introduced Dorper sheep flocks at Debre Birhan Agricultural Research
Center (DBARC), Amhara Region, Ethiopia. Blood samples were taken from 186 sheep and the
serum was sent to the National Animal Health Diagnostic and Investigation Center in Sebeta,
Ethiopia for examination. All serum samples were examined using blocking ELISA with a
sensitivity of 94.3% and specificity of 93.7% to detect antibodies for Border Disease Virus. Of
these, 55 (29.6%) tested seropositive and the rest 131 (70.04%) tested seronegative for Border
Disease virus antibodies. The seroprevalence of sheep born in the research center was found
23.97% (29/121) whereas the sero prevalence of those sheep imported directly from abroad was
found 40% (26/65). In this research, higher prevalence of Border Disease virus antibodies was
recorded in the imported sheep than the sheep which are born in the research center and the
difference in prevalence between the two groups was found statistically significant (P<0.05).
Therefore, it is utmost important to study the status of the disease from indigenous breeds of
sheep and utilize more sensitive tests such as RT-PCR.

Keywords: Border Disease Virus, Antibodies, Debre Birhan, Dorper Sheep, Sero prevalence

INTRODUCTION

Border disease (BD) is a viral disease of sheep and goats first reported in sheep in 1959 from the
border region of England and Wales (Hughes et al., 1959). The disease is caused by a virus,
Border disease virus (BDV), which is closely related to a similar virus of cattle - bovine viral
diarrhoea virus (BVDV) and classical swine fever virus (CSFV). They are classified together as
ruminant pestiviruses which readily cross between the two species so that border disease in sheep
can be caused by either BDV or by BVDV (Nettleton and Entrican, 1995). Border Disease Virus
infections are most commonly associated with sheep but all domesticated and wild cloven-
hoofed artiodactyls are likely to be susceptible (Feknous et al., 2018).

224
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Distribution of the virus is world-wide and the prevalence varies in sheep from 5% to 50%
between countries and from region to region within countries (Nettleton et al., 1998; Krametter-
Froetscher, 2008). It has been reported throughout Europe, Australasia, North and South
America, Asia and Africa. Border disease virus spreads naturally among sheep by the oro-nasal
route and by vertical transmission (OIE, 2017). It is principally a cause of congenital disease in
sheep and goats, but can also cause acute and persistent infections (OIE, 2017). Infection is less
common in goats, in which persistent infection is rare as abortion is the main presenting sign
(OIE, 2017). Sheep can become infected with BVDV from cattle (Carlsson, 1991), and in some
countries, BVDV can be a more common cause of BD than BDV.

Oro-nasal infection in healthy adults or neonates causes mild or in-apparent disease (OIE, 2017).
The consequences of infection are primarily reproductive barren ewes, abortions, stillbirths and
stunted, weak lambs with variable degrees of nervous dysfunction. Other occasional effects
include ‘hairy’ and malpigmented wool, tremor, skeletal abnormalities, and immune suppression
with subsequent secondary bacterial infection (Nettleton et al., 1998; Pratelli et al., 1999).

Vertical transmission is important in the epidemiology of BD. The infection of fetuses in early
gestation can result in the birth of persistently infected (PI) lambs. These PI lambs are viraemic,
BDV antibody negative and constantly excrete virus. They are a potent source of infection and
ensure the spread of virus within susceptible populations (OIE, 2008). The control of BD is
hampered by a lack of vaccines (Volkan et al., 2014). Therefore, effective control has to rely on
the identification of PI animals and the prevention of infection of susceptible pregnant dams,
especially during the first half of gestation.

Due to low productivities of the indigenous breeds, the Ethiopian government had been
introducing different exotic sheep and goat breeds in the past (Tibbo, 2006: Aschalew, 2006).
Despite the efforts on genetic improvement, unexpected diseases have been introduced like
Maedi-visina (Tsegaw and Adem, 2012). Recently, incidences of new cases with undefined
etiology and characterized by typical clinical signs similar to BD that include barren ewes and
the birth of small weak lambs, abnormal body conformation, tremor fleece changes, diarrhea, ill
thrift and death from immune incompetence has been observed in introduced exotic Dorper
sheep breed nucleus flock maintained at Debre Birhan Agricultural Research Center.

225
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

According to Monies et al. (2004), enteric diseases which are characterized by diarrhoea and ill-
thrift are common problems in lambs persistently infected with Border disease virus (BDV). The
reproductive and neonatal diseases of small ruminants like border disease are far important
because they can cause huge economic disasters (Yilmaz et al., 2014). Therefore, the
establishment of diagnostic systems and control strategy after disease survey should be
performed in order to continue the expected genetic improvement program through cross
breeding at the farmer level.

Hence, this particular research activity is initiated to verify the presence and to determine the
prevalence of Border Disease Virus antibodies in the newly introduced Dorper sheep breed
maintained at Debre Birhan Agricultural Research Center.

MATERIAL AND METHODS

Study Area

The study was conducted in Debre Birhan Agricultural Research Center (DBARC). Debre
Birhan Agricultural Research Center is found in North Shewa Administrative Zone of Amhara
National Regional State, North eastern part of Ethiopia. It is located in the central part of the
Nation, at a road distance of about 120 kilometers from Addis Ababa, the capital city of the
country. Geographically, the area lies between 090o35’45” to 090o36’45” north latitude and
39o29’40” to 39o31’30” east longitude with an average elevation of about 2828 meters above sea
level. It has an average annual rainfall of about 897.8mm and mean annual temperature of about
19.9°C (Debre Birhan Agricultural Research Center, 2016).

Study Population and Sample Size

Apparently healthy, newly introduced Dorper breed sheep that are found in Debre Birhan
Agricultural Research Center, Amhara Region, in Ethiopia were considered as study animals for
this research. All the female Dorper breed sheep in the flock (186), with the age one year and
above were sampled to check the sero-prevalence of Border Disease Virus antibodies.

226
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Study Design and Data Collection

Cross-sectional study design was considered for this serological study. Blood samples were collected
in 10 ml vacutainer tubes without adding any anticoagulant, allowed to clot overnight and serum was
separated from each collected sample. After serum separation, it was kept at -20oC until laboratory
testing. The samples were sent to the National Animal Health Diagnosis and Investigation Center,
Sebeta, Ethiopia for the examination. Blocking Enzyme Linked Immunosorbent Assay (Ab ELISA)
with the sensitivity of 94.3% and specificity of 93.7% was used for the detection of serum antibodies
against Border Disease Virus.

Data Management and Analysis

Descriptive statistics of the variables was calculated and the overall sero-prevalence of Border
disease in the sampled sheep and the prevalence of the same disease in the sheep that were directly
imported from abroad and in those that were born in the research center were determined. Data
analysis was performed using SPSS software and determination of statistically significant differences
in disease prevalence between age groups was tested using Chi-square statistical method and
declared if P<0.05.

RESULTS AND DISCUSSION

One of the most important persistent viral infections affecting small ruminant populations globally is
BDV. The presence of persistently infected animals in flocks and the births of persistently infected
off-springs mean that flocks contain animals without clinical signs of infection but continuously
infect others, making it impossible to eliminate infection from the flock.

Based on the individual antibody detection using blocking ELISA, the overall sero-prevalence of
Border Disease Virus antibodies in Dorper breed sheep was recorded as 29.60% (Table 1). This
result agrees with the finding of Valdazo et al. (2008) who reported that depending on the country
and the region, sero-prevalence in sheep varies from 5% to 50%. Similarly, Krametter-Froetscher et
al. (2007) investigated sero-positivity rates of 29.4% in Australian samples using ELISA, which also
agrees with the findings of this work. However, Burgu et al. (1987) reported sero-positivity of 0.06 -
3% particularly in samples from aborted sheep and Yavru et al. (2014) reported pestivirus antigens in
8.82% of fetuses and fetal tissue samples (liver, lung and brain), both of which were by far lower
than the finding in this study.

227
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In contrary to these reports, Ameen and Karapinar (2018), Gür (2009) and Yilmaz et al. (2014)
reported sero-prevalence of pestivirus-specific and BDV antibodies of 45.2%, 78.5% and
74.57%, respectively, in sheep from different regions and countries. The differences between the
prevalence of BDV antibodies reported by this work and other research activities may be due to
variations in flock size & structure, population density, types of samples taken, the study design
and types of sampling implemented to investigate the antibodies and diagnostic methods used.

Of the186 samples tested for Border Disease Virus antibodies, 65 sheep were imported from abroad
and the rest 121 sheep were born in the research center. From the imported sheep, 40% (26/65) were
sero-positive while in sheep born in the research center; there were 23.97% (29/121) seropositive
samples. From this result, it can be observed that the prevalence is higher in imported Dorper sheep
than the sheep which were born in the research center and the chi square test showed that the
difference in prevalence of Border Disease Virus antibodies between the two groups was statistically
significant (P<0.05). The finding recorded in this study agrees with the reports of Shohreh et al.
(2014) who indicated that the number of sero-positive animals increases with the age because the
directly imported Dorper sheep were older (above 6 years) than the same breed of sheep which were
born in the research center (1 to 2 years) during the sampling year.

The tendency to higher risk among older sheep as compared to younger sheep is probably due to the
fact that Border Disease Virus antibodies in most cases are life long (Shohreh et al., 2014).
Therefore, the older the sheep, the higher is the probability to be infected during its life. According to
the result, it can be concluded that it is likely the presence of persistently infected (PI) animals within
the imported Dorper breed sheep were responsible for the presence of antibodies in the flock. Since
vaccination against BDV was not practiced in the Dorper sheep in history of the research center, the
serological response recorded in this work reflected natural infection or infection before they were
imported to Ethiopia.

Table 1: Prevalence of border disease virus antibodies in dorper sheep nucleus flock at Debre
Birhan Agricultural Research Center
Sample N N+ (Prevalence in %) N- (Prevalence in %) χ2 p-value
Center born 121 29 (23.97) 92 (76.03) 5.219 0.022
Imported 65 26 (40.00) 39 (60.00)
Total 186 55 (29.60) 131 (70.43)
N: number of animals sampled, N+: number of sero-positive animals, N-: number of sero-
negative animals

228
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

Border Disease Virus, which is common in small ruminant flocks worldwide, is important due to
the economic losses it causes. The serum antibodies of this disease were found in the newly
introduced Dorper sheep flocks at Debre Birhan Agricultural Research Center. However, this
work is a serological survey and detected the presence of antibodies only in Dorper sheep flocks
at one station using blocking ELISA test. The prevalence was higher from imported sheep than
sheep borne in the center. Therefore, further research should be conducted especially on more
sensitive tests including RT-PCR techniques to check the presence of the disease in both native
and cross breeds of sheep at regional as well as national level. Using more sensitive tests can also
provide more clear pictures on the extent of antibody or disease distribution on wider areas
which will encourage more effective preventive and control measures including proper
identification of persistently infected animals and their removal from flocks. In addition, up to
date diagnostic techniques and most effective quarantine procedures should be used and followed
to prevent the introduction of disease, especially those diseases that have not been reported
before in the country.

REFERENCES

Aschalew, T. 2006. Assessment of Feeding Systems and Evaluation of Feed Supplementation on

Body Weight and Fleece Production of Sheep in Ethiopia. PhD Dissertation, Kasetsart
University, Bangkok.
Burgu, İ., Öztürk, F. and Akca, Y. 1987. Investigations on the occurrence and impact of bovine
viral diarrhea virus infections in sheep in Turkey. Dtsch. Tierärztl.Wschr, 94, 292-294.
Carlsson U. 1991. Border disease in sheep caused by transmission of virus from cattle
persistently infected with bovine virus diarrhoea virus. Vet. Rec., 128, 145–147.
Feknou, N., Hanon, J.B., Tignon, M., Khaled, H., Bouyoucef, A. and Cay, B. 2018.
Seroprevalence of Border Disease virus and other pestiviruses in sheep in Algeria and
associated risk factors, BMC Veterinary Research, 14 (339).
Gür, S. 2009. An investigation of border disease virus in sheep in Western Turkey. Trop. Anim.
Health Prod., 41, 1409-1412.
Hughes, L.E., Kershaw, G.F. and Shaw, I.G. 1959. Border Disease.An undescribed disease of
sheep.The Veterinary Record, 71, 313-317.

229
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Krametter-Froetscher, C. Schmitz, C., Benetka, V., Bago, Z., Moest, K., Vanek, E. and
Baumgartner, W. 2008. First descriptive study of an outbreak of Border disease in a sheep
flock in Austria – a high risk factor for Bovine viral diarrhea virus free cattle herds: a case
report. VeterinarniMedicina, 53 (11): 625–628.
Krametter-Froetscher, R., Loitsch, A., Kohler, H., Schleiner, A., Schiefer, P., Moestl, K., Golja,
F. and Baumgartner, W. 2007. Serological survey for antibodies against pestiviruses among
sheep in Austria.Veterinary Record, 160, 726–730.
Mohammad Ameen, P.S. and Karapına, Z. 2018. Seroprevalence of Visna-Maedi Virus (VMV)
and Border Disease Virus (BDV) in Van province and around.Arq.Bras. Med. Vet. Zootec.,
70, 1029-1035.
Monies, R.J., Paton, D.J. and Vilcek, S. 2004. Mucosal disease-like lesions in sheep infected
with Border disease virus. Veterinary Record, 155, 765–769.
Nettleton P.F. and Entrican G. 1995. Ruminant pestiviruses, Br. Vet.J., 151, 615-642.
Nettleton, P.F., Gilray, J.A., Russo, P. and Dlissi, E. 1998. Border disease of sheep and
goats.Vet. Res., 29, 327–340.
OIE. 2008. Border disease. Chapter 2.7.1.In Manual of diagnostic tests and vaccines for
terrestrial animal health.World Organization for Animal Health (OIE).
OIE. 2017. Border disease. Chapter 2.7.1. Terrestrial Mannual.World Organization for Animal
Health (OIE).
Pratelli, A., Bollo, E., Martella, V., Guarda, F., Chiocco, D. and Buonavoglia, C. 1999.
Pestivirus infection in small ruminants: virological and histopathological findings.
Microbiologica.22, 351–356.
Shohreh, B., Hajinejad, M.R. and Yousefi, S. 2014. Prevalence of Border Disease Virus
Antibodies among Native and Imported Sheep Herds in Zabol.Trends in Life Science, 3 (2).
Tibbo, M. 2006. Productivity and Health of Indigenous Sheep Breeds and Cross- breeds in the
Central Ethiopian Highlands. PhD Dissertation, Swedish University, Uppsala.
Tsegaw, F. and Adem, Z. 2012. Serological survey of Maedi-visna virus infection in highland
sheep at ranches and small-holder farms in eastern Amhara region, Ethiopia.
Valdazo González, B., Alvarez, M. and Sandvik, T. 2008. Prevalence of border disease virus in
Spanish lambs. Vet. Microbiol., 128, 269-278.
Volkan, Y., Yakup, Y. and Nuvit, C. 2014. Molecular and serological investigation of border
disease virus infection in sheep in the Kars District of Turkey. ACTA VET. BRNO., 83:
175–179.
Yavru, S., Avci, O. and Atli, K. 2014. Koyunfötuslarında border disease virus varlığınınimmun
peroksidazilesaptanması. Eur asian J. Vet. Sci., 30, 222-226.
Yilmaz, V., Yildirim, Y. and Coskun, N. 2014. Molecular and serological investigation of border
disease virus infection in sheep in the Kars District of Turkey.Acta Vet. Brno, 83: 175–179.

230
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Seroprevalence and Associated Risk Factors of Maedi-Visna Virus in Sheep Population of

Selected Areas of Eastern Amhara

Liuel Yizengaw1*, Negus Belayneh1, Asres Zegeye1, Fasil Aklilu3, Alemu Kefale1, Enyiew
Alemnew2*, Tadiwos Asfaw2, Chekol Demis2 and Yeshitla Wondifra2
1
Sirinka Agricultural Research Centre, P.O. Box 74, Sirinka, Ethiopia
2
Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia
3
National Animal Health Diagnostic and Investigation Centre, P.O. Box 04, Sebeta, Ethiopia

All authors from Sirinka and Debre Birhan Agricultural Research Center are listed accordingly
to their order of contribution

ABSTRACT

This cross-sectional study was conducted to determine the prevalence and identify associations
with potential risk factors of Maedi-Visna virus infection in selected areas of eastern Amhara
region. A total of 2668 sheep blood serum were collected in the period from November, 2017 to
October, 2018 and examined using indirect enzyme linked immune-sorbet assay (i-ELISA) to
screen specific antibodies against Maedi-Visna virus. The sensitivity and specificity of the test
were 91.7% and 98.9%, respectively. From a total sample tested 8.2% (220/2668) were positive
for the presence of antibodies against Maedi-Visna virus in the areas. The seroprevalence of
Maedi-Visna was statistically significantly different (P<0.05) between associated risk factors of
age and production system while risk factors such as breed, sex and body condition score were
not statistically significant different (P>0.05). Semi-intensive production systems and adult
sheep were found to be important risk factors associated with higher rate of infection. Sheep
breed multiplication and improvement station and other cross breed sheep distributor farms were
incriminated as a source for Maedi-Visna virus infection and effective control measures have to
be implemented through annual or semi-annual testing and culling of all seroreactor ewes and
their progeny and annual or semi-annual screening test should be carried out during introduction
of new flocks and before distribution of Awassi and Dorper cross breed rams particularly from
ranches and research center to smallholder farms. In addition, further epidemiological study
should be done in sheep producing areas of the country to know the level of infection at regional
and national level.

Keywords: Eastern Amhara, Maedi-Visna virus; Risk factors, Seroprevalence and Sheep

231
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTIONS

Sheep population in Ethiopia is estimated at about 31.30 million, out of which, 99.81% are
indigenous breeds. Roughly 75% of the populations are found in the highlands and 25% in the
lowland (CSA, 2017/18). The sheep population in the Amhara regional state is about 11,086,083.
This contributes 35.4% to the total population of the country (CSA, 2017/18).

Small ruminant production is an important agricultural activity and has a substantial contribution
to smallholder farmers in generating income and securing food in developing countries. They
have fast growing and return rates and are able to give twins and triple births with short lambing
intervals (Kosgey et al., 2006). In Ethiopia they provide a significant value of the national meat
and skins production (Tibbo, 2006). However, due to low productivities of the indigenous
breeds, the government had been introducing exotic sheep breeds since the early 1970’s to
upgrade the genetic makeup of the local sheep breeds (Gizaw et al., 2013; Getachew et al.,
2016).

Imported sheep were stocked and crossed with the local Menz and Horro breed sheep in Debre
Berhan and Amed Guya sheep breeding and multiplication centers and crossbred rams were
selected for distribution (Getnet et al., 2010; Getachew et al., 2016). The distribution was
primarily for sheep farms established by rural kebeles in different parts of the country with the
intention that the associations could distribute to other local breeders easily and among them,
Agarfa and Arsi Rural Development Unit (ARDU), which owned the rams (Corriedale and
Awassi) in 1981 (Gizaw et al., 2013; Getachew et al., 2016) can be mentioned.

Despite the genetic improvement, an incidence of a new case with undefined etiology
characterized by a respiratory embarrassment appeared in Agarfa and Arsi Rural Development
Unit (ARDU) sheep farms in 1990`s (BoA, 2000; Tibbo et al., 2001). The disease caused 10%
mortality affecting mainly adults and had no response to antibiotic treatments. Finally, serum
samples were sent to Pirbright laboratory (UK) and specific antibodies for the Maedi-Visna virus
(MVV) or Ovine Progressive Pneumonia (OPP) virus were detected in 39 out of 43 tested
samples (MoA, 1991, unpublished). Therefore, the occurrence of MVV in Ethiopia was first
detected in imported breeds in 1986 at Agarfa sheep ranch (Ayelet et al., 2001).

232
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Maedi-Visna /OPP is a slowly progressive disease of sheep and rarely goats reported first in the
Iceland in 1939 and subsequently eradicated, has been reported in major sheep rearing countries
throughout the world except Australia and New Zealand (Bellavance et al., 1974; Pálsson et
al.,1990; Radostits et al., 2000; Straub, 2004). It has been reported from most of continental
Europe, the United Kingdom, Canada, the United States, Kenya, South Africa, Israel, India and
the southern regions of the former U.S.S.R (Jones and Hunt, 1983; Clements and Zink, 1996;
Vorster et al., 1996; Murphy et al., 1999; Radostits et al., 2000; Kahn et al., 2005).

Maedi-Visna (MV) is a chronic disease of adult sheep characterized by progressive interstitial

pneumonia and other syndromes such as meningoencephalitis, indurative mastitis and arthritis
(Lujan et al., 1994; Radostitis et al., 2007). It is caused by a non-oncogenic retrovirus, which
belongs to the subfamily lentivirinae (Radostitis et al., 2007). Transmission occurs more readily
between dams and lambs via colostrum and milk, and among confined individuals probably via
respiratory secretions (Preziuso et al., 2009).

Since 1988-1989 the detection of the virus in Ethiopia, it has been assumed that Maedi-Visna is
an emerging disease introduced to the country through the imported sheep breeds (BoA, 2000;
Ayelet et al., 2001). Previous reports from the assessment of the disease in and around the
stocking and rearing centers of North Shewa showed that the disease became one of the most
important diseases of the respiratory system of sheep in central Ethiopia. The infection is
persistent, so antibody detection is a valuable tool for identifying virus carriers (Ayelet et al.,
2001; Tibbo et al., 2001; Woldemeskel et al., 2002; Getnet et al., 2011).

In the country an outbreak of unidentified diseases often occurs and a considerable number of
sheep die with signs of respiratory embarrassment. When MVV is introduced into a new area,
the mortality rate may reach 20 - 30% (Ayelet et al., 2001). The mortality rate is low in regions
where MV is endemic; annual losses rarely exceed 5% in a flock, even when nearly 100% of the
flock is infected (Peterhans et al., 2004). The economic losses of the disease are due to mortality
associated with clinical disease, poor value of the removed animals and reduction of economic
life. Effects of subclinical infection on the reproductive potential should also be added to the
economic losses (Cutlip et al., 1982).

233
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

There is no treatment and effective vaccine against the disease but by increasing the quality and
efficiency of diagnostic tests, there is the possibility of eradicating the disease (Shuaib et al.,
2010). So far in Ethiopia, outbreaks of unidentified diseases often occur and a considerable
number of sheep die with signs of respiratory embarrassment (Ayelet et al., 2001). Although
farms and breeding centers have been reporting Maedi-Visna cases in Eastern Amhara, Ethiopia
(Nigussie and Belay, 2016), the extent to which the disease disseminated has been established,
but there was lack of information on the status and losses associated with Maedi Visna and very
little attention has been given to the role of Maedi-Visna virus as the cause of disease and
production losses in sheep in Ethiopia. Therefore, taking into account the significance of the
disease as one of the most important causes of economic losses and the scarcity of information in
the country, the present study was designed to estimate the seroprevalence status and point out
the associated risk factors of Maedi-visna infection in eastern Amhara region.

MATERIALS AND METHODS

Study Areas

Three administrative zones (South Wollo, North Wollo, and North Shewa) were selected
purposely based on the history of Awassi cross bred sheep population. From each zone
representative districts were again selected purposely based on Awassi cross bred sheep
distribution, their location, proximity and accessibility to road. Usually Awassi cross bred sheep
were distributed around the highland areas of eastern Amhara which practiced the crop-livestock
mixed production system.

Debre Birhan Agricultural Research Center, Faji, and Debre Birhan Sheep Multiplication and
Breed Improvement Center are located around Debre Birhan town at a distance of 110 - 130 Kms
North of Addis Ababa, between 9o30’26’’ to 9o64’92’’N latitude and 39°14’32’’ to 39°27’37’’E
longitude. The study areas are found in the central highland of the country at an altitude of above
2770 m. The annual rainfall of the study areas ranges from 950 - 1200mm. The mean annual
minimum and maximum temperatures are 1.5 and 23.3°C, respectively and the area experiences
a bimodal rainfall patterns with a short rainy season which occurs from January to March and
long rainy season which starts at the end of June and ends at early November (Tefera and
Mulate, 2016).

234
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Legambo district: is situated about 500 Km North of Addis Ababa at a Latitude, 11°00'0"N and
Longitude, 39°00'0"E and an elevation of range from 1500 to 3700 m.a.s.l. The sheep populations of
the area were 120, 993 (Woreda livestock agency report, 2018). The annual rainfall of the area ranges
from 950 - 1200mm. The mean annual minimum and maximum temperatures are 1.5 °C and 23.3°C,
respectively and the area experiences a bimodal rainfall patterns with a short rainy season which
occurs from January to March and long rainy season which starts at the end of June and ends at early
November (Tefera and Mulate, 2016).

Gazobelay district: (Wadila former) is found about 600 Km of Addis Ababa at a latitude, 11°50′N
and longitude, 38°50′E and an elevation of ranges from 700 to 3200 meters above sea level. The
sheep populations of the area were 143,133 (Woreda livestock agency report, 2018). The rainfall
pattern is bimodal, with two-rainfall season, belg (February/March - April) and meher (July -
October/November) and the mean annual rainfall amount is on average about 950 mm. This area has
three seasons of sub-humid agro-climatic zone with mean daily temperature ranges from 16 - 21°C.
Months of June to August are main rainy seasons; months of September to February are dry seasons,
while months of March to May are short rainy season (Zegeye et al., 2014).

Mehal Meda District: Amed guya /Awassi cross sheep multiplication center/ is situated in the
Mehal Meda town of North Shoa zone of the Amhara region. It has a geographical coordinate of a
latitude, 10°18'0"N and longitude, 39°40'0"E with an altitude of 3132 meter above sea level. It is
located 180 Km north of Debre Birhan town and 361Km northeast of Addis Ababa. The average
temperature is 12.2°C. The average annual rainfall of the study area is 1149 mm and the area
experiences bimodal rainfall patterns with a short rainy season which occurs in winter and long rainy
season in summer (Gebreyohannes et al, 2013).

Study Animals and their Management

The animals used for this study were short fat tailed indigenous breeds (Solomon et al., 2011),
Awassi cross, Dorper cross and pure Awassi sheep breeds. Sheep above 6 months of age were
sampled. The age of sheep was classified into three; young, adult and old based on dentations (Geoff,
2016). The body condition of sheep was classified into three; good, medium and poor based on
fatness (Jodie, 2010).

235
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In this study the production system was classified into two based on management systems of sheep
owners in the study areas. In an extensive system, sheep were spent all the day on grazing pasture, on
fallow lands and crop residues usually with no extra-supplement and sheltered during the night. This
management system was basically practiced in smallholder sheep producers. In semi-intensive
production system owners supplemented extra feed sources in addition to grazing. This production
system was practiced by the sheep multiplication stations (Stephen, 2017).

Study Design and Sampling Strategy

Cross-sectional sero-epidemiological study was conducted from November, 2017 - October,

2018. A multistage stratified cluster sampling method was used in smallholder farmers of Gazo,
Legambo and Faji districts. Finally, study units were sampled using a simple random sampling
method both in districts of smallholder farmers and Debre birhan agricultural research center
sheep multiplication station. In both Amedguya and Debre Birhan sheep multiplication stations
all sheep flock (census) samples were sampled (Martin et al., 1988 and Thrusfield, 2007). At
smallholder farmer levels three districts; Faji from North Shoa zone, Legambo from South Wollo
and Gazobelay from North Wollo zone were selected purposely based on accessibility and the
history of Awassi and high population of sheep. From each district a representative sample
animal was selected.

Sample Size Determination

Since there was a previous study conducted in the study areas with a prevalence rate of 4% by
Nigussie and Belay (2016), so the desire sample size for this study was determined based on the
previous prevalence 4%, the 5% desired absolute procession and 95% confidence interval (CI))
according to Thrusfield (2005).

N = (1.96)2Pexp (1 - Pexp)
d2

Where: n is the required sample size, Pexp is the expected prevalence, d is the desired absolute
precision and 1.962 is the value of z at 95% Confidence level.

Accordingly, 59 sheep samples were required. But to increase the precision and accuracy, the
sample size was maximized to more than 13 fold (817) from smallholder farmers.

236
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A total 1851 blood samples were collected from sheep multiplication centers. From these 280
from local, 2166 from Awassi cross, 131 from Dorper cross and 91 from pure Awassi breed were
collected. The number of animals sampled from each smallholder farmer in each study district
was proportionally distributed based on sheep population. Overall 2668 serum samples were
tested, analyzed and interpreted.

Blood Sample Collection

Blood samples were collected directly from the jugular vein of each sheep. Sterile vacutainer
tubes and needles were used for each animal and about 5 ml of blood was taken. Each sample
from each animal was labeled by using codes describing the specific animal. The tubes were kept
overnight at a room temperature to allow clotting. Next morning the clotted bloods in the tubes
were centrifuged to obtain clear serum. Then serum were separated from the blood and collected
into 1.8 ml cryo-vial and were preserved at -20ºC in a deep refrigerator until they were processed
and analyzed in the National Animal Health Diagnostic and Investigations Center, Sebeta,
Ethiopia (Nigussie and Belay 2016).

Serological Examination and Analysis

The serum samples were tested for the presence of specific antibody against Maedi-visna virus
using Indirect Enzyme-linked immunosorbent assay test (I-ELISA), Maedi-Visna/CAEV serum
verification version VISNAS ver 1217 EN (IDvet, 310, Rue Louis Pasteur - Grabels - France).
The I-ELISA kit allows a joint detection of the antibodies directed against both MVV and
CAEV. The test was performed according to the manufacturer`s manual. The results of the test
were considered valid only if optical density of a positive control serum (ODPC) was higher than
0.350 and ODPC was more than three times higher than optical density of a negative control
serum (ODNC). The optical density of a serum sample (OD sample) was recalculated into
percentage of ODPC (S/P %) adjusted by ODNC with the formula: S/P% = (OD sample -
ODNC) / (ODPC - ODNC) × 100%. The interpretations of samples presenting and S/P %, equal
or below 50% are considered as negative, between 50% and 60% are considered as doubtful and
equal or above 60% are considered as positive. At manufacturer cut-off of 50% sensitivity and
specificity were 91.7% and 98.9% respectively (Nowicka et al., 2014).

237
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Management and Analysis

All data collected for this study were entered in MS-Excel spreadsheet, arranged and analyzed
using STATA (version 14) software. Descriptive statistics were used to estimate the
seroprevalence of Maedi visna virus antibodies in the areas. Binary logistic regression was used
to identify significantly important risk factors such as breed, age, sex, body condition score and
production system, for the occurrence of Maedi visna virus. Tests were considered statistically
significant when P<0.05.

RESULTS

From the total samples tested, 8.2% (220/2668) were positive for the presence of antibodies
against Maedi-Visna virus (MVV) in the area. The seroprevalence was higher in station/ ranch
(10.3%) than smallholder farmers (3.5%). The Maedi-Visna seroprevalence distribution had
varied among the districts and production system. There was a statistical significance difference
in sero-positivity among sheep of different production systems (P<0.05, OR = 10.2) with
prevalence of positive case was higher in semi intensive (station/ranch, 10.3%) than extensive
(smallholder farmers, 3.5%) (Table 2). The highest and the lowest sero-prevalence were in Debre
Birhan on station (63.5%) and Faji districts (0.8%), respectively. The seroprevalence of Maedi-
Visna infection in each production system and districts is summarized in Table 1.

Table 1: Seroprevalence of Maedi-Visna virus in each production system and districts of eastern
Amhara region
Production
systems Zone Districts No of samples No of positives Prevalence (%)
Small holder South Wollo Legambo 419 16 3.8
North Wollo Gazobelay 273 12 4.4
North Shoa Faji 125 1 0.8
Total 817 29 3.5
Station or North Shoa On station 52 33 63.5
ranches Amedguya 501 90 18
Debre Birhan sheep 1298 68 5.2
multiplication center
Total 1851 191 10.3
Overall 2668 220 8.2

238
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

During the statistical analyses of all risk factors, the first level of each independent variable was
used as a reference category. The univariable logistic regression odd ratio model analysis of
attribute risk factors indicated that there was statistical significance difference in sero-positivity
among sheep of different age group (P<0.05, OR = 5.5) 2 - 4 years (adult) are more affected, and
different production system (P<0.05, OR = 10.2) with prevalence of positive case was higher in
semi intensive station /ranch (10.3%) than extensive smallholder farmers (3.5%). There was no
significant difference in the prevalence of MV infection between the male sheep (7.6%) and
female sheep (8.3%) (P>0.05). Similarly, no significant difference in sero-prevalence among
different sheep breeds (P>0.05) and body condition score (P>0.05) (Table 2).

Table 2: Univariable logistic regression analyses (LR) of risk factors with Maedi-Visna sero-
positivity in sheep in eastern Amhara region
No. of
animal No. of Prevalence
Risk factors sampled positive (% ) P-value OR CI (95%)
Breed Local (Indigenous)* 280 23 8.2 - -
Awassi cross 2166 187 8.6 0.81 1.06 0.7 - 1.7
Pure Awassi 91 3 3.3 0.12 0.38 0.1 - 1.3
Dorper cross 131 7 5.3 0.30 0.63 0.3 - 1.5
Sex Female* 2314 193 8.3 -
Male 354 27 7.6 0.65 0.91 0.6 - 1.4
Age < 2 years (young)* 553 11 2 - - -
2 - 4 years (adult) 2038 206 10.1 0.000 5.54 2.9- 10.2
> 4 years (old) 77 3 3.9 0.29 1.99 0.5 - 7.3
Body Poor* 109 6 5.8 - - -
condition Moderate 2218 201 9.1 0.25 1.6 0.7 - 3.8
score Good 346 13 3.8 0.38 0.6 0.2 - 1.7
Production Extensive* 817 29 3.5
system Semi- intensive 1851 191 10.3 0.00 3 2 - 4.4
Total 2668 220 8.2
* Reference category

DISCUSSIONS

Seroprevalence of Maedi-Visna (MV) study has been conducted at different times from various
countries including Ethiopia. The result of the present study showed an overall seroprevalence of
Maedi-Visna in sheep to be 8.2%. This seroprevalence study indicated the occurrence and wide
distribution of Maedi Visna (MV) virus infection in sheep across production systems in selected
areas of eastern Amhara region.

239
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The 8.2% prevalence of MV in this study was comparable with the reports of Getnet et al. (2002)
6% in north Omo, Ethiopia, but moderately higher than Tefera and Mulate (2016) 3.2% in
Eastern Amhara, Ethiopia, Mahin et al. (1984) 2.7% in Morocco, Sihvonen et al. (2000) 1.6% in
Finland, Shuaib et al. (2010) 2.41% in Manitoba, Canada and Aslantas et al. (2002) 1.5 - 2.6% in
Hatay region, Turkey. However, the seroprevalence result of the present study was much lower
than many of the previous reports in Ethiopia, which was 74% in central Ethiopia (Woldemeskel
et al., 2002); 62.5% in central cool highland (Garedew et al., 2010); 70.4% in Sheno agricultural
research center (Seyoum et al., 2011); 15.6% in eastern Amhara region (Tsegaw and Ademe,
2012); and 88% and 20% in Debre Birhan sheep breeding center and Arsi, respectively (Getnet
et al., 2010).

This sero-prevalence result was also much lower than in other countries of the world. A
prevalence of 15.3% was reported in Turkish sheep (Preziuso et al., 2010), 15.6% in culled ewes
in Alberta, Canada (Fournier et al., 2006), 29.6% in Khorasan-Razavi province, Iran (Norouzi et
al., 2015), 28.8% in Germany (Hüttner et al., 2010), 50% in Palestine (Hananeh and Barhoom,
2009), 19% in Canada (Simard and Morley, 1991), 19.4% in Kirikkale district, Turkey (Azkur et
al., 2011) and 18% in Wyoming sheep, USA (Gerstner et al., 2015). Such inconsistency in the
prevalence rates of Maedi Visna (MV) might be due to the variation in the diagnostic tests,
sampling method used, the prevalence variability within the population studied, the
characteristics of the animals forming the population, susceptibility of different breeds to the
disease, management practices and measures taken to control the disease.

This study showed a variation in seroprevalence of MV between different study districts (0.8%
to 63.5%). Similar results were obtained in different parts of Quebec (14.5% to 69%) (Shuaib et
al., 2010); in Turkey (3.8% to 41.2%) (Alkan and Tan, 1998); in Iran (6.7% to 72.2%) (Norouzi
et al., 2015); and indifferent parts of Ethiopia (0.6% to 88%) (Getnet et. al., 2010), (2.4 % -
6.8%), (Tefera and Mulate, 2016) and (2.2% - 87.4%) (Tsegaw and Ademe, 2012). The spatial
difference in distribution of positive cases might be explained by the introduction of carrier
animals from an infected area to disease free zones, the management / production practices and
the bio-security followed by farm owners.

240
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The seroprevalence found in Faji (0.8%), Legambo (3.8%) and Gazobelay (4.4%) districts was
amazing. These districts are geographically located far from severely affected sheep station/
ranches and it is suggested that the disease might have been spread along with the distribution of
cross breed rams as other study investigated sero reactor rams in the villages obtained from
Amedguya and Debre Birhan station/ ranches a year ago.

There was no statistically significant difference in seroprevalence between male and female
sheep (P>0.05), which was in agreement with findings of Woldemeskel et al. (2002), Seyoum et
al. (2011) and Tefera and Mulate (2016). This sex sero-prevalence similarity can probably be
both sexes had equal chance of exposure and susceptibility to Meadi Visna virus infection. Also,
the possible explanation for this similarity could be due to the fact that animals rearing
conditions in the study areas were similar and both sexes were herded together without
separation and in direct contact with each other.

The breed related seroprevalence of Maedi-Visna virus in present study showed no statistical
significant difference among breeds (P>0.05) which was in line with Tefera and Mulate, (2016).
The possible explanation for this similarity could be due to the fact that animals rearing
condition in the study areas was similar and different breeds were herded together without
separation and sheep of different breeds were in direct contact with each other. In contrast to this
finding, susceptibility difference was reported between Menz and Awassi sheep breeds by
Seyoum et al. (2011), Tsegaw and Ademe (2012). Snowder et al. (1990) and Schaller et al.
(2000) have also reported a prevalence difference among breeds. This breed susceptibility
difference could be related to the influence of traits of particular family lines, the strain of the
virus and the result of one or more recessive genes (Simard and Morley, 1991).

There was no significant difference in seroprevalence among body condition scores (P>0.05).
This finding disagreed with the report of Pritchard and Dawson (2000), Ayelet et al. (2001),
Getnet et al. (2010), Tsegaw and Ademe (2012) and Tefera and Mulate (2016). The possible
explanation for this similarity could be due to the fact that animals rearing conditions in the study
areas were similar and different body condition score animals were herded together without
separation and in direct contact with each other.

241
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A significant seroprevalence difference (P<0.00) among age groups was observed. Adult and
older animals were 5.3 and 2 (95% CI = 3- 10.2, X2 = 51.3) times more likely to be infected as
compared to younger sheep, respectively. This result was in line with the result reported by other
studies in Canada (Arsenault et. al., 2003; Simard and Morley, 1991), in Ethiopia (Ayelet et al.,
2001; Tsegaw and Ademe, 2012; Tefera and Mulate, 2016), in Turkey (Preziuso et al., 2010) and
in Iran (Norouzi et al., 2015). This age sero-prevalence discrepancy can probably be explained
by the longer exposure to horizontal transmission and development of detectable levels of MV
antibodies can vary from months to years (Radostits et al., 2000). Thus, the older the animals
was the greater the potential for a greater proportion of sheep to become infected with MV. This
was consistent with the works of Cutlip et al. (1988) and Snowder et al. (1990) who reported 4%
and 11% at one year of age and 34% and 93% at above 4 years of age, respectively. The
incubation period of this subclinical infection is usually more than two years and its clinical
signs appear when the animal is 3 to 4 years old (OIE, 2012).

There was significant variation (P<0.00) between the production systems with semi-intensive
(10.3%) was 3 (95% CI = 2 - 4.4) times more likely infected than the extensive (3.5%) system.
This variation in husbandry and production system is in agreement with previous reports. Ayelet
et al. (2001) has reported lower prevalence (3.7%) of MV in village flocks and relatively higher
(7%) in sheep on-station at Debre-birhan, north Shewa. Woldemeskel et al. (2002) also reported
higher seroprevalence (74%) of MV in clinically morbid sheep at ranches and Tsegaw and
Ademe (2012) higher seroprevalence (30%) in ranches higher than in smallholder farmer (6.6%).
The seroprevalence difference between the semi intensive and extensive production system was
probably associated with the husbandry and management practices. The large flock size,
confining animals for longer hours during cold seasons, mixing different breeds and age groups,
keeping a high proportion of older animals in ranches have been incriminated to contribute to the
disease occurrence and transmission in higher rates. Baumgartner et al. (1990) have suggested
unfavorable housing conditions such as insufficient room, bad climatic conditions and crowding
behavior in sheep promote a high incidence of the disease.

242
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSIONS AND RECOMMENDATIONS

Even though the sero prevalence of Maedi-Visna (MV) in the present study was low; it is widely
distributed in sheep producing areas of eastern Amhara region. The seroprevalence was
significant P<0.05 in the semi-intensive production systems (sheep breeding and multiplication
ranches) than the extensive system and it was also higher in the adult and older sheep than
younger ones. The finding of positive serological reactors does not only suggest the occurrence
of the disease in sheep population of the study area, but also indicates the presence of foci of
infection that could serve as source of infection for the spread of disease into unaffected animals
around the study region and elsewhere in the sheep producing areas which receive cross breed
sheep for upgrading of local sheep and also through marketing. Sheep breeding and
multiplication stations were incriminated as a home-base for Maedi-Visna virus infection and
effective control measures should be implemented, through screening tests during introduction of
new flocks and before distribution to smallholder farms. Annual or semi-annual testing and
culling of all seroreactor ewes and their progeny should be employed. In addition, further
epidemiological study using more sensitive tests should be done in sheep producing areas of the
country to know the level of infection at regional and national level.

ACKNOWLEDGMENTS

We would like to thank all the livestock research directorate staff of Sirinka and Debre Birhan
Agricultural Research Center and all staff of microbiology and virology from National Animal
Health Diagnostic and Investigation Center for laboratory analysis; and special thanks goes to
Solomon Gebre and Yeshawork Gashaw. We would like to extend our sincere thanks to the head
and coordinator Debre Birhan Sheep Multiplication and Breed Improvement Center (DBSMBIC)
and Amed Guya Sheep Multiplication and Breed Improvement Center (AGSMBIC) and to all
their staff members for their willingness to help us in our research by giving the necessary
information (data). Finally, our acknowledgment goes to Amhara Agricultural Research Institute
(ARARI) for funding of this activity.

243
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Alkan F. and Tan M. 1998. Comparative study on the diagnosis of Maedi-Visna infection in
serum and colostrums samples using Agar gel immunodiffusion (AGID) technique.
Dtsch.Tierarztl.Wochenschr, 105:276-278
Arsenault J., Girard C., Dubreuil P., Daignault D., Galarneau J., Boisclair J., Simard C. and
Elanger D. 2003. Prevalence of and carcass condemnation from Maedi-visna,
paratuberculosis and Caseous lymphadenitis in culled sheep from Quebec, Canada.
Preventive Veterinary Medicine, 59: 67–81.
Aslantas O., Pinar D. and Gungor B. 2002. Serologic investigation of Maedi-Visna infection in
Hatay region. Veteriner Hekimler Mikrobiyoloji Dernegi., 2:31-34.
Ayelet G., Roger F., Tibbo M. and Tembely S. 2001. Survey of Maedi-Visna (MV) in Ethiopia
highland sheep. Veterinary.Journal, 161:208-210.
Azkur A., Gazyagci S. and Aslan E. 2011. Serological and Epidemiological Investigation of
Bluetongue, Maedi-Visna and Caprine Arthritis-Encephalitis Viruses in Small Ruminant in
Kirikkale District in Turkey. Kafkas Universitesi. Veteriner Fakultesi. Dergisi, 17 (5):803-
808.
Baumgartner W., Reckinger M., Pernthaner A. and Leitold B. 1990. The occurrence and
distribution of Maedi-Visna virus infection in lower Australian sheep breeding
establishments. Dtsch. Tierar. Wochen. 97: 465-469.
Bellavance R.D., Turgeon J., Phaneuf B. and Sauvageau R. 1974. Pneumonie interstitielle et
progressive du mouton. Can Vet J 1974; 15:293–297
BoA (Bureau of Agriculture). 2000. Report on the preliminary investigation of sheep disease at
sheep ranches and ram distribution sites in North Shoa. Disease Investigation Report. Bahir
Dar, Ethiopia.
Central Statistical Authority. 2017/18. Central Statistical Agency Agricultural Sample Survey,
Federal Democratic Republic of Ethiopia, Central Statistical Authority, volume II report on
livestock and livestock characteristics.
Clements J. E. and Zink M. C. 1996. Molecular biology and pathogenesis of animal lentivirus
infections. Clinical Microbiology Review, 9: 100-117.
Cutlip R., Lehmkuhl H., Schmerr M. and Brogden K. 1988. Ovine progressive pneumonia
(maedi-visna) in sheep. Veterinary Microbiology, 17: 237-250.
Fournier D., Cambell J. and Middleton D. 2006. Prevalence of Maedi-Visna infection in culled
ewes in Alberta. Canadian Veterinary Journal, 47: 460-466.
Garedew G., Ayelet G., Yilma R., Zeleke A. and Gelaye E. 2010. Isolation of diverse bacterial
species associated with maedi-visna infection of sheep in Ethiopia. African Journal of
Microbiology Research, 14 (1): 14-21.
Gebreyohannes M., Demek Y. and Kebede E. 2013. Ovine fasciolosis: prevalence and associated
risk factors in Menz Gera Midr woreda of north shoa, Ethiopia. J. Anim.Prod. Adv., 2013,
3(6): 203-207

244
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Geoff, 2016. How to tell the age of sheep. Primefact 1481, second edition. NSW Department of
primary industries, July 2016. ISSN 1832-6668.
Gerstner S., Adamovicz J., Duncan J., Laegreid W., Marshall L., Logan J. and Schumaker, B.
2015. Prevalence of and risk factors associated with ovine progressive pneumonia in
Wyoming sheep flocks. Journal of the American Veterinary Medical Association, 247(8):
932-937.
Getnet A., Asegedech S. and Hassen C. 2010. Seroepidemiological study on Maedi-Visna in
selected areas of Ethiopia. Ethiopian Veterinary Journal, 14 (1): 101-111.
Gizaw S., Abegaz S., Rischkowsky B., Haile A., Okeyo A.M. and Dessie T. 2013. Review of
sheep research and development projects in Ethiopia. Nairobi, Kenya: International
Livestock Research Institute (ILRI).
Hananeh W. and Barhoom S. 2009. Outbreak of MaediVisna in Sheep and Goats in Palestine.
World Applied Sciences Journal, 7(1):19-23.
Herrmann L. M., Cheevers W. P. and Marshall K. L. 2003. Detection of serum antibodies to
ovine progressive pneumonia virus in sheep using a caprine arthritis-encephalitis virus
competitive-inhibition enzyme-linked immunosorbent assay. Clin Diag Lab Immunol,
10:862–865.
Hüttner K., Seelmann M. and Feldhusen F. 2010. Prevalence and risk factors for Maedi-Visna in
sheep farms in Mecklenburg-Western-Pomerania. Berliner und Münchener tierärztliche
Wochenschriftärztl, 123:10–14
Jodie penninggton, 2010. Body condition scoring of sheep and goats. AG NEWS & VIEWS,
417-455.
Jones C. and Hunt R. 1983. Veterinary Pathology, 3rd Edition, Lea and Febiger, Philadelphea,
466-468.
Kahn C., Line S. and Aiello S. 2005. Merck Veterinary Manual, National Publishing Inc.,
Philadelphia, 1233-1235.
Kosgey I.S., Baker R.L., Udo H.M.J. and Van Arendonk J.A.M. 2006. Successes and failures of
small ruminant breeding programmes in the tropics: a review. Small Rumin. Res. 61:13-28.
Lujan L., Begara I., Collie, D. and Watt N. 1994. Ovine lentivirus (maedi-visna virus) protein
expression in sheep alveolar macrophages. Veterinary pathology, 31(6): 695-703.
Mahin, Chadli and Houwers. 1984. A preliminary report on the occurrence of Maedi-visna in
sheep in Morocco. Veterinary quarterly, 6(2):104.
Martin S. W., Meek A. H. and Welleberg P. 1988. “Veterinary epidemiology/ principles and
Methods,” Veterinary epidemiology/ principles and Methods, pp. 30-31, 1988. View at:
Google Scholar
Ministry of Agriculture (MoA). 1991. The occurrence of MVV in imported breed in 1986 at
Agarfa sheep ranch (Bale province) Ethiopia. (unpublished)
Murphy F. A., Gubbs E. P., Horzinek M. C. and Studdert M. J. 1999. Veterinary Virology. 3rd
Edition, Academic publisher, San Diago, 175-385.

245
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Nigussie T. and Belay M. 2016. Seroprevalence of Maedi -Visna in sheep in selected districts of
Amhara region, Ethiopia. bull. anim. hlth. prod. afr., (2016), 64, 423-43
Norouzi B., Razavizade A., Azizzadeh M., Mayameei A. and Mashhadiv S. 2015. Serological
study of small ruminant lentiviruses in sheep population of Khorasan-e-Razawi province in
Iran. Veterinary Research Forum, 6(3):245-249.
Nowicka D., Czopowicz M., Mickiewicz M., Szalus-Jordanow O., Witkowski L., Bagnicka E.
and Kaba J. 2014. Diagnostic performance of ID screen MVV-CAEV Indirect screening
ELISA in identifying small ruminant lentivirus- infected goats. journal of veterinary science
vol.17, no.3 (2014)501-506.
Office International des Epizooties (OIE). 2008. Caprine Arthritis- Encephalitis and Maedi-
Visna. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Office
International des Epizooties (OIE), http://www.oie.int/eng/normes/mmanual/a_summry.htm.
Office International des Epizooties (OIE). Maedi-Visna. Available at: www.oie.int/doc/
ged/d1226.pdf. Accessed Sep 06, 2012.
Peterhans E, Greenland T, Badiola J, Harkiss G, Bertoni and G, Amorena. 2004. Routes of
transmission and consequences of small ruminant lentiviruses infection and eradication
schemes. Veterinary Research, 35(3): 257-274.
Pálsson PA., Pétursson G. and Hoff-Jørgensen R. (editors).1990. History and clinical description
of Maedi-visna and Related Diseases. Boston: Kluwer Academic Publishers; 1990. p. 3-17.
Preziuso S., Erman O., Gimmarioli M., Kyar A., Feliziani F., Gonul R., Farneti S., Yaramis C.,
Valente C. and Cuteri V. 2010. Maedi-Visna virus in Turkish sheep: a preliminary
serological survey using ELISA test. Turkish Journal of Veterinary and Animal. Sciences,
34(3):289-293.
Preziuso S., Sanna E., Sanna M., Loddo C., Cerri D., Taccini E., Mariotti F., Braca G. and Rossi
G. 2009. Association of Maedi Visna virus with Brucella ovis infection in rams. European
Journal of Histochemistry, 47(2), 151-158. https://doi.org/10.4081/821.
Radostitis, O.M., D.C. Blood and C.C. Gay. 2007. Ovine progressive pneumonia (Maedi-
Visna).In: Veterinary Medicine. A Textbook of Cattle, Sheep, Goats, Pigs and Horses, 9 ed.
London, Bailliereth, Tindall, pp: 1071-1075.
Radostits O. M., Gay C. C., Blood D. C. and Hinchcliff K. W. 2000. Veterinary Medicine. A
textbook of diseases of cattle, sheep, pigs, goats and horses. 9th ed., Harcourt publishers
limited, London Philadelphia, 701-967.
Schaller P., Vogt H., Strasser M., Nettleton P., Peterhans E. and Zanoni R. 2000. Seroprevalence
of Maedi-visna and border disease in Switzerland. Schw. Arch. Teirh. 42: 145-153.
Seyoum Z., Molalegne B., Mekonen T. and Esayas G. 2011. Evaluation of control program of
Maedi- Visna by foster feeding with cow colostrums and other measures. Global Veteriaria,
6(1): 96-96.
Shuaib M., Green C., Rashid M., Duizer G. and Whiting T. 2010. Herd risk factors associated
with seroprevalence of Maedi-Visna in the Manitoba sheep population. Canadian Veterinary
Journal, 51:385-390.

246
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sihvonen L., Nuotio L., Rikula U., Hirvelä-Koski V. and Kokkonen U. 2000. Preventing the
spread of maedi–visna in sheep through a voluntary control programme in Finland.
Preventive veterinary medicine, 47(3): 213-220.
Simard C. and Morley R. 1991. Seroprevalence of Maedi-Visna in Canadian Sheep. Canadian.
Journal of Veterinary Research, 55: 269-273.
Snowder G., Gates N., Glimp H. and Gorham J. 1990. Prevalence and effect of subclinical ovine
progressive pneumonia virus infection on ewe wool and lamb production. Journal of
American Veterinary Medical Association, 197: 475-479.
Solomon Gizaw, Komen, H., Hanote, O., van Arendonk, J.A.M., Kemp, S., Aynalem Haile,
Mwai, O. and Tadelle Dessie. 2011. Characterization and conservation of indigenous sheep
genetic resources: A practical framework for developing countries. ILRI Research Report
No. 27. Nairobi, Kenya, ILRI.
Straub O.C. 2004. Maedi-Visna virus infection in sheep. History and present knowledge.
Comparative Immunology, Microbiology and Infectious Disease, 27(1):1-5).
T. Getachew, A. Haile, M. Wurzinger, B. Rischkowsky, S. Gizaw, A. Abebe, and J. Sölkner
2016. Review of sheep crossbreeding based on exotic sires and among indigenous breeds in
the tropics: An Ethiopian perspective. Vol.11(11), pp. 901-911, March 2016.
https://doi.org/10.5897/AJAR2013.10626
Tefera N. and Mulate B. 2016. Seroprevalence of Maedi-Visna in sheep in selected districts of
Amhara region, Ethiopia. Animal Health and Production, 64: 423-430.
Thrusfield M. 2007. Sampling In: Veterinary Epidemiology, 3rd ed London: Black Well Science
Ltd 179-284.
Tibbo M. 2006. Productivity and health of indigenous sheep breeds and crossbreds in the central
Ethiopia highlands. PhD dissertation. Department of Animal Breeding and Genetics, Faculty
of Veterinary Medicine and Animal Sciences, Swedish University of Agricultural Science
(SLU), Uppsala, Sweden.
Tibbo, M., Woldemeskel M. and Gopilo A. 2001. An outbreak of respiratory disease complex in
sheep in central Ethiopia. Trop. Animal Health Prod., 33: 355-365.
Tsegaw F. and Ademe Z. 2012. Serological survey of Maedi-visna virus infection in highland
sheep at raches and smallholder farms in eastern Amhara region, Ethiopia. Bulletin of
Animal Health and Production in Africa, 60(3):287-295.
Vorster J. H., Dungu B. B., Marias, L. C., York D. F., William R. R. and Boshoff C. H. 1996. A
Perspective on Maedi/Visna in South Africa, J. S. Afr. Vet. Assoc., 67: 2-3.
Woldemeskel M., Tibbo M. and Potgieter L. 2002. Ovine progressive pneumonia (Maedi-
Visna):An emerging respiratory disease of sheep in Ethiopia. Deutsche tierärztliche
Wochenschrift, 109:486-488.
Zegeye Asres, Belayineh Nigus, Shinkute, Abiy, Lakew Mesfin and Tesfaye Abebe. 2014. On-
Station Retrospective Epidemiologic Study of Small Ruminant Disease in Sirinka
Agricultural Research Center, Sheep and Goat Breeding, Evaluation and Distribution Site.
Academic Journal of Animal Diseases 3(3): 39-44, 2014.

247
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The Effect of Strategic Deworming Programs on the Prevalence of Gastrointestinal

Strongyles and Fasciola in Sheep at Selected Districts of North Shewa Zone, Ethiopia

Chekol Demis, Tadiwos Asfaw, Erdachew Yitagesu and Enyiew Alemnew

Debre Birhan Agricultural Research Center, P.O. Box 112, Debre Birhan, Ethiopia

ABSTRACT

On farm and on station study was conducted from September 2014 to August 2018 in Menz Gera
and Menz Mama Districts of North Shewa Zone and in Debre Birhan Agricultural Research
Center (DBARC), Amhara Region, Ethiopia to evaluate the effect of strategic deworming
programs on the prevalence of major gastrointestinal strongyle nematode and Fasciola eggs in
naturally infected sheep. Sheep were dewormed four times per year in September, December,
March and June using locally available anthelmintic drugs that include Albendazole,
Tetraclozan, Ivermectin and Triclabendazole according to the manufacturers’ recommendations.
A total of 6728 fecal samples were collected and examined both before and after deworming
using flotation technique for nematode eggs and sedimentation technique for Fasciola eggs.
Before deworming, the on farm prevalence of nematode and Fasciola eggs was 38.8% and
43.3%, respectively and the on station prevalence of the same parasite eggs was 29.2% and
33.8%, respectively. At the end of the study, after four successive years of strategic deworming
programs, the on farm prevalence of nematode and Fasciola eggs was reduced to 1.1% and
23.2% and the on station prevalence of these parasitic eggs was reduced to 2.6% and 19.0%,
respectively, and the difference was found statistically significant (P<0.05). Similarly, significant
(P<0.05) reduction in the on farm and on station prevalence of nematode and Fasiola eggs was
recorded after the application of strategic deworming programs. The coprological investigation
in this study revealed that, irrespective of the production system and various risk factors,
application of the strategic deworming programs can significantly reduce the parasite load in
sheep. However, more tight and integrated parasite control programs should be implemented for
better success in reduction of parasitic burden in sheep flocks of highland areas with further
studies on species identification.

Keywords: Deworming, Fasciola, Nematode, Strongyle, North Shewa, Prevalence, Sheep

248
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Small ruminants play a very important role in human nutrition and they have the potential of
serving as tools for poverty reduction through provision of meat, milk, household income,
manure and skin. However, the productivity per animal is still low due to multitude of factors
(Okorafor et al., 2015). Parasites are among these factors which are responsible for a number of
economic losses in a variety of ways. Parasites cause losses through lower fertility, a reduction in
feed intake, lower weight gains, lower milk and meat production, treatment costs and mortality
in heavily parasitized animals (Carmichael, 1972; Akerejola et al., 1979; Nahed-Toral, 2003).
Gastrointestinal parasites are worldwide problems but their impact is greater in Sub–Saharan
Africa including Ethiopia due to the availability of a wide range of agro- ecological factors
suitable for diversified host and parasite species (Fikru et al., 2006).

Recently, in Amhara National Regional State, there are encouraging works on sheep breed
improvement for mutton purposes, both through community based breeding programs at the
farmer level and at the research centers. Despite the breed improvement programs of the region,
several sheep diseases, particularly those diseases which are caused by internal parasites are the
commonest problems in the areas. Among these parasitic diseases, the ones that are caused by
gastrointestinal nematodes and Fasciola are the primary challenges of the study area.

Hence, four times per year (September, December, March and June) deworming strategic
programs are implemented in sheep flocks of the study areas as part of health monitoring and
surveillance works that will help breed improvement activities. The strategic deworming is
applied using broad spectrum anthelmintic drugs like albendazole, tetraclozan, tetramizole and
ivermectin as well as narrow spectrum drugs mainly triclabendazole following the manufacturer
recommended doses.

According to El-Bahy et al. (2009), strategic parasite control programs are usually designed
based on identification of major types, incidence and seasonality of parasites, rainfall and
temperature of an area as well as availability of anthelmintic drugs in the target region. The same
authors indicated that such treatment of parasites at critical times of the year induce enormous
effects on their eradication, minimize the parasitic stress on the animal and improve its general
health condition and productivity.

249
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Several strategic deworming programs are applied in different countries all over the world. These
programs include two times treatment during the year (Maingi et al., 2002; Magona et al., 2004),
three times treatment program per year (Guimaraes et al., 2000; Magona et al., 2004), four times
treatment program per year and even monthly treatment (Bukunzi and Serumaga-Zke, 2000).
However, selection of the effective time for massive chemotherapy prophylaxis is different from
one country to another depending on the rain fall, atmospheric temperature and seasonality of
parasites (El-Bahy et al., 2009).

Therefore, the objective of this study was to evaluate the effect of available broad and narrow
spectrum anthelmintic drugs in the reduction of the prevalence of nematode and Fasciola eggs
when they are applied in a strategic approach in sheep under the community based breeding
programs at the farmer level and at Debre Birhan Agricultural Research Center.

MATERIALS AND METHODS

Study Areas

The study was conducted from September 2014 to August 2018 in Menz Gera and Menz Mama
Districts of North Shewa zone and in the station of Debre Birhan Agricultural Research Center
(DBARC), Amhara Region, Ethiopia. All of the study areas are found in the central highlands of
the country at road distances of 120 to 360 kilometers from Addis Ababa, Ethiopia.
Geographically, the areas lie between 09°35’45” to 10°18’0”N latitude and 39°29’40” to
39°40’0’’E longitude with average elevation ranges of 2800 to 3200 meters above sea level and
with mean annual temperature ranges of 12.2 to 19.9°C. The average annual rainfall of the areas
ranges from 897.8 to 1149 mm and it is characterized by bimodal pattern with cold, harsh
climate that occasionally has frost, particularly between November and January. There are four
climatic seasons which include the Kiremt (summer or meher), Belg (autumn), Bega (winter) and
Tsedey (spring). Kiremt ranges from June to August, Belg ranges from September to November,
Bega ranges from December to February and Tsedeey ranges from March to May (DBARC,
2014: Menz Gera and Menz Mama Livestock Offices, 2014 and www.Ethiopian treasures. co.uk/
pages /climate.htm).

250
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Study Animals and their Management

Indigenous breeds (Menz, Bonga and Washera breeds), cross breeds (Dorper and Awassi breeds
crossed with indigenous breeds) and exotic breeds (Dorper and Awassi breeds) were included in
the study. The study animals were kept under traditional extensive management systems in the
areas of community based breeding programs at the farmers’ level and in semi-intensive
management systems in the research center. The sheep in the community based breeding
programs at the farmers level depend mainly on grazing natural pasture in private and communal
grazing lands while sheep in the research center provided harvested hay and commercial
concentrate feed (maize, oil sees cake, wheat bran and salt) in addition to the morning and
afternoon pasture grazing.

Moreover, these animals were treated using anthelmintic drugs that include albendazole,
tetraclozan, tetramizole, ivermectin and triclabendazole. The drugs were applied in four rounds
per year in September, December, March and June following manufacturers’ recommendations
and deworming months were selected based on the epidemiological cycle of targeted parasite
groups and the laboratory findings. They are also vaccinated against major infectious diseases
which include pasteurellosis, sheep and goat pox and pestidespetitis ruminants.

Study Design and Sample Size

This study is a continuation of the first longitudinal monitoring study conducted from 2011 to
2014 before full implementation of strategic deworming programs. In that study, the baseline
(before deworming) prevalence of major gastrointestinal nematode and Fasciola parasite eggs
was recorded in sheep from the same study areas and sample size. Then, implementation of
strategic deworming programs were started in September 2014 and continued up to August 2018
to evaluate the impact of strategic deworming programs on the prevalence of major
gastrointestinal nematode and trematode parasites. These activities were conducted through
longitudinal follow up based on laboratory analysis of monthly collected fecal samples for the
presence of parasite eggs. Study districts and villages were selected purposely based on the
presence of breed improvement works, while sheep were selected through stratified random
sampling method.

251
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sheep were stratified before sampling by their breed, age and sex and then animals were sampled
based on the repeatable Fecal Egg Count (FEC) monitoring guideline of Abbott et al. (2012).
According to this guideline, from sheep grazing together in the same field, at least 10 sheep in
the strata of each group should randomly be sampled. For this particular study however, about
15% (proportional sampling) of animals were sampled from each breed, age and sexusing
systematic random sampling method. Accordingly, a total of about 6728 (1848 from the on farm
and 4880 from the on station) sheep were sampled and followed for four successive years from
September, 2014 to August, 2018.

Fecal Sample Collection and Examination

Fecal samples were collected in the first week of every month directly from the rectum of sheep
under study and these samples were put in universal sampling bottles containing 10%
chlorhexidine preservative and all the necessary information was labeled. The collected samples
were transported to the parasitology laboratory of Debre Birhan Agricultural Research Center
and examined for the presence of nematode and Fasciola eggs using simple test tube floatation
and sedimentation techniques as described by Charles (2006).

Data Management and Analyses

The data collected from the community based sheep breed improvement villages and the
research center as well as from the laboratory investigation were coded with appropriate
variables and analyzed using Statistical Package for Social Science (SPSS) statistical software
version 20. Descriptive statistics was used to summarize the data and the prevalence was
calculated by dividing the number of positive animals by the total number of animals examined
and multiplied by 100. Percentage was used to measure prevalence and Chi Squared (X2) test
was used to measure the association between the prevalence of the parasite eggs before and after
the application of strategic deworming programs and the risk factors including age, sex, breed
and season. In all analyses, confidence level was held at 95% and P<0.05 was set for statistical
significance.

252
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

The changes in prevalence of Nematode and Fasciola eggs in the on-farm and on-station sheep
before and after the application of strategic deworming programs were demonstrated in Tables 1
and 2, respectively. According to the result of this study, the initial (before deworming) on-farm
and on-station prevalence of nematode eggs was 38.8% and 29.2%, respectively, with an overall
prevalence of 31.8%. After four years of intervention using strategic deworming programs, the
on-farm prevalence of these eggs reduced to 1.1% and the on-station prevalence of the same eggs
reduced to 2.6% with an overall prevalence of 2.3%. The respective difference in the on-farm
and on-station prevalence of Nematode eggs reached to 37% and 26.6% with an overall
difference of 29.5% after application of strategic deworming programs.

In addition, the initial on-farm and on-station prevalence of Fasciola eggs was 43.3% and 33.8%,
respectively, with an overall prevalence of 36.6%. However, the on-farm prevalence of these
eggs reduced to 23.2% and the on-station prevalence reduced to 19.0% after the deworming
programs with an overall prevalence of 20.1%. In a similar trend to nematodes, the on-farm and
on-station prevalence of Fasciola eggs also showed respective differences of about 20.1% and
14.8% with an overall difference of 16.1%. The statistical analysis of this result indicated that the
difference in the on-farm and on-station prevalence of both nematode and Fasciola eggs before
and after the strategic deworming programs was found statistically significant (P<0.05) (Table 1
and 2).

Table 1: On-farm and on station prevalence of nematode eggs of sheep in selected districts of
North Shewa Zone
N+ (Prevalence in %)
Risk Factors N Before After X2 P-value
Production System
On farm 1848 717(38.8) 20 (1.1) 66.218 0.000
On station 4880 1425(29.2) 127 (2.6)
Total 6728 2142 (31.8) 147 (2.3)

253
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: On-farm and on station prevalence of Fasciola eggs of sheep in selected districts of
North Shewa Zone
N+ (Prevalence in %)
Risk Factors N Before After X2 P-value
Production System
On-farm 1848 801(43.3) 428 (23.2) 103.190 0.000
On-station 4880 1649(33.8) 927 (19.0)
Total 6728 2450 (36.4) 1355 (20.1)

Concerning the risk factors, the difference in the on-farm and on-station prevalence of both
nematode and Fasciola eggs before and the application of strategic deworming programs was
found statistically significant (P<0.05) for age, sex, age and climatic seasons.

The statistically significant (P<0.05) reduction in the on-farm prevalence of nematode and
Fasciola eggs is demonstrated in Tables 3 and 4, respectively.

Table 3: On-farm prevalence and risk factors of nematode infection of Menz sheep in selected
districts of North Shewa Zone
N + (Prevalence in %)
Risk Factors N Before After X2 P-value
Age
<1 year 414 202 (48.8) 6 (1.4) 23.923 0.000
≥1 year 1434 515 (35.9) 14 (1.0)
Sex
Male 482 218 (45.2) 7(1.6) 12.787 0.002
Female 1366 499 (36.5) 13 (1.0)
Season
Sept – Nov 464 192 (41.4) 7 (1.5)
Dec – Feb 457 166 (36.3) 4 (0.9) 20.469 0.002
Mar – May 472 153 (32.4) 4 (0.8)
June – Aug 455 206 (45.3) 5 (1.1)
Total 1848 717 (38.8) 20 (1.1)

254
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 4: On-farm prevalence and risk factors of Fasciola infection of Menz sheep in selected
districts of North Shewa Zone
N + (Prevalence in %)
Risk Factors N Before After X2 P-value
Age
<1 year 414 169 (40.8) 72 (17.4) 19.401 0.000
≥1 year 1434 632 (44.1) 356 (24.8)
Sex
Male 482 179 (37.1) 94 (19.5) 28.511 0.000
Female 1366 622 (45.5) 334 (24.6)
Season
Sept – Nov 464 204 (44.0) 103 (22.2)
Dec – Feb 457 181 (39.6) 96 (21.0) 23.290 0.001
Mar – May 472 190 (40.3) 115 (24.4)
June – Aug 455 226 (49.7) 114 (25.1)
Total 1848 801 (43.3) 428 (23.2)

The results in Tables 5 and 6 demonstrated a significant (P<0.05) reduction in the on-station
prevalence of nematode and Fasciola eggs, respectively, after application of strategic deworming
programs.

Table 5: On-station prevalence and risk factors of nematode infection of sheep in selected
districts of North Shewa Zone
N + (Prevalence in %)
Risk Factors N Before After X2 P-value
Age
<1 year 716 237 (33.1) 28 (3.9) 13.225 0.001
≥1 year 4164 1188 (28.5) 99 (2.4)
Sex
Male 1150 293 (25.6) 31 (2.7) 10.096 0.006
Female 3730 1132 (30.3) 96 (2.6)
Breed
Dorper 1277 456 (35.7) 59 (4.6)
Dorper x Local 933 275 (29.5) 13 (1.4)
Menz 2045 493 (24.1) 40 (2.0)
Bonga 139 44 (31.7) 4 (2.9) 92.2596 0.000
Washera 196 63 (32.1) 4 (2.0)
Awassie 50 18 (36.0) 0 (0.0)
Awassie x Local 240 76 (31.7) 7 (2.9)
Season
Sept – Nov 1224 382 (31.2) 56 (4.6) 66.801 0.000
Dec – Feb 1232 291 (23.6) 19 (1.5)
Mar – May 1208 340 (28.1) 21 (1.7)
June – Aug 1216 412 (33.9) 31 (2.5)
Total 4880 1425 (29.2) 127 (2.6)

255
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 6: On-station prevalence and risk factors of Fasciola infection of sheep in selected districts
of North Shewa Zone
N + (Prevalence in %)
Risk Factors N Before After X2 P-value
Age
<1 year 716 272 (38.0) 153(21.4) 14.536 0.001
≥1 year 4164 1377 (33.1) 774 (18.6)
Sex
Male 1150 371 (32.3) 167 (14.5) 28.386 0.000
Female 3730 1278 (34.3) 760 (20.4)
Breed
Dorper 1277 521 (40.8) 271 (21.2)
Dorper x Local 933 297 (31.8) 173 (18.5)
Menz 2045 594 (29.0) 358 (17.5)
Bonga 139 57 (41.04) 29 (20.9) 95.371 0.000
Washera 196 82 (41.8) 42 (21.4)
Awassie 50 20 (40.0) 10 (20.0)
Awassie x Local 240 78 (32.5) 44 (18.3)
Season
Sept – Nov 1224 437 (35.7) 244 (19.9)
Dec – Feb 1232 375 (30.4) 222 (18.0) 22.563 0.001
Mar – May 1208 387 (32.0) 227 (18.8)
June – Aug 1216 450 (37.0) 234 (19.2)
Total 4880 1649 (33.8) 927 (19.0)

In the current study, a highly significant (P<0.05) reduction in the on-farm and on-station prevalence
of both nematode and Fasciola eggs was recorded for all risk factors considered after the application
of four times per year strategic deworming programs. This result is in agreement with the reports of
EL-Bahy et al. (2009) who indicated absolute freedom of the animals from their previous parasitic
infection after four times per year treatment regime. The current result is also in line with previous
reports by Dalton (1999) who indicated that the prevalence of liver flukes in sheep was reduced
significantly from 75 to 1% after five anthelmintic treatments per annum during a 3 year period.
Similarly, Francisco and Gonzalez (2009) indicated a marked decrease in the prevalence of Fasciola
hepatica parasites from 63.16 to 13.64% over a two years intervention period. The study conducted
in cattle by Jakob et al. (2000) indicated that the overall prevalence of Trichostrongyle faecal eggs
was 32% in the treated animals and 47% in the untreated control animals. A large-scale, randomized
intervention field study conducted by Zinsstag et al. (1997) also showed that two annual
fenbendazole treatments increase live weights of one to four years old animals between 8 and 17%.
This was also in agreement with the works of Bukunis and Serumagas-Zake (2000) which indicated
that repeated treatment of animals lead to gradual decrease in the number of infective stages of
parasites shed from treated animals.

256
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A study conducted by Maes (1993) showed reduction in the prevalence of liver fluke infections
from 93 to 5% even when the annual treatments were reduced from four times per year to three
times per year in the second year of the strategic parasite control campaign. The findings of this
study are also in agreement with the works of Maichomo et al. (2004) and Magona et al. (2004)
that indicated improvements in general health status of animals as a result of strategic application
of anthelmintic drugs. As Jakob et al. (2000), two annual strategic treatments during the peak of
gastrointestinal nematode excretion reduce contamination with Trichostrongyle eggs by 31 -
57%. Maingi et al. (2006) also reported that strategic anthelmintic treatment leads to
improvements in some physiological parameters of treated animals in the form of high weight
gain and Packed Cell Volume (PCV).

According to ILCA (1990), reduction in prevalence of parasites through strategic regular

deworming programs is much more important in highland areas than the midland and lowland
areas found elsewhere in the country because gastrointestinal parasites contribute for about 50%
of all on-farms sheep morbidities in the Ethiopian highlands.

However, the fluke prevalence of this study is higher than the findings of Njau et al. (1990) who
reported a Fasciola infection rate of 10% after regular deworming programs at the International
Livestock Centre for Africa research station at Debre Birhan. These relative differences in
prevalence of parasites might arise due to the existence of different climatic and environmental
factors (ecological diversity and climatic variation in temperature and rainfall) that could support
survival and development of the infective larval stage of parasites.

In addition, multitude of other factors including over stocking, poor nutrition (starvation), poor
housing and hygiene, the sample size considered, types of techniques utilized, frequent exposure
to the contaminated communal grazing lands and resistance of parasites to the frequently used
anthelmintic drugs could also be responsible for the variation in the prevalence of parasite eggs
recorded in this and other studies.

257
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

Strategic application of available broad and narrow spectrum anthelmintic drugs in sheep
according to the epidemiological cycle of the parasites results in significant reduction in
prevalence of both nematode and Fasciola eggs. In the current study, highly significant reduction
in the on farm and on station prevalence of both nematode and Fasciola eggs was recorded
irrespective of the risk factors considered (age, sex, breed and climatic season). Thus, four times
per year application of anthelmintic treatments contributed to the reduction of parasitic burden in
sheep. Therefore, community based strategic deworming programs and periodic monitoring of
the prevalence of parasite eggs in sheep should be expanded to other areas of similar conditions.
However, for better results in the area of small ruminant health monitoring activities, such
programs have to be integrated with the management practices of animals and continuous
training and technical support to the smallholder farmers.

REFERENCES

Abbott, K.A., Taylor, M. and Stubbings, L.A. 2012. Sustainable Control of Parasites in Sheep.A
Technical Manual for Veterinary Surgeons and Advisers, 4th Edition, United Kingdom.
Akerejola, O. O. Schillhorn van Veen, T. W. and Njoku, C.O. 1979. Ovine and Caprine diseases
in Nigeria: a review of economic losses. Bull.Anim. Hlth. Prod. Afr., 27: 65-70.
Bakunzi, F.R. and Serumaga-Zake, P.A.E. 2000. The effect of strategic anthelmintic treatment
on internal parasites in communally grazed sheep in a semi-arid area as reflected in the
faecal nematode egg count. Trop. Anim. Health Prod., 32: 295-302.
Carmichael, I. H. 1972. Helminthiasis in domestic and wild ruminants in Botswana: Preliminary
investigations. Trop. Anim. Hlth. Prod., 4: 175-181.
Charles, M. H. 2006. Diagnostic Veterinary Parasitology.3rd ed. Elsevier Science.
Dalton JP. 1999. Fasciolosis.Cab International (ed.) Wallingford, Oxon, UK.
El-Bahy, M.M., El-Bahy, N.M. and El-Shehawy, S.A. 2009. A Model for Strategic Control of
Sheep Parasites at Qassim Region, Saudi Arabia. Research Journal of Parasitology, 4:33-44.
Fikru, R., Teshale, S., Reta, D. and Yosef, K. 2006. Epidemiology of gastrointestinal parasite of
ruminants in Western Oromia Ethiopia.International Journal of Applied Research in
Veterinary Medicine, 4: 51-57.
Francisco, R. and Sergio, G. 2009. Strategic Control and Prevalence of Fasciola hepatica in
Cajamarca, Peru.A pilot Study.Intern J Appl Res Vet Med., 7, ( 4): 145-152.
Guimaraes, M.P., Ribeiro, M.F., Facuri-Filho,E.J. and Lima, W.S. 2000. Strategic control of
gastrointestinal nematodes in dairy calves in florestal, minas gerais, Brazil.Vet. Res.
Commumu., 24: 31-38.

258
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ILCA. 1990. Annual Report and Programme Highlights.International Livestock Centre for
Africa, Addis Ababa, Ethiopia.
Jakob, Z., Philippe, A., Mamud, N., Pradeep, I., Vincent, M., Johannes, K., Tom, S., Vijay, S.P.
and Kurt, P. 2000. Effect of strategic gastrointestinal nematode control on faecal egg count
in traditional West African cattle.Vet. Res., 31:259–266.
Maes L Veys, P., Geerts, H. and Chiarsoli, O. 1993. Field trials with closantel within an
integrated control strategy against faciolosis in sheep.Rev Med Vet.; 144: 781-6.
Magona, J.W., Musisi, G., Walubengo, J. and Olaho-Mukani, W. 2004. Effect of strategic
deworming of village cattle in Uganda with Moxidectin pour-on on faecal egg count and
pasture larval counts. J. S. Afr. Vet. Assoc., 75: 189-192.
Maichomo, M.W., Kagira, J.M. and Walker, T. 2004. The point prevalence of gastro-intestinal
parasites in calves, sheep and goats in magadi division, south-western Kenya. Onderstepoort
J. Vet. Res., 71: 257-261.
Maingi, N., Munyua, W.K. and Gichigi, M.N. 2006. Strategic use of moxidectin or closantel in
combination with levamisole in the control of nematodes of sheep in the highlands of central
Kenya. Acta. Trop., 84: 93-100.
Maingi, N., Otieno, R.O., Gichohi, V.M. and Weda, E.H. 2002. Strategic control of
gastrointestinal nematodes of sheep in the highlands of central Kenya. Onderstepoort J. Vet.
Res., 69: 229-235.
Nahed-Toral, J., Lopez-Tirado, Q., Mendoza-Martinz, G., Aluja-Schunemann, A. and Trigo-
Tavera, F.J. 2003. Epidemiology of parasitosis in the Tzotzil sheep production system.
Small Ruminant Research; 49 (2):199-206.
Njau, B.C., Scholtens, R.G. and Kassali, O., 1990. Parasites of sheep at the international
Livestock Centre for Africa DebreBerhan station, Ethiopia. Preventive Veterinary Medicine,
9 (4): 267-277.
Okorafor, U. P., Obebe, O. O., Unigwe, C. R., Atoyebi, T.J. and Ogunlaye, O. K. 2015. Studies
on the gut parasites of small ruminants reared in some selected farms in Ido Local
government area of Oyo state Nigeria. Applied research journal, 1 (3):153-159.
Zinsstag J., Ankers P., Dempfle L., NjieM.,Kaufmann J., Itty P., Pfister K. and Pandey V.S.
1997. Effect of strategic gastrointestinal nematode control on growth of N’Dama cattle in
Gambia, Vet. Parasitol, 68:143-153.

259
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Retrospective Analysis of Clinical Cases of Caprine and Ovine in Sirinka Agricultural

Research Centre, Amhara Region

Liuel Yizengaw, Negus Belayneh, Asres Zegeye, Zeleke Tesema, Almu Kefale, Shamble Kiross
and Mekonnen Shibesh
Sirinka Agricultural Research Centre, P.O. Box 74, Woldia, Ethiopia

ABSTRACT

A retrospective epidemiological study of sheep and goats diseases was conducted at Sirinka
agricultural research center, small ruminant breeding, evaluation, and distribution site, from
available data recorded from 2013 to 2018 to determine the temporal distribution of diseases and
associated risk factors of clinical cases. From the total 664 clinical cases, gastrointestinal
parasitism (24.55%), septicemia (16.11%), and pneumonia (10.99%) were the most frequently
occurred cases while, keratoconjunctivitis (1.51%), external parasitism (2.26%) and mastitis
(2.41%) were the least. The prevalence of cases was higher in crossbred goat (39%) and
crossbred sheep (36.14% than local sheep (11.14%) and local goat (10.24%). Female goats
(34.49%) and female sheep (34.34%) were highly susceptible to diseases. The highest clinical
cases were observed in young (50.75%), then adult (43.67%), and old animals (5.57%). The
young animals were more susceptible to pneumonia, septicemia, unpredicted cases, and external
parasitism while adult animals were highly susceptible to reproductive diseases, abscess,
mastitis, and wound. The highest clinical cases were observed in the month of September,
August, and January. There were higher cases in the dry season (56.63%) followed by the main
rainy season (29.22%) and the short rainy season (14.16%). The highest prevalence was recorded
in the year 2017 (31.5%), 2018 (21.5%), and 2016 (20.9%). Generally, clinical cases of
Gastrointestinal parasitism, septicemia, pneumonia, and skin disease were the most important
health problems in the farm, and from these all cases considerable numbers of sheep and goats
were recovered (73.64%) and (26.36%). were dead. Therefore, appropriate measures and
intervention should be given for these disease conditions, and strictly follow-up, monitoring, and
reporting of the case should be continued strongly. Finally, an investigation of unknown cases
should be done.

Keywords: Caprine, Clinical Cases, Ovine, Prevalence, Retrospective, Sirinka and Risk factors

260
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

The diverse agro-climatic conditions of Ethiopia make it very suitable for the production of different
kinds of livestock (Belachew and Jemberu, 2002). Ethiopia is believed to have the largest livestock
population in Africa. The estimated livestock population of Ethiopia is 60.39 million heads of cattle,
31.30 million heads of sheep, 32.74 million heads of goats, 2.01 million horses, 8.85 million
donkeys, 0.46 million mules, 1.42 million camels, 56.06 million poultry and 6.52 million hives
(CSA, 2017/18). Sheep and goat are known to serve as a source of animal protein, skin, manure,
income, and some cultural values (Tefera et al., 2009). This economic advantage is primarily
associated with their ease of handling as it favors small scale investment, minimum risk of loss, and
high reproductive efficiency (Omoike, 2006).

Despite a large number of livestock populations in the Amhara region, the sector is characterized by
low productivity and, hence, income derived from this sector of agriculture could not impart a
significant role in the development of the region’s economy. Factors affecting livestock production in
sub-Saharan countries include diseases, poor management, and lack of proper breeding policies
(Adebowale, 2012). These problems have also been observed as major constraints to livestock
production in Ethiopia. Among these constraints, diseases have numerous influences on the
production and reproductive performance of sheep and goats. The effect of livestock diseases could
be expressed in terms of losses due to mortality and morbidity, loss of weight, slow growth rate, poor
fertility. The annual economic loss due to disease, mortality, reduced reproductive and productive
performance was estimated to be 150 million USD (Berhanu, 2002). In Ethiopia, 5 - 7 million sheep
and goats die each year due to disease, and the overall economic loss from the meat industry due to
parasitic diseases is estimated at 400 million annually (Sileshi and Lidetu, 2003; Frehiwot and
Samson, 2010).

Prevalence and intensity of pathogenic infections are often seasonal (Nelson et al., 2002). This could
be linked to changes in the host immunity or increase in the preponderance of either the pathogen or
vector or both (Nelson et al., 2002). The knowledge of the pattern of disease occurrence is important
as it suggests the period of likely outbreak. It also helps in planning towards ensuring proper
preventive measures (Thrusfield, 2007). Thus, the aim of the study was to assess the type,
distribution, and pattern of sheep and goat diseases and health conditions in the breeding farm.

261
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of Study Area

The study was conducted at Sirinka Agricultural Research Center (SARC) which is located 508
km northeast of Addis Ababa. The site is located at an altitude of 1850 m.a.s.l. The rainfall
pattern is bimodal, with the two rainfall season, ‘Belg’ (February/March - April) and ‘Meher’
(July - September), and the mean annual rainfall amount is on average about 950 mm. The area
is a moderately warm temperature zone with a mean daily temperature range of 20.05°C. Months
were classified as June to August are the main rainy seasons; months of September to February
are dry seasons, while months of March to May are short rainy season (Zeleke et al., 2018).

69.0 3600.0
66.0 3400.0
63.0
Temprature (◦C) and Relative humidity (%)

3200.0
60.0
57.0 3000.0
54.0 2800.0
51.0 2600.0
48.0
45.0 2400.0
42.0 2200.0

Rainfall (mm)
39.0 2000.0
36.0
1800.0
33.0
30.0 1600.0
27.0 1400.0
24.0 1200.0
21.0
18.0 1000.0
15.0 800.0
12.0 600.0
9.0
400.0
6.0
3.0 200.0
0.0 0.0
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec

Min.temp Max.temp Humidity Rainfall

Figure 1: Rainfall, humidity and temperature distribution across months in Sirinka BED site

262
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

80.0 1600.0
Relative humidity (%) and Temprature (0C) 75.0
70.0 1400.0
65.0
60.0 1200.0
55.0

Rainfall (mm)
50.0 1000.0
45.0
40.0 800.0
35.0
30.0 600.0
25.0
20.0 400.0
15.0
10.0 200.0
5.0
0.0 0.0
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Humidity Max.temp Min.temp Rain fall

Figure 2: Rainfall, humidity and temperature distribution across year in Sirinka BED site

Study Animals and their Management

Clinically sick, sheep and goats were the study subjects. In Sirinka sheep and goat breeds
evaluation and distribution site, animals were kept in the intensive and semi-intensive
management system. Lambs and kids weaned at 3 months of age. Concentrate supplementation
was provided based on their age group and physiological condition in an erratic way. Lambs and
kids had access to concentrate feed in addition to their dam’s milk before weaning. After
weaning, however, they are supplemented with 100 g/day concentrate until they are able to graze
actively. Pregnant ewes and does as well as ewes and does before the onset of mating were
supplemented with concentrate mixture while weaned female lambs and kids, non-pregnant ewes
and does before the mating season were kept on grazing land. In the main rainy season, indoor
housing has been implemented. They had free access to drinking water.

Sheep and goats were vaccinated against Sheep and Goat pox, Anthrax, Pestis Des Petites
Ruminitis, and Ovine Pasteurellosis diseases. They were treated regularly for internal and
external parasites. All animals were sprayed for ectoparasite (ticks, mites, fleas) control and
prevention with Deltametrine (1%), Diazinon (60%), Amitrazine (12.5%), Albendazole,
Tetramisole, Tetraclozash, Levamisole, Fasinex, and injected ivermectin were used for internal
parasites treatment and prophylaxis measure.

263
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Generally, there was a regular follow up of sick animals and the causes of illness and type of
disease were registered in a case recording book and appropriate treatment of antibiotic,
anthelmintic, acaricide, insecticide, and multivitamins was employed accordingly.

Study Design and Case Definition

Retrospective epidemiological study was employed using six years of health recorded data
(August 2013 to February 2018) from Sirinka Agricultural Research Center, sheep and goat
breeding, evaluation, and distribution site. The diagnosis was based on history, clinical, faecal,
and postmortem examination. In this study the age of diseased animals were grouped as young
(up to one year), adult (1.5 - 3 year) and old (>3 years) (Hassan and Nwannenna, 2009). Breeds
were categorized into local indigenous sheep (Wollo Tikur bred), cross sheep (pure Dorper x
local indigenous sheep) and exotic (pure Dorper), local goat (central high land indigenous bred),
crossbred goat (pure Boer x local indigenous goat) and exotic (pure Boer).

Overall 664 clinical cases were used for this study. Clinical cases were grouped into fifteen
major categories based on their nature. These were; enterotoxaemia, anthrax, blackleg, arthritis,
colibacillosis, acute infections were categorized as a case of Septicemia. All internal parasites
and enteritis were considered as a case of gastrointestinal parasitism. Ticks, mange mites, fleas,
and sheep kids were categorized as a case of external parasitism. Contagious orf (mouth sore),
sheep and goat pox, foot root, dermatophilosis and dermatophytosis, tumor, and any skin
eruption were under skin disease. Caseous lymphadenitis, pus containing wound, swelling,
lumpy jaw (actinomycosis), and actinobacillosis (wooden tongue) were under local abscess
cases. Pasteurellosis, contagious caprine pleuropneumonia, pest des petites ruminitis, sheeping
fever, and any respiratory embarrassment case were grouped into a case of pneumonia. Case of
abortion, retained placenta, stillbirth, orchitis, phimosis and paraphimosis, vaginal discharge, and
any reproductive system related disease were grouped to reproductive disease. Clinical and
subclinical teat and udder disease/infection were grouped under the mastitis case. Colic, bloat,
pregnancy toxemia, acidosis, ketosis, malnutrition, and other mineral deficiency cases were
grouped into the metabolic disease.

264
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Cases of heartwater, listreosis, tetanus, circling disease (coenerosis), paralysis, were grouped into
the nervous disease. The case of eye and its related disease were grouped as keratoconjunctivitis.
Cases due to mechanical and physical trauma and its effect were grouped into wounds. Case of
plant toxicity, snake bite, and other intoxication was grouped into the poisoning. Cases due to
sudden death, unknown and doubtful cases were grouped into accidental/ unpredicted cases and
cases of more than two are grouped into the mixed disease. Some disease or infection which
occurred as an outbreak and seasonal year-round tick infestation in the site starting from late
May to early November were recorded in a separate case and were not included in this report.

Data Management and Analysis

The data were checked manually for obvious inconsistencies, recording errors, or missing data.
Data were statistically analyzed by the Statistical Package for Social Science (SPSS) software
version 20.0. Descriptive statistics were used to determine the prevalence of diseases; seasonal
patterns and distribution.

The proportion of different animal’s health problems was expressed as a percentage by dividing
the total number of animals positive to a specific health problem to the total number of animals
that showed clinical disease. Chi-square (X2) test with P-value was applied to test the existence
of association and to see the level of significance between observed health problems and
associated risk factors respectively.

RESULTS

The overall proportional prevalence of different clinical case occurrences according to

decreasing order was gastrointestinal parasitism followed by septicemia, pneumonia, skin
disease, abscess, metabolic disease, reproductive disease, nervous disease, accidental and
unpredicted, wound, mixed, mastitis, poisoning, ecto parasitism and keratoconjunctivitis with a
prevalence of 24.5%, 16.11%, 10.98%, 8.28%, 5.27%, 5.12%, 4.51%, 4.51, 4.51%, 4.21%, 3.3%,
3.16%, 2.41%, 2.41%, and 1.5%, respectively (figure 3).

265
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

clinical cases across species

90
80 χ2 = 50.7081
70 P = 0.000
Proportion

60
50
40
30
20
10
0 ovine
caprine

Clinical cases

Figure 3: the frequency of all clinical cases in sheep and goat of Sirinka Agricultural Research center

The overall case occurrence, magnitude, and distributions across animal level risk factors
(species, sex, age and body condition score) were explained. The most dominant cases were
presented table 1. The highest clinical cases were observed in September, August, January, and
December and also in the dry season followed by main rainy and short rainy seasons (Figure 4).
Higher clinical cases have occurred in years of 2017, 2018, and 2016 with a proportional
prevalence of 31.5%, 21.5%, and 20.9%, respectively (figure 5).

266
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: The overall frequency and percentages of frequently occurred cases in sheep & goat across associated risk factors at SARC
Clinical cases (frequency (n) and percentage (%))
Skin Metabolic Reproductive Nervous Accidental
Risk factors / variable Septicemia Git parasitism Pneumonia disease Abscess disease disease disease cases
Species Ovine Local 12 (11.2) 16 (9.8) 8 (11) 8 (14.5) 1 (2.9) 3 (8.8) 8 (27.7) 1 (3.3) 5 (16.7)
Cross 41 (38.3) 62 (38) 28 (38.4) 25(45.5) 1 (2.9) 10 (29.4) 10 (33.3) 16 (53.3) 9 (30)
Exotic 4 (3.7) 2 (1.2) - - - - - 3 (10) 1 (3.3)
Total 57 (53.3) 80(49.1) 36 (49.3) 33 (60) 2 (5.7) 13 (38.2) 18 (60) 20 (66.7) 15 (50)
Caprine Local 5 (4.7) 14 (8.6) 6 (8.2) 3 (5.5) 8 (22.9) 2 (5.9) 7 (23.3) 2 (6.7) 6 (20)
Cross 44 (41.1) 67 (41.1) 29 (39.7) 19(34.5) 22(62.9) 19 (55.9) 5 (16.7) 7 (23.3) 9 (30)
Exotic 1 (0.9) 2 (1.2) 2 (2.7) - 3 (8.6) - - 1 (3.3) -
Total 50 (46.7) 83 (50.9) 37 (50.7) 22 (40) 33(94.3) 21 (61.8) 12 (40) 10 (33.3) 15 (50)
Sex Female Ovine 33 (30.8) 49 (30.1) 23 (31.5) 28(50.9) 1 (2.9) 10 (29.4) 18 (60) 12 (40) 9 (30)
Caprine 27 (25.2) 43 (26.4) 23 (31.5) 17(30.9) 30(85.7) 15 (44.1) 12 (40) 9 (30.0) 11 (36.7)
Total 60 (56.1) 92 (56.4) 46 (63.1) 45(81.8) 31(88.6) 25 (73.5) 30 (100) 21 (70) 20 (66.7)
Male Ovine 24 (22.4) 31 (19.0) 13 (17.8) 5 (9.1) 1 (2.9) 3 (8.8) - 8 (26.7) 6 (20)
Caprine 23 (21.5) 40 (24.5) 14 (19.2) 5 (9.1) 3 (8.6) 6 (17.6) - 1 (3.3) 4 (13.3)
Total 47 (43.9) 71 (43.6) 27 (36.9)) 10(18.2) 4(11.4) 9 (26.5) - 9 (30) 10 33.3)
Age Young Ovine 42 (39.3) 43 (26.4) 26 (35.6) 6 (11) 1(2.9) 7(20.6) - 10 (33.3) 8 (26.7)
Caprine 34 (31.8) 59 (36.2) 27 (37) 8 (14.6) 5 (14.3) 10(29.4) 1 (3.3) 7 (23.3) 7 (23,3)
Total 76 (71) 102 (62.6) 53 (72.6) 14(25.5) 6 (17.1) 17 (50) 1 (3.3) 17 (56.7) 15 (50)
Adult Ovine 15 (14) 30 (18.4) 10 (13.7) 26(47.3) 1 (2.9) 6 (17.7) 17 (56.7) 10 (33.3) 7 (23.3)
Caprine 13 (12.2) 21 (12.9) 5 (6.9) 11 (20) 24(68.6) 9 (24.5) 11 (36.7) 3 (10) 7 (23.3)
Total 28 (26.2) 51 (31.3) 15 (20.5) 27(49.1) 25(71.4) 15 (44.1) 28 (93.3) 13 (43.3) 1446.7)
Old Ovine - 6 (3.7) - 1 (1.8) - - 1( 3.3) - -
Caprine 3 (2.8) 3 (1.84) 5 (6.9) 3 (5.5) 3 (14.3) 2 (5.9) - - 1 (3.3)
Total 3 (2.8) 9 (5.5) 5 (6.8) 4 (7.3) 3 (8.6) 2 (58.8) 1 (3.3) 1 (3.3)
Body Ovine Poor 6 (5.6) 15 (92) 2 (2.7) 2 (3.6) - 2 (5.9) - 1 (3.3) 2 (6.7)
condition Moderate 5 (4.7) 14 (8.6) 3 (41.1) 5 (9.1) - 1 (2.9) 2 (6.7) 3 (10) -
score Good 13 (12.1) 13 (7.8) 10 (13.7) 5 (9.1) 1 (2.9) 1 (2.9) - 6 (20) 3 (10)
Na/ miss 33 (30.8) 38 (23.3) 21 (28.8) 21(38.2) 1 (2.9) 9 (26.5) 16 (53.4) 10 (33.3) 10 (33.3)
Total 57 (53.3) 80 (49.1) 36 (49.3) 33 (60) 2 (5.7) 13 (38.2) 18 (60) 20 (66.7) 15 (50)
Caprine Poor 1 (0.9) 36 (22.1) 3 (41.1) 5 (9.1) 1 (2.9) 5 (14.7) - - 1(3.3)
Moderate 10 (9.3) 17 (10.4) 14 (19.2) 6 (10.9) 6 (17.1) 2 (5.9) - 1 (3.3) -
Good 12 (11.2) 4 (2.5) 8 (11) 1 (1.8) 1 (2.9) 1 (2.9) - 2 (6.7) 1 3.3)
Na/ miss 27 (25.2) 26 (16) 12 (16.4) 10(18.2) 25(71.4) 12 (35.3) 12 (40) 7 (23.3) 13 (43.3)
Total 50 (46.7) 83 (50.9) 37 (50.7) 22 (40) 33(94.3) 21 (61.8) 12 (40) 10 (33.3) 15 (50)

267
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

120
Clinical cases across dfferent seasons
100

χ2 = 63.77 Dry season

80 P = 0.000
Proportion

Short rainy
60
Main rainy season
40

20

Clinical cases
Figure 4: Clinical cases of sheep and goat in sirinka ARC across seasons

80
Frequncy of cases across year
70 2013
60
χ2 = 185.91 2014
Frequncy

50
P = 0.000
40 2015
30
2016
20
10 2017
0 2018

Clinical cases
Figure 5: Clinical cases of sheep and goat in sirinka ARC across years

From all clinical cases the considerable numbers of animals 26.36% have died, while 73.64%
have recovered. Accidental/unpredicted cases were the leading followed by poisoning and
metabolic disease for the occurrences of the highest death while abscess, mastitis,
keratoconjunctivitis and wound were the most recovered cases (Table 2).

268
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: The frequency of responses from all clinical cases sheep and goat in sirinka ARC
Frequency of response
Recovered (%) Died (%)
Types of Cases Sheep Goat Total Sheep Goat Total Overall case χ2 (P-value)
Septicemia 43 (40.2) 43 (40.2) 86 (80.4) 14 (13.1) 7 (6.5) 21 (19.6) 107 (16.1) 55.14 (0.00)
GIT parasitism 68 (41.7) 72 (44.2) 140 (85.9) 12 (7.4) 11(6.7) 23 (14.1) 163 (24.6)
Ecto parasitism 6 (40.0) 2 (13.3) 8 (53.3) 5 (33.3) 2 (13.3) 7 (46.7) 15 (2.3)
Skin disease 24 (43.6) 8 (14.5) 32 (58.2) 9 (16.4) 14 (25.5) 23 (41.8) 55 (8.3)
Abscess 2 (5.7) 32 (91.4) 34 (97.1) - 1 (2.9) 1 (2.9) 35 (5. 3)
Pneumonia 25 (34.2) 27 (37.0) 52 (71.2) 11 (15.1) 10 (13.7) 21 (28.8) 73 (11)
Reproductive disease 16 (53.3) 9 (30) 25 (83.3) 2 (6.7) 3 (10.0) 5 (16.7) 30 (4.5)
Mastitis 5 (31.3) 10 (62.5) 15 (93.8) 1 (6.3) - 1 (6.3) 16 (2.4)
Metabolic disease 10 (29.4) 8 (23.5) 18 (52.9) 3 (8.8) 13 (38.2) 16 (47.1) 34 (5.1)
Nervous disease 10 (33.3) 8 (26.7) 18 (60.) 10 (33.3) 2 (6.7) 12 (40) 30 (4.5)
Keratoconjuctivities 7 (70.0) 2 (20.0) 9 (90.0) 1 (10) - 1 (10) 10 (1.5)
Wound 9 (32.1) 16 (57.1) 25 (89.3) 1 (3.6) 2 (7.1) 3 (10.7) 28 (4.2)
Poisoning 2 (12.5) 1 (6.3) 3 (18.8) 3 (18.8) 10 (62.5) 13 (81.3) 16 (2.4)
Accidental/Unpredicted 2 (6.7) 9 (30.0) 11 (52.4) 13 (43.3) 6 (20.0) 19 (90.5) 30 (4.5)
Mixed 7 (31.8) 6 (27.3) 13 (59.1) 7 (31.8) 2 (9.1) 9 (40.9) 22 (3.3)
Total 236 (35.5) 253 (38.1) 489 (73.6) 52 (7.8) 20 (3) 175 (26.4) 664 (100)

269
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

DISCUSSIONS

All clinical cases (664) were arranged in 15 groups based on their nature, similarities of origin,
and clinical presentations (organ or system affected). From this retrospective study
gastrointestinal parasitism (24.55%), septicemia (16.11%), pneumonia (10.99%), skin disease
(8.28%) were the most dominant cases, whereas, keratoconjunctivitis (1.51%), external
parasitism (2.26%), mastitis (2.41%), and poisoning (2.41%) were the least cases in the study.
Both species, ovine 328 (49.4%) and caprine 336 (50.6%) had shown an equal number of cases.
This was due to a similar chance of exposure to infection and any management systems including
feeding and watering point. This result showed almost similar prevalence of internal (24.55%)
parasite cases as compared with previous work of Zegeye et al. (2014) who reported the highest
prevalent case of internal (git) parasite (17.3%). However, the prevalence of external parasite in
our study was by far lower (2.26%), than Zegeye et al. (2014) who reported (21.4%) this might be
due to grazing land management /burning/, cleaning measures of the farm /making the floor
cement/, as a result, there was a seasonal occurrence of external parasites mainely ticks.

The overall prevalence of external and internal parasite of ovine and caprine in this study was still
much lower (26.81%) than the value 95.6% reported by Abebe and Esayas (2001), 48.1% by
Kumsa et al.(2012), 38.7% by Zegeye et al. (2014), and 45.5% by Zewdu et al. (2015). This
variation might be due to the management practice difference when compared to smallholder
farmers and extensive management or husbandry system. Despite regular deworming of animals
against internal parasites was practiced on the farm, the highest cases were recorded. This could
lead to suspect the occurrence of anthelmintic drug resistance in the farm and frequently grazed
on the same grazing land might be reinfection of animals by internal parasites the tactic and
strategic deworming protocol may not work there. The occurrence of septicemic cases was
expected as regular vaccination were given only for four major infectious diseases of sheep and
goats, which were similar to the report of Zegeye et al. (2014)

Regarding the occurrence of cases on sex and breed as a risk factor, the farm is established to
conduct breeding and evaluate the performance of Dorper x local sheep and Boer X local goats.
Hence, there were low numbers of exotic rams and bucks, males, and even local sheep and goats
in the farm which were directly correlated with the number of clinical cases recorded on each
group of animals on the farm. So, in this study, the highest clinical cases were registered in cross
goats (39%) followed by cross sheep (36.14%), local sheep (11.14%) and local goats (10.24%)
from the overall cases.

270
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The highest incidence of abscess in Goat was due to the aggressive nature and the frequent
occurrences of Caseous lymphadenitis in the site. All cases were recorded in the crossbreds, but
in the exotic breed only a few were recorded, this was due to the very small number of exotic
breeds compared to the others. The result disagrees with the finding of Zegeye et al. (2014), who
reported that metabolic disease particularly hypocalcemia (pregnancy toxemia) was not recorded
in crossbred goats.

In this study, female animals were more infected than males based on the recorded case. This
result agrees with Zegeye et al. (2014). Reproductive disease and Keratoconjunctivitis cases were
not recorded in both male Sheep and Goats. The reason for this variation might be a large number
of females were kept for breeding purposes while males were distributed to the farmers, thus less
number of bucks and rams were left on the farm. The finding was in line with the result of
Sefinew and Bider (2011), who reported that female animals have a low level of immunity that
associated with physiological nature and the susceptibility to infection in a female animal is
maximized during periparturient and parturition where relaxation of resistance and suppression of
immunity happens.

Animals at a young age were more significantly susceptible to infection which is similar to the
previous result of Zegeye et al. (2014) This could be associated with less immunity and exposure
to the infectious agent during weaning shock caused by sudden feed change (Taye et al., 2009).
The case of pneumonia (72.6%), septicemia (71.03%) accidental (66.67%), and ecto parasitism
(66.67%) were the most cases in young animals than other age groups, which was similar to
Zegeye et al. (2014). The effect of age on the highest occurrence of pneumonia was agreed with
the report of Donklin et al. (2004) that recorded kids were susceptible to pneumonia.

Most cases occurred and recorded in September, August, January, and December while the
lowest number of cases observed in March, April, and May. The factor attributes to the highest
and lowest observation of cases might be the changes in temperature, humidity, and feed
availability which directly determines disease occurrence. The highest and most cases were
recorded in the dry season (56.63%) followed by the main rainy season (29.22%) and the short
rainy season (14.16%) from all cases. The finding was similar to the report of Yusuf et al. (2016)
who reported that many infectious disease occurrences during the dry season tend to rise. This
might be associated with inadequate pasture which may lead to starvation and reckless
scavenging of food. The available pastures tend to have lost their nutritive value. It is thus evident
that this condition affects animals and this lack of nutrients during such seasons impairs the
ability of animals to mount an immune response to fight infections.

271
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

But it is contrasting from the report of Zegeye et al. (2014) who reported that the highest
clinical cases have occurred in the short rainy season while the lowest were in the main rainy
season. The deviation might be associated with temporal variation as the distribution of the
disease-causing agent was influenced by climatic factor like; rainfall, humidity and
temperature and management system used for the animal and disease-causing factor
including the intermediate host, all determine the epidemiology of the disease as well
(Kusiluka and Kambarage, 2006).

The most prevalent case observed in the study was gastrointestinal parasitism. Normally GIT
parasites are dormant and undergo hypobiosis in the dry season. Septicemia, pneumonia, and
skin disease were recorded in the dry season. This might be associated with the shortage of
quality feed, increased chance of the spread of infectious agents through wind easily and
reduced immunity as a result of the opportunistic microflora of the animal becoming an
infectious agent. In agreement with Lopez (1995) and Radostits et al. (2007), the highest rate
of poisoning was observed in the dry season. This is directly related to the feed shortage
during the dry season that forced the animals to fed unusual feed source plants including the
toxic plant since they remain green throughout the year.

The highest prevalence was observed in 2017 (31.5%) followed by 2018 (21.5%), 2016
(20.9%) and 2015 (14.0%), 2014 (8.4%) and the least was 2013 (3.6%). The inconsistency
number of cases might be due to the population dynamic of the flock and climatic condition
which has erratic nature of the area and an outbreak of cases like mouth sore (contagious
erythema, orf) in the farm during 2017.

Considerable numbers of animals died (26.36%) even though a higher number of animals
have recovered (73.64%) from their illness. Accidental /unpredicted cases were the leading
fatal cases followed by poisoning and metabolic diseases while abscess, mastitis,
keratoconjunctivitis, and wound were the most recovered cases. Those cases which account
for the high frequency of death were found difficult to treat once they happened/occurred/.
This might be due to the acute and per-acute nature of cases. However, gastrointestinal
parasitism, septicemia, and pneumonia that have a high prevalence in the study area, their
fatality rate was relatively low. This might be associated with the effectiveness of the
treatment given during the time of intervention.

272
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

The study revealed that gastrointestinal parasitism, septicemia, pneumonia, and skin disease
were the most dominant sheep and goats health problems of Sirinka agricultural research
center, breeding evaluation, and distribution site. These cases are mostly observed during the
dry season followed by the main rainy and short rainy seasons. Young, female, and crossbred
animal groups were more susceptible than others. While the distributions of cases are affected
by temporal factors, the distributions of the disease in both species were almost the same.
From those clinical cases, accidental /unpredicted/ cases were the first fatal case, followed by
poisoning and metabolic diseases. Therefore, grazing land management and efficacy of
commonly used anthelmintic drugs and antimicrobial susceptibility study should be done on
the site. Better management practices should also be made to physiologically susceptible
young, female, and crossbreed animals, especially in the dry season. Close follows up,
immediate interventions and strong health record-keeping need to be continued. Besides,
further investigation has to be done to identify the sources of accidental and poisoning cases.

ACKNOWLEDGEMENT

We would like to thank all the livestock research directorate staff of Sirinka Agricultural
Research Center for their technical support during the study period.

REFERENCES

Abebe W. and Essays G. 2001. Survey of ovine and caprine gastro-intestinal helminthosis in
eastern part of Ethiopia, during the dry season of the year. Vet., 152: 379-384.
Adebowale A.L. 2012. Dynamics of ruminant livestock management in the context of the
Nigerian Agricultural System. In Livestock Production (Eds. Khalid Javed). Intech.
USA. ISBN978-953-51-0814-6. 61.
Belachew Hurrissa and Jemberu Eshetu, 2002. August. Challenges and opportunities of
livestock marketing in Ethiopia. In Proceedings of the 10 th annual conference of the
Ethiopian Society of Animal Production (ESAP). Addis Ababa, Ethiopia pp. 1-13).
Berhanu A. 2002. Welcome address; Animal health and poverty reduction strategies. In:
proceedings of the 16th Annual Conference of the Ethiopian Veterinary Association
(EVA) held 5-6 June, ,Ghion Hotel, Addis Ababa, Ethiopia, 117-137.

273
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Central Statistical Authority. 2017/18. Central Statistical Agency Agricultural Sample

Survey, Federal Democratic Republic of Ethiopia, Central Statistical Authority, volume
II report on livestock and livestock characteristics.
Donkin E.F., Boyazoglu P.A. and Eglan. 2004. Diseases and mortality of goat kids in South
Africa milk goat herd. South Africa Journal of Animal Science, 34: 258-261.
Hassan Z.A. and Nwannenna A.I. 2009. Animal handling and restraint. An instructive
Approach 1st ed Ndahi Press: 61. Kumsa B., Beyecha K. and Geloye M. 2012.
Ectoparasites of sheep in three agro-ecological zones in central Oromia, Ethiopia’,
Onderstepoort Journal of Veterinary Research, 79(1).
Kusiluka L. and Kambarage D. 2006. Diseases of Small Ruminants, A Handbook, Common
Diseases of Sheep and Goats in Sub-Saharan Africa. pp: 8-11.
Lopez A. 1995. Respiratory System. In: W.W. Carlton and M.D. McGavin: Thomson's
Special Veterinary Pathology, 2, 116-174.
Mukasa-Mugerwa E. 1989. Review of reproductive performance in female Bas indicus
(Zebu) cattle. ILCA monograph. No.6.ILCA.Addis Ababa.
Nelson J.R., Demas E.G., Klein L.S. and Kriegsfeld J.L. 2002. Seasonal patterns of stress,
immune function and disease. 1st ed. Cambridge University Press,
Omoike A. 2006. Prevalence of disease among sheep and goats in Edo State, Nigeria. J. Agri.
Social Res. 6(2).: 23- 31.
Radostits O.M., Gay C.C., Hinchcliffand K.W. and Constable P.D. 2007. Veterinary
Medicine, a Textbook of the Diseases of Cattle, Horses, Sheep, Pigs and Goats. 10th Ed.
London: Saunders Elsevier. pp: 119- 1966.
Samson L. and Frehwot M. 2010. Prevalence of Small Ruminant Trypanosomosis and Tsetse
Fly Challenge in Upper Didessa Valley, Ethiopia, Global Veterinaria, 5 (4): 215-219
Sefinew A. and Bider Z. 2011. Occurrence and associated risk factors of clinical diseases of
farm animals presented to Gondar University Veterinary Clinic. American-Eurasian
Journal of Agricultural & Environmental Sciences, 11(2), 237-241.
Sileshi Z. and Lidetu D. 2003. Control of internal parasites in sheep and goats.
www.esgpip.org/PDF/Technical bulletin no .3.
Taye M., Abebe G., Gizaw S, Lemma S., Mekoya A. and Tibbo M. 2009. Growth
performances of Washera sheep under smallholder management systems in
Yilmanadensa and Quarit districts, Ethiopia. Trop. Anim. Heal. Prod.

274
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Taye, M., Abebe, G., Gizaw, S., Lemma, S., Mekoya, A. and Tibbo, M. 2010. Growth
performances of Washera sheep under smallholder management systems in
Yilmanadensa and Quarit districts, Ethiopia. Tropical animal health and production,
42(4), pp.659-667.
Tefera M., Batu G. and Bitew M. 2009. Prevalence of gastrointestinal parasites of sheep and
goats in and around Bedelle South Western Ethiopia. Internat J. Vet. Med. 8(2): 1-3.
Thrusfield M. 2007. Veterinary epidemiology 3rd Edition, Oxford: Black Science Ltd. pp
228-246.
Yusuf Audu, Jamila Dauda, Yusuf Madaki, Aliyu Haruna and Mustapha Bamanga. 2016. A
Retrospective Study of Cases of Disease Condition of Large and Small Ruminants
Handle at the State Veterinary Hospital Maiduguri, Borno State, Nigeria. Greener
Journal of Epidemiology and Public Health. 4 (3) 055-060,
Zegeye Asres, Belayineh Negus, Shenkutie Abiy, Lakew Mesfin and Tesfaye Abebe. 2014.
On-Station Retrospective Epidemiologic Study of Small Ruminant Disease in Sirinka
Agricultural Research Center, Sheep and Goat Breeding, Evaluation and Distribution
Site, Academic Journal of Animal Diseases 3(3): 39-44.
Tesema, Z., Tilahun, M., Yizengaw, L., Zegeye, A., Bisrat, A. and Abebe, A. 2018. Growth,
carcass and non-carcass characteristics of Central Highland and Boer x Central Highland
goats under different levels of supplementation. Livestock Research for Rural
Development, 30, 11.
Zewdu Seyoum, Tsegaye Tadesse and Agerie Addisu. 2015. Ectoparasites Prevalence in
Small Ruminants in and around Sekela, Amhara Regional State, Northwest Ethiopia,
Journal of Veterinary Medicine (2015), 6.

275
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

A Case Control Study of Infectious and Non-Infectious Causes of Abortion in sheep and
Goats of Abergelle and Ziquala Districts of Amhara Region, Northern Ethiopia

Adane Bahiru1, Ayalew Assefa1, Biruk Alemu2, Hiwot Desta2, Barbara Wieland2 and Abebe
Sahle3
1
Sekota Dryland Agricultural Research Center, P.O. Box 62, Sekota, Ethiopia
2
International Livestock Research Institute, P.O. Box 5689, Addis Ababa, Ethiopia
3
Ziquala Woreda Office of Agricultural, Ziquala, Ethiopia

ABSTRACT

A case-control study was conducted in Abergele and Ziquala districts in June 2016 to assess
the occurrence and magnitude of abortion in small ruminants, and to identify the major
causes of small ruminant abortion. Case village defined as villages having an abortion storm
and control village as villages with low abortion storm. Questionnaire and focus group
discussion was used to see differences between case and control villages in causes of abortion
storm. Rose Bengal plate test used to test Brucellosis and ELISA used to test Coxiella and
Chlamydophila. All the 30 household flocks in the case villages affected by abortion (a mean
abortion of 13 animals in Abergele and 9 in Zukuala), however only 3 households reported
abortion in the control villages. A total of 176 blood samples were collected from the case
and control villages for further laboratory diagnosis of possible causes of abortion. The
serological investigation for the prevalence of Coxiella burnetii was 17%, Toxoplasma gondii
was 10.2%, and Chlamydophila abortus was 2.8%. However, the Brucella melitensis
antibody, which is the well-known cause of abortion, was not detected in this study. While
the seroprevalence was higher in some infectious agents of abortion, there was no significant
difference between the case and control villages in the occurrence of disease. The logistic
regression analysis showed no significant difference between sex groups and study districts
for the presence of organisms regardless of case and control groups. However, there was a
significant (P = 0.017) difference between the study districts in the occurrence of Coxiella
burnetii. It is clear that the abortion problem in the study villages is complex and likely
presents a mix of poor husbandry practice and the presence of infectious diseases, which are
confounded by harsh climatic conditions. To better understand the underlying causes of
abortion, there is a need to conduct a longitudinal study involving more disease and
household-level sample collection with improved husbandry practices.

Keywords: Abergelle, Abortion, Case-control study, infectious, non-infectious, Ziquala

276
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Sheep and goats are highly adaptable to a broad range of environmental conditions.
Moreover, the low cost of production, little land requirement, and higher prolificacy made
them an attractive asset for development. Investment in sheep and goats avoid losses due to
high inflation rates that are found in unstable economies of many developing countries like
Ethiopia (Mohammed et al., 2017). Ethiopia is an agriculturally based country and owns a
considerable number of small ruminants, which is estimated to be over 61 million heads of
sheep and goats (CSA, 2017). Despite these large-small ruminant population sizes, the
country fails to optimally utilize these resources as the sector is suffering lower productivity
due to various factors in which diseases stand front line (Mohammed et al., 2017). One of the
diseases that hamper the productivity of small ruminants is abortion, which has major
economic and public health impacts (Mohammed et al., 2017).

Reproductive failure harms livestock production, animal health and welfare, and, ultimately
rural economies. There are several potential underlying causes of reproductive failure broadly
categorized into infectious and non-infectious causes (Gebremedhin et al., 2013). Abortion,
stillbirth, and delivery of weak lambs or kids are important issues of sheep and goat breeding
(Szeredi et al., 2006). Abortion in sheep and goats has become increasingly important as the
potential zoonotic significance of commonly involved pathogens; in particular, Coxiella
burnetii and Chlamydophila abortus are better understood. Coxiella burnetii is well known as
the cause of Q fever in human beings (Hazlett et al., 2013).

Toxoplasmosis is a parasitic disease of humans, wild and domestic animals caused by

Toxoplasma gondii, a cosmopolitan protozoan of the sarcocystidae family with animals of the
Felidae family as a definitive host. Toxoplasmosis is usually subclinical, with the only
common signs of hyperthermia, dyspnea and neurological problems. It is considered as the
major causative agent of abortion. Toxoplasmosis also has a zoonotic effect (Klauck et al.,
2016). The difference in Climatic and geographical factors and human and animal customs
affects its prevalence from place to place (Al et al., 2019). The disease can transmit between
animals through the ingestion of contaminated oocytes. Consumption of tissue cysts from
undercooked meat and oocysts intake from food gives access for a human to be infected with
toxoplasmosis. Besides, it can also be transmitted congenitally (Bachan et al., 2018).

277
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The disease Q fever is caused by the bacteria Coxiella burnetii, ubiquitous, gram-negative
bacteria capable of infecting humans as well as a wide range of animals. The disease is
characterized by febrile condition and can cause abortion in dairy animals (Mohammed et al.,
2014). Chlamydiae are obligate intracellular bacteria, which have an increased presence and
prevalence in animals. They infect ocular, genital and respiratory tissues leading to a variety
of diseases including abortion, pneumonia, gastroenteritis, encephalomyelitis, kerato-
conjunctivitis, arthritis, orchitis, seminal vesiculitis and epididymitis (Rodolakis and
Laroucau, 2015).

Brucellosis is an infectious disease of worldwide importance in domestic ruminants and

causative bacteria (Brucella abortus and Brucella melitensis in sheep and goats) are
transmitted to humans through contact with infected or by consumption of contaminated
dairy products. Because of the economic importance of cattle, diagnosis, research activities
and prophylaxis measures have been made on B. abortus intensively (Gupta et al., 2007). In
contrast, although goats are crucial in the economy of developing countries and B. melitensis
is a common cause of human infection, the brucellosis of goats has received comparatively
little attention.

Abortion is probably the most common complaint in goats in Abergele and Ziquala district of
the Amhara region. The consequence of an abortion storm can have a devastating economic
impact in the community where goat production is the basis for livelihood. Given the recently
reported increasing levels of abortion incidence, a detailed investigation was conducted with
the following objectives.

• To assess the occurrence and magnitude of abortion in small ruminants

• To identify the major causes of small ruminant abortion

MATERIAL AND METHODS

Description of the Study Area

The study was carried out in two selected districts (Ziquala and Abergelle) of the Amhara
region, Northern Ethiopia. The study sites were selected purposely based on the reports of
abortion outbreaks. Abergele district is characterized as hot, warm sub-moist lowland and
located at an average altitude of 1450 masl. The average annual rainfall varies from 350 to
650 mm, and the temperature ranges from 18 to 42°C (Assefa and Bahiru, 2018).

278
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 1: Location map of study sites

Study Design and Sample Size Determination

Case control study design of households affected by recent abortion storms and those with no
reports of recent abortions were included in the study. Sample size of cases and control
groups was determined by the following formula.

r + 1 ( p )(1 − p )( Z  + Z/2 )
2

n=( )
r (p1 − p2 ) 2

With Expected proportion in controls = 0.05, Assumed odds ratio = 3, Confidence level =
0.95, and study Power = 0.8, the sample size per group was 88 with a total sample size of 176
animals in both groups (Thrusfield and Christley, 2018).

Questionnaire Survey and FGD

A questionnaire survey in 30 households and 2 FGDs were conducted with small ruminant
producing farmers in two case villages of ICARDA research program sites (Sazeba
(Abergelle district) and Bilaque (Ziquala district)), where abortion storms reported. An
additional 28 households were interviewed in two control villages at Tsetseka (Ziquala
district) and Mehal Genet/ Fertata Bakuna (Abergelle district), where no abortion storms
reported.

279
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sample Collection

A total of 176 blood samples were collected in all villages for further laboratory diagnosis of
the reproductive disease, which may cause the observed abortion storms. One kebele from
each district was included in the study. The blood sample was collected by bleeding from the
jugular vein using vacutainer needles with their respective plastic cap, adjusted to 8ml
vacuum tubes for serum (without anticoagulant). Tubes were transported in coolers keeping a
temperature between 5 and 10°C. For serum separation, blood samples were allowed to stay
for 24 hours at room temperature. The serum was frozen at -20°C until laboratory procedure
was done.

Details of Each Laboratory Procedure

Rose Bengal Plate Test

Rose Bengal plate test Kit was used to detect Brucella melitensis bacteria-specific antibodies
in goat serum samples. The brief steps were as follows. Bring the antigen to room
temperature 30 minutes before use. Shake the antigen bottle gently to obtain a homogeneous
suspension. Bring the test and control sera to room temperature 30 minutes before use. Place
25-30 µl of neat serum and the antigen side by side on a plate and Mix thoroughly and
rapidly the serum and the antigen. Shake for about 4 minutes and observe for agglutination.
The results were expressed as no agglutination (negative), visible agglutination (positive) and
Flocculates (false agglutination) as interpretable or unreadable (Swai and Schoonman, 2010).

Enzyme-Linked Immunosorbent Assay (ELISA)

Direct ELISA was used to detect specific antibodies for Chlamydophila abortus, Coxiella
burnetii. Serum samples were processed according to the manufacturer's protocol. The sera
were diluted 1:10 for the analysis of each agent. For an adequate interpretation, the average
of the optical densities (OD) of the positive controls, and the difference between standards of
ODs of positive and negative control sera were calculated. Serum positive percentages (S/P)
were calculated according to OD data from the different serum samples and the average of
OD of the positive control sera, using the following formula: S/P = (OD of sample×100):
(average OD of positive control). As recommended by the manufacturer, the serum samples
with S/P percentages <40% were considered as negative; samples with S/P values between 40
and 50% were scored as inconclusive (considered as doubtful in this study) and sera with S/P
values >50% were determined as positive (Gokce et al., 2007).

280
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

For Toxoplasma the Toxoplasma gondii Antibody Test Kit, (IDEXX®Switzerland AG, 3097
Liebefeld-BernSwitzerland) was used.

Data Management and Analysis

All the records were coded in Microsoft excel, and analysis was done using the STATA
software program. Descriptive statistics and inferential statistics (chi-square test and logistic
regression) were used to summarize results. A p-value of less than 0.05 was considered a
statistically significant association between variables.

RESULTS AND DISCUSSION

Small Ruminant Herd Structure of Surveyed Households

In the interviewed households, the proportion of young (particularly the yearlings) stock was
low compared to adults in the flock that could be fewer kids/lambs being born or pre-weaning
mortality was high during the study. The herd structure of surveyed households is
summarised in Table 1. There were a higher number of Does and Ewes than the other age
groups, which is similar to the study in the Dawro zone where ewes and Does were
significantly higher than the others (Alilo et al., 2018). The presence of a higher number of
ewes/does in the flock is possibly due to the households keep more productive females than
other age and sex groups.

The Occurrence of Abortion

Most of the households reared cattle, sheep, goats, and donkeys. The population of the goat is
the highest among all other species in each household. Although the abortion problem had
presented in previous years, farmers reported an increased number of abortions that occurred
in 2016. The abortion storm was reported to occur explicitly in goats except for very few
abortions in sheep and cattle, which were not significant in number. All the 30 household
flocks included in the case villages were affected by abortion with different magnitude
compared to that of the control villages that reported only 3households out of 28. It was
evident that the scope of abortion is much higher in case villages as the cases and controls
defined purposely. A total of 339 animals in the study sites, 327 from the case, and 12 from
the control village aborted. On average, 13 sheep/goats had aborted per household in Sazeba
village, with a minimum of 3 and a maximum of 30 ewes/Doe aborted per household (Table
2).

281
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

During the FGD, farmers described the abortion storms occurring in two episodes. The first
phase of abortion happened from October to November 2015. They described this
phenomenon as of late abortion since kidding/lambing is mostly expected in November and
December. They reported big sized fetuses; some were full-term kids with skin cover.

The second phase of abortion occurred between March and June 2016, with a peak in May.
Many abortions were challenging to detect since they only saw blood and noted vaginal
discharge. They recognized this as early abortion. They described the situation with local
language directly as 'Ye Jerat Dem Hone', which indicates that they only noticed uterine
blood and discharge staining the wool in the perineal region, specifically on the tail.

282
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Herd structure owned by study participants in Ziquala and Abergele districts
Mean number of sheep per household Mean number of goats per household
District Kebele Status Total Ewes Lambs yearlings Rams Total Does Kids Yearlings Bucks
Abergele Sazeba case 15 8 9 2 2 50 32 16 3 2
FertataBaquna control 16 9 3 2 2 28 15 7 5 1
Ziquala Bilaque case 18 8 6 3 1 33 18 10 5 2
Tsitsika control 8 3 2 1 1 11 6 3 1 1

Table 2: Number of abortions reported in Abergele and Ziquala districts

Mean no. of animals Total no. of

District Kebele Status aborted at HH Min Max animals aborted Remark
Abergele Sazeba Case 13 3 30 193 Abortion occurred in all 15 HH studied
Fertata Baquna Control 0.3 0 5 5 Abortion reported in a single HH of the 15 studied
Zequala Bilaque Case 9 1 20 134 Abortion occurred in all 15 HH studied
Tsetseka Control 0.5 0 5 7 Abortion reported from 2 HH of 13 studied
Total 339
HH = households

283
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The result of the household survey confirmed this description. From the total animals aborted,
69% of the animals aborted from March to June and 20% from October to December (Figure 2).
According to respondents, 44% of abortion occurred during the early phase of pregnancy, less
than one month, while 30% reported abortion during the late stage of pregnancy greater than four
months (Figure 3).

Number of abortions reported

August 2015 - July, 2016
100
80
60
40
20
0

Figure 2: Abortion distribution of goat and sheep with months of the year

50
44.2
% of respondant reported

45
40
34.9
35 30.2
abortion

30
25
20
15
10
4.7
5
0
< 1 month 1-4 month > 4month Any time
Pregnancy Time

Figure 3: The occurrence of abortion in sheep and goat with different pregnancy stage

The commonly mentioned clinical signs in aborted animals were rough hair coats, fever, vaginal
discharge, and retained placenta. 30% and 38% of the respondents also encountered stillbirth and
weak kids/lambs, respectively during the abortion outbreak.

284
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Management of Aborted Fetus

There was no practice of proper disposal of the fetus after an abortion. The majority of
respondents, 79% stated that they gave aborted fetuses to dogs, and 19% said they left the
aborted fetus as it is. Only 2% of the respondents reported burying aborted material. This
condition can be the reason for the continued occurrence of abortion since some infectious
causes of abortion can transmit to the other herd.

Causes of Abortion Mentioned

During FGD, farmers suggested that the cause for the abortions in October - December was due
to disease, even though they couldn’t observe any new disease sign during this period than the
previous years. They mentioned the occurrence of PPR and Blackleg in some flocks by
observing the signs they observed and translated to the scientific name by researchers. They
noted that weather conditions, feed, and water availability were relatively pleasant during this
time as an outcome of the main rainy season. In contrast, most farmers relate the abortions from
March to June with the drought season (harsh sun), which affected the animals immediately after
mating. They mentioned 'Tegage,' a weather condition characterized as very hot and 'dull' air.
This weather occurs about every 2 or 3 years and causes drying of the vegetation, the shrubs, and
trees. Based on the household survey, only 6% reported feed change during the dry season by
providing supplements in the form of concentrate feed. There was no change in the water source
and animal movement compared to previous years. Around 53% of the respondents thought the
cause for abortion was the extended dry season, drought and feed scarcity, and environmental
stress due to the intense sun. Only 13% assumed the cause could be a disease.

Mitigation Strategies

About 59% of respondents described that during times of abortion storms, they try to keep
pregnant animals around the homestead in the shed to protect them from the sun and supplement
with crop residues like millet and maize stover and 56% of them feel this helped much to reduce
the abortion level. Still, they face a shortage of feed to manage large flocks.

285
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Differences Noted By Farmers in Control Villages

During the FGD, farmers in the control villages indicated that they have a water source in the
form of a river which is found near to their village. They mostly use irrigated forage like cowpea,
elephant grass, and sesbania trees to feed their animals. There are also tree sheds available to
keep animals. Flock size is relatively small compared to the villages where abortions occurred.

Infectious Causes of Abortion

Serological investigation of samples collected revealed that the prevalence of Coxiella burnetii
was (17%), Toxoplasma gondii 10.2% and Chlamydophila abortus 2.8%. However, the well-
known cause of abortion, Brucella melitensis, was not detected. This result agrees with the
findings of (Gebretensay et al., 2019) who reported 89.17% prevalence of chlamydiosis, 68.33%
C. burnetii, and 70.83% toxoplasmosis in Menz and Horo areas. Besides, (Tilahun et al., and
Shimelis, 2018) reported even higher prevalence of toxoplasmosis. These differences can be
attributed to geographical and production system variations as well as diagnosis test accuracies.

Even though higher seroprevalence of the organism was noted, there was no statistical difference
in the occurrence of organisms between the case and control groups. This can be because animals
from both cases and control groups can serologically positive even if an abortion storm was not
observed in that particular year.

From the organisms isolated, Coxiella burnetii is suspected to be the major cause of abortion,
followed by Toxoplasma gondii and Chlamydophila abortus. The overall prevalence of
organisms is depicted in the following table in both cases and controls groups (Table 3).

Table 3: Seroprevalence of isolated organisms in Abergele and Ziquala districts

Grouping Toxoplasma gondii Coxiella burnetii Chlamydophila abortus
variable (n=18/176) (n=30/176) (n=5/176)
Sex Cases Controls Cases Controls Cases Controls
Male 1 4 14 3 1 1
Female 6 7 5 8 1 2
District Ziquala 6 9 4 4 0 1
Abergele 1 2 15 7 2 2
Total (n/N) 7/88 11/88 19/88 11/88 2/88 3/88

286
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Logistic regression analysis of the occurrence of organisms, regardless of case and control groups,
showed that there is no significant difference between sex groups and study districts. However, the
occurrence of Coxiella burnetii showed a substantial difference between the two-study districts (P =
0.017). Its occurrence was much higher in Abergele than Ziquala district. (Table 4)

Table 4: Association of seroprevalence with the difference in sex and district

Toxoplasma gondii Coxiella burnetii Chlamydophila abortus
P-Value COR 95%CI P-Value COR 95%CI P-Value COR 95%CI
Sex Female Ref - - Ref - - Ref - -
Male 0.55 0.72 0.24-2.12 0.34 0.65 0.27 - 1.58 0.77 1.29 0.21 -7.99
District Zequala Ref Ref. Ref.
Abergele 0.45 0.68 0.25-1.83 0.017* 2.90 1.21- 6.94 0.746 1.35 0.22 - 8.28
COR, crude odds ratio *showed significant difference at the p-value of 0.05

Brucella melitensis was not detected in this particular study area in both cases and control groups.
The finding of this study was in line with a study conducted in Menze and Horro small ruminants. In
these locations, the prevalence of small ruminant brucellosis was zero (Gebretensay et al., 2019).
However, the findings of this study disagree with reports elsewhere in regards to the prevalence of
this organism. Galgal et al. (2013) reported that the overall incidence of brucellosis in small
ruminants was to be 1.88% in the Yabello district. A meta-analysis report indicated that the overall
pooled prevalence of brucellosis in goats in Ethiopia was found to be 2.7% (Tadesse, 2016). This
variation can be attributed to the differences between study localities and production systems as well.

CONCLUSION AND RECOMMENDATION

From the data collected, it seems clear that the abortion problem in the enrolled villages is
complex and likely presents a mix of poor husbandry practices and the presence of infectious
diseases, which are confounded by harsh climatic conditions. To better understand the
underlying causes and understand the relative importance of the husbandry system and infectious
diseases, the following activities should be implemented.

• Conducting longitudinal study that confirms the cause of abortion with community
participation (with better household-level data collection strategies and microbiological
diagnosis)
• Improving the feeding and goat management practice (to avoid abortions related to feed and
climate stress)

287
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Al, A., Habib, I., Barnes, A. & Robertson, I. 2019. Veterinary Parasitology : Regional Studies
and Reports Risk factors associated with seropositivity to Toxoplasma among sheep and
goats in Northern Iraq. Veterinary Parasitology: Regional Studies and Reports,
15(September 2018), 100264. https://doi.org/10.1016/j.vprsr.2019.100264
Alilo, A. A., Beyene, A. & Mola, M. 2018. Advances in Dairy Research Assessment of Sheep
and Goat (Small Ruminants) Production System in, 6(4). https:// doi.org/ 10.4172/ 2329-
888X.1000215
Assefa, A. & Bahiru, A. 2018. Ethnoveterinary botanical survey of medicinal plants in
Abergelle, Sekota and Lalibela districts of Amhara region, Northern Ethiopia. Journal of
Ethnopharmacology, 213, 340–349. https://doi.org/10.1016/j.jep.2017.11.024
Bachan, M., Ranjan, A., Ranjan, B. & Sudhakar, N. R. 2018. Veterinary Parasitology : Regional
Studies and Reports High seroprevalence of Toxoplasma gondii in goats in Jharkhand state
of India, 12(July 2017), 61–68. https:// doi.org/10.1016/ j.vprsr. 2018.02.004
CSA. 2017. Federal democratic republic of Ethiopia central statistical agency: Agricultural
sample survey 2016/17 (Vol. II). Addis Ababa, Ethiopia.
Gebremedhin, E. Z., Agonafir, A., Tessema, T. S., Tilahun, G., Medhin, G., Vitale, M. & Di
Marco, V. 2013. Some risk factors for reproductive failures and contribution of Toxoplasma
gondii infection in sheep and goats of Central Ethiopia: A cross-sectional study. Research in
Veterinary Science, 95(3), 894–900. https:// doi.org/ 10.1016 /j.rvsc. 2013.08.007
Gebretensay, A., Alemayehu, G., Rekik, M., Alemu, B., Haile, A., Rischkowsky, B. & Wieland,
B. 2019. Risk factors for reproductive disorders and major infectious causes of abortion in
sheep in the highlands of Ethiopia. Small Ruminant Research, 177(October 2018), 1–9.
https://doi.org/10.1016/j.smallrumres.2019.05.019
Gokce, H. I., Kacar, C., Genc, O. & Sozmen, M. (2007). Seroprevalence of Chlamydophila
abortus in aborting ewes and dairy cattle in the north-east part of Turkey. Bulletin of the
Veterinary Institute in Pulawy, 51(1), 9–13.
Gupta, V. K., Verma, D. K., Singh, S. V. & Vihan, V. S. 2007. Serological diagnostic potential
of recombinant outer membrane protein ( Omp31 ) from Brucella melitensis in goat and
sheep brucellosis, 70, 260–266. https://doi.org/ 10.1016/ j. small rum res. 2006. 01.012
Hazlett, M. J., McDowall, R., DeLay, J., Stalker, M., McEwen, B., van Dreumel, T. & Cai, H. Y.
2013. A prospective study of sheep and goat abortion using real-time polymerase chain
reaction and cut point estimation shows Coxiella burnetii and Chlamydophila abortus
infection concurrently with other major pathogens. Journal of Veterinary Diagnostic
Investigation, 25(3), 359–368. https://doi.org/10.1177/1040638713484729
Klauck, V., Pazinato, R., Radavelli, W. M., Cust, E., Bianchi, A. E., Camillo, G. & Tonin, A. A.
2016. Microbial Pathogenesis Toxoplasma gondii infection in dairy ewes : Vertical
transmission and influence on milk production and reproductive performance, 99, 101–105.
https://doi.org/10.1016/j.micpath.2016.08.012

288
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Mohammed, M., Mindaye, S., Hailemariam, Z., Tamerat, N. & Muktar, Y. 2017. Sero-
Prevalence of Small Ruminant Brucellosis in Three Selected Districts of Somali Region ,
Eastern Ethiopia, 5(1), 1–6. https://doi.org/10.15744/2348-9790.5.105
Mohammed, O. B., Jarelnabi, A. A., Aljumaah, R. S., Mohammed, A., Bakhiet, A. O., Omer, S.
A. & Hussein, M. F. 2014. Coxiella burnetii, the causative agent of Q fever in Saudi
Arabia : molecular detection from camel and other domestic livestock. Asian Pacific Journal
of Tropical Medicine, 7(9), 715–719. https://doi.org/10.1016/S1995-7645(14)60122-X
Rodolakis, A. & Laroucau, K. 2015. Chlamydiaceae and chlamydial infections in sheep or goats.
Elsevier B.V. https://doi.org/10.1016/j.vetmic.2015.07.010
Swai, E. S. & Schoonman, L. 2010. The use of rose bengal plate test to assess cattle exposure to
Brucella infection in traditional and smallholder dairy production systems of Tanga Region
of Tanzania. Veterinary Medicine International, 2010. https://doi.org/10.4061/2010/837950
Szeredi, L., Jánosi, S., Tenk, M., Tekes, L., Bozsó, M., Deim, Z. & Molnár, T. 2006.
Epidemiological and pathological study on the causes of abortion in sheep and goats in
Hungary (1998-2005). Acta Veterinaria Hungarica, 54(4), 503–515. https://doi.org/
10.1556/AVet.54.2006.4.8
Tadesse, G. 2016. Brucellosis Seropositivity in Animals and Humans in Ethiopia: A Meta-
analysis. PLoS Neglected Tropical Diseases, 10(10), e0005006. https:// doi.org/ 10.1371/
journal.pntd.0005006
Thrusfield, M. & Christley, R. 2018. Veterinary Epidemiology (4th ed.). Oxford: Wiley.
Tilahun, B., Tolossa, Y. H., Tilahun, G., Ashenafi, H. & Shimelis, S. 2018. Seroprevalence and
Risk Factors of Toxoplasma gondii Infection among Domestic Ruminants in East Hararghe
Zone of Oromia Region, Ethiopia. Veterinary Medicine International, 2018. https:// doi. org/
10.1155/2018/4263470

289
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Evaluation of the Propagation Techniques of a Native Becium Grandiflorum in Wag-

Himra Zone, Ethiopia

Ayalew Girmay, Agazhe Tsegaye, Alemu Tsegaye and Meressa Lemma

Sekota Dryland Agricultural Research Center, P.O. Box 62, Sekota, Ethiopia

ABSTRACT

Becium grandiflorum is one of the major honey plants and sources of pollen and nectar for honey
bees in the eastern part of the Amhara region, especially in the Wag-himra zone. Nowadays the
plant has been found in depletion and extinction from its natural vegetation area due to human
interference, as they are using it for different cultural household tools. This study was conducted
at the Jinqaba apiary site of Sekota Dryland Agricultural Research center in 2017 - 2019 to
evaluate different propagation techniques of Becium grandiflorum and select the best
propagation method. The propagation techniques (treatments) were direct sowing, seedling,
cutting using pot, plant splitting, ground, and air layering. Each treatment was replicated three
times in a randomized complete block design. The treatments (planting materials) were planted
at 1.5m distance between plants, 1.5m distance between rows, and 2m distance between blocks.
All the young plants or seedlings deriving from the different propagation techniques were
planted in pits of 20cm in diameter and 20cm depth. There were significant differences (p<0.05)
among the propagation techniques in canopy cover, height, number of flowers, branches per
plant. Propagation of Becium grandiflorum from seedlings using pot is the best in attaining the
maximum number of 4427.8 flowers per plant, while sowing is the least technique with 1415.9
flowers per plant. Therefore, planting seedlings using pot was found to be the best method for
wider production of Becium grandiflorum under its natural growing areas.

Keywords: Becium grandiflorum, bee flora, flower number, honey bees, treatments

290
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Becium grandiflorum is one of the endemic plants to the highlands of Ethiopia and Eritrea. It is a
medium-sized, aromatic woody shrub that belongs to the family of Lamiaceae. The species
grows on eroded soils, particularly in rocky slopes and sandy soil, in mountain bush land and
pastures. The species spans over altitude ranges of 1600 and 3100 m.a.s.l (Vivero et al., 2006).

In the Tigray region, B.grandiflorum is the only highly abundant and most frequently available
bee flora species in the dry season (Niguse, 2015). The plant also ranked as the best honey bee
forage by beekeepers (Equar et al., 2016). Honey bee flora inventory study in the Amhara region,
Wag-himra zone also showed as one of the top 10 abundant major bee forage species (Abebe and
Alemu, 2016). In the area, the species is also the most important source of honeybee forage in
terms of preference by honeybees and abundance (Alemu et al., 2017).

Honey bees visit flowers of the plant for collecting pollen and/or nectar. The color of the honey
is creamy white and granulates rapidly. Because of its attractive color and also light to taste, the
honey is preferred by many consumers. The species has also added value as fuel wood, broom
for cleaning threshing ground, roofing of traditional houses, food flavoring, traditional medicine
against malaria, and for soil and water conservation (Taddele and Nejdan, 2008).

Nowadays honey bees are exposed to a serious bee flora shortage which is mainly caused by
colony absconding. The possible reasons behind the declining population of different bee flora
types in the area are due to population pressure and farmland expansion, as well as use for a
variety of cultural household tools. On top of this, increased demand for fuel wood has led to
intensified extraction of mature plants from their natural habitats (Fikadu and Tilaye, 2017).

As a result, natural regeneration of the species from seeds has become very difficult due to
widespread human interference and also due to climate change, thus further plantation is needed
to this promising plant with appropriate propagation technique. Therefore, this activity was
initiated to evaluation the possible propagation techniques of B. grandiflorum and select the best
method for wider production of the plant in Wag-Himra zone.

291
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIAL AND METHODS

Description of the Study Area

Agro-ecologically, the study area (Wag-Himra zone) is categorized into lowland (63.3%)
midland (31.43%) and highland (5.54%) with a total area of 878,195.3 hectares. Regarding the
topography of the study area, it is located at an elevation between 500 and 3500 m.a.s.l. And the
annual rainfall distribution varies between 150 to 700 mm, which is an erratic type of rainfall. Its
temperature ranges from 15 - 40°C (BoFED, 2007).

In the Wag-himra zone, there is a huge potential of beekeeping, which is an integral part of the
animal husbandry though it has not been well exploited yet. Moreover, the forest and bush cover
is concentrated in the utmost specific areas most of which are communally owned or possessed
by churches. On another hand, the topography dominated by a number of deep gorges, ups and
downs and series of the rugged massif (uneven mountains) that creates good opportunities for
beekeeping practice (Alemu et al., 2017).

Seeds and Mother Plant Collection

B. grandiflorum mother plants with the same provenance were selected from Jinqaba watershed
to be used as a source of planting material before the actual experiment was conducted. Seeds
were collected from selecting mature fruits and packed in perforated polyethylene bags and
allowed to dry for one month at room temperature. Then seeds were taken out of these packages,
for propagating the plant in plastic pots and direct sowing. The planting materials were collected
at the beginning of the rainy season and transplanted within the same day.

Experimental Management

Treatments (propagation techniques) that were applied in the experiment include direct sowing,
seedling, cutting using pot, plant splitting, air layering, and ground layering.

• Seeds were sowed in rows in the experimental field. To keep proper spacing and avoid
nutrient competition, thinning of seedlings was done on the 30th day from the time of sowing.

292
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

• For seedling techniques, plastic sleeves (diameter, 8; length, 15cm) were prepared at Wolh
nursery site. Dried grass cover over the ordered plastic sleeves was done to conserve water.
Watering was applied regularly for about 45 day’s period which was the end of the nursery
age of the seedlings.
• For cutting using pot, 60 cuttings with a height of 40cm were collected early in the morning
from naturally matured young and healthy mother plants. To avoid the direct entrance of
water during planting and growing period, the cuttings were prepared in such a way as the
top ends have a slant surface (angle of 45°) vertically, and contain a minimum of two nodes.
Finally, transplantation of the seedlings to the experimental plot was done at 55 days at a
survival rate of 75%. Then after, cuttings were inserted in plastic sleeves (diameter, 8; length,
15cm) and well managed for days in the nursery to use on the main experimental field.
• For air layering, the stem was enclosed via a moist rooting medium soil at the point of its
node. Then plastic wrap and proper sealing were done to keep the layer from drying out and
to retain the moisture around the node area for 25 days. Finally, the simple cuttings of the
branch were done close to the new plant.
• For ground layering to raise new plants from layering each stem was attached to the parent
plant by bending selected branches. The middle portion of the selected branches was put in
the ground under the existing soil. The stem bend to the ground was covered part of it with
the above soil mixture and the rest was left exposed. Cutting of the stem was done just below
the rooted zone on the 25th day. Lastly, the seedlings were transplanted to the experimental
plot directly after cutting the branches from their mother plants.
• For plant splitting, mother plants that had more young branches were selected from the
established mother plants. Then split the branches easily from their mother plants without
injuring their roots.

Each propagation technique was considered as a treatment. Each treatment was replicated three
times in three blocks. The newly raised planting materials (treatments) were planted at 1.5m
distance between plants, 1.5m distance between rows, 2m distance between blocks. All the
young plants or seedlings deriving from the different propagation techniques were planted in pits
of 20cm in diameter and depth.

293
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Photo1: Seedlings Photo 2: Ground Layering

Photo 3: Cutting with Pot Photo 4: Air Layering

Data Collection

To collect the data three plants were taken randomly per plot and a total of nine plants per each
treatment were used. The data were collected on the following parameters.

• The height of the plant was measured from ground level to the tip of the largest branch with the
help of a measuring tape. The crown depth of the plant was measured from the first branched
area to the tip of the largest branch with the help of a measuring tape. The number of branches
per plant was determined by counting.
• Canopy cover was calculated by CC = (D1 +D2)/2. Where D1 is the diameter of the plant toward
the larger canopy coverage, D2 is the diameter of the plant towards the small canopy coverage
and CC represents canopy cover of the plant in cm. And then the canopy cover of the plant was
expressed in terms of the average diameter in cm.
• The total numbers of flowers (TF) per plant were calculated by counting the total head flower per
plant and the number of flowers per head flower. The total number of flowers per plant was
calculated by TF = H.F *NFH, Where HF is the number of head flowers per plant; NFH is the
number of flowers per head flowers. To count the number of flowers per head flower, a sample of
10 head flowers per plant was taken randomly.
• Survival rate was calculated by SR = PS*100/PP, Where SR = survival rate, PS = number of
plants survive, PP = number of plants planted.

294
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Analysis

The collected data were statistically analyzed using one-way ANOVA analysis of variance and
least significant difference was calculated to identify significant difference among the treatments
using SAS (version 20) statistical software. Correlation analysis was also applied for specific
variables

RESULTS AND DISCUSSION

The study revealed that B.grandiflorum could able propagated in six successful propagation
techniques, which were direct sowing, seedling, plant splitting, cutting using pot, air layering,
and ground layering. Similar propagation techniques for B. grandiflorum had reported in Tigray
(Haftom and Kebebew, 2013). Comparatively, splitting methods had the probability of mother
plant damage than the other methods. It is because plant root splitting destroys separating of the
new split from the mother plant in doing propagation.

Differences (p<0.00) were found among the treatments in plant height (Table 1). The plant
height ranges from 77.6m (cuttings) to 118.3m (seedling). Similarly, Haftom and Kebebew
(2013) stated that the height of the plant ranges from 104m (cuttings) to 116m (splitting). There
was a significant (p<0.01) difference among techniques in the number of branches per plant
(Table 1). Plants propagated through cutting had the highest number of branches (21), while air
layering had the smallest branches (10.3) per plant. Haftom and Kebebew (2013) stated that
ground layering had the highest number of branches per plant (12) whereas seedlings had the
smallest number of branches (7) per plant.

The study showed a significant difference in the canopy cover of the plant among the techniques
(p<0.01). The difference in canopy cover between the treatments might be due to the difference
in the number of branches. Canopy cover and number of branches per plant had a significant
positive correlation (r = +0.82, P<0.001) (Table 2). Similarly, Haftom and Tesfay (2014a)
reported the canopy coverage of seedlings was the highest (121.2cm), while air layering was the
smallest (77.7cm).

295
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Differences were also found in flowers number per plant (p<0.001). Plants propagated through
seedling had the highest number of flowers per plant (4427.8), whereas plants propagated
through air layering had the smallest flower number (1415.9). Differently, Haftom and Kebebew
(2013) stated that plants propagated from cuttings had the highest number of flowers (12,910),
whereas plants propagated through sowing had the smallest number of flowers (5637) per plant.
The variation in flowers number may be due to the number of branches per plant.

Table1: Mean height, crown depth, number of branches, canopy cover, root collar diameter,
flower number and survival rate of different propagation techniques
Parameters
Techniques H (cm) CD (cm) NB CC (cm) RCD (cm) FN SR
b
Direct sowing 97.8 93.4b 18ab 94bc 1.7b 3066c 92.6
a a
Seedlings 118.3 115.6 20a 121.2a 1.8b 4427.8a89
Splitting 89.7c 83.2c 13.2bc 94bc 2.4a 3066c 89
d d
Cuttings 77.6 74 21a 110ab 1.8b 3952.5b85.2
Air layering 91.2c 88bc 10.3c 77.7c 1.1c 1416e 70.4
c cd
Ground layering 86.1 79.4 13.4bc 86c 1.5b 1805.5d74.1
Sign (5%) *** *** ** ** *** *** NS
CV 3.7 5.6 17 9.8 12.4 5.8 13.6
*** and ** = significant at P<0.001 and p<0.01, respectively; NS = non significant, CV =
coefficient of variation. Mean values followed by the same letters are not significant difference,
H = plant height, CD = crown depth, NB = no of branches, CC = canopy cover, RCD = root
collar diameter, FN = no of lowers and SR = Survival rate in %.

Correlation Analysis

Height of the plant had none significant weak positive correlation with number of branches per
plant (r = +0.16, p>0.05), canopy cover (r = +0.38, p>0.05) and flower number per plant (r =
+0.36, p>0.05) (Table 2). This might be due to the growth habit of the plant. Branches orient
horizontally than growing upward and more flower biomass was found from the branched plants.
Haftom and Kebebew (2013) reported that the flower number of B. grandiflorum had a negative
correlation with the height of the plant. Similarly, it was reported earlier by Haftom and Tesfay
(2014b).

296
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Branches of a plant significantly positively correlated with canopy cover (r = +0.82, p<0.001)
and flower number (r = +0.78, p<0.001). Similarly, Haftom and Kebebew (2013) declared that
the plant branch had a positive correlation with canopy cover and number flowers of a plant. The
same authors also reported a positive correlation between the number of flowers and the number
of branches on Hypoestes forskaolii.

Table 2: Correlation coefficient(r) between variables

FN
RCD (cm) 0.58*
CC (cm) 0.76*** 0.21
NB 0.78*** 0.22 0.82***
CD (cm) 0.37 0.09 0.41 0.19
H (cm) 0.36 0.03 0.38 0.16 0.98***
FN RCD (cm) CC (cm) NB CD (cm) H (cm)
*** and ** = significant at P<0.001 and p<0.01, respectively, H = plant height, CD = crown
depth, NB = no of branches, CC = canopy cover, RCD = root collar diameter and FN = no of
lowers

CONCLUSION AND RECOMMENDATIONS

This study indicated that B. grandiflorum can be propagated through direct sowing, seedling,
cuttings, splitting, ground and air layering. These different propagation techniques will allow
multiplying of the plant in different growing areas using the skill and areas of application
acquired. B.grandiflorum propagated through seedling had the highest canopy coverage and
flower biomass, while plant propagated through cutting with pot had the highest number of
branches followed by a seedling method. The highest plant height found in plants propagated
through seedlings, whereas the smallest was in the cutting method. Besides, plant height
negatively correlated with plant branch number, canopy cover, and flower biomass. Therefore,
among these techniques, for broader production of the plant, the seedling method is
recommended to use as a best propagation technique of the plant under its natural growing areas,
because of its flower number and respective correlated parameters.

297
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

REFERENCES

Abebe Jenberie, Alemu Tsegaye and Addisu Bihonegn. 2016. Identification and Characterization of
Honeybee Flora in Wag-Lasta Area. In Solomon Abegaz and Likawent Yeheyis (Eds). 2016.
Proceeding of the 9th Annual Regional Conference on Livestock Completed Research Activities
9-20 March, 2015, Amhara Agricultural Research Institute, Bahir Dar, Ethiopia.
Alemu T., Seifu E. and Bezabih A. 2017. Honeybee production practices in Sekota district , northern
Ethiopia This article is licensed under a Creative Commons license , Attribution 4 . 0
International (CC BY 4.0) Beekeeping is a sustainable form of agriculture, which is beneficial to
the e. (May 2016), 263–275.
BoFED. 2007. GIS team of Amhara National Regional State Bureau of Agriculture and Economic
Development. Working Paper. 55–60.
Equar G., Berhanu A., Hayal L., Solomon A. and Tsehaye A. 2016. Honey bee flora diversity and
their impact on honey production in Tigray region of Ethiopia. Livestock Research for Rural
Development, Vol. 28.
Fikadu A. A. and Tilaye K. A. 2017. Adoption and intensity of use of modern beehives in Wag
Himra and north Wollo zones, Amhara region, Ethiopia. Ethiopian Journal of Economics, 26(2),
1–30.
Haftom G. and Kebebew W. 2013. Identification and evaluation of propagation techniques of a
native bee forage, “tebeb” (becium grandflorum). Livestock Research for Rural Development,
25(1), 1–7.
Haftom G. and Tesfay B. 2014a. Determining suitable size of stem cutting for propagating Becium
grandiflorum. 24(10), 1–6.
Haftom G. and Tesfay B. 2014b. Effect of pruning on vegetative and flowering patterns of Becium
grandiflorum. 24(12), 2010–2015.
Haftom G. and Yaynishet T. 2012. Identification and evaluation propagation techniques of Hypoestes
forskaolii (Grbia) as bee fodder for smallholder farmers. Livestock Research for Rural
Development, 24(9), 18.
Niguse G. 2015. Assessment of Honey Bee Flora in Eastern Zone of Tigray, North. 5(1), 53–57.
Taddele and Nejdan. 2008. Beekeeping and honey production in Tigray region, Ethiopia.
Vivero, J. L., Kelbessa, E. and Demissew, S. 2006. The Red List of Endemic Trees & Shrubs of
Ethiopia and Eritrea. 28. https://doi.org/10.13140/2.1.3139.6168

298
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Evaluation and Characterization of Physico-Chemical Quality Parameters of Stingless Bee

(Apidae Meliponini) Honey in Amhara Region

Yesuf Ebrahim1, Alemu Tsegaye1 and Engdaw Malede2

1
Sekota Dryland Agricultural Research Center, P.O. Box 62, Sekota, Ethiopia
2
Andassa Livestock Research Center, P.O. Box 27, Bahir Dar, Ethiopia

ABSTRACT

This study aimed to evaluate the physicochemical parameters of honey samples of stingless bees
(Meliponinae). Following the honey flow season, the honey samples collected from 9 potential
districts in the Amhara region. About 41 honey samples evaluated for physicochemical
parameters of pH, free acidity, electrical conductivity, hydroxymethylfurfural, moisture content,
specific rotation, ash content, color, reducing sugars, and sucrose. The Melissopalynological and
sensory analysis had been used as a complement to physicochemical analysis. The composition
analysis showed that the mean values of moisture content were 29.69±3.53; electric conductivity,
0.78±0.14; pH, 3.38±0.19; ash, 0.5±0.17; free acidity, 67.47±23.85; HMF, 8.38±4.47; specific
rotation, 2.24±4.79; total reducing sugar, 55.27±4.24 and Sucrose, 3.17±1.5. The color of honey
ranges from amber (97.6%) to dark amber (2.4%). The values compared against the standard
Apis mellifera honey quality parameters and varied for some parameters based on botanical
origin. The parameters pH, electric conductivity, sucrose, ash and HMF in the honey samples
comply with the requirements of the previous global reports. The moisture content, reducing
sugar, and free acidity do not comply with the limit compared to Apis mellifera honey, implying
that this product can undergo fermentation quickly if not properly stored after harvesting. This
fact showed the necessity of proper honey harvesting, creating specific legislation for stingless
bee honey, and justifies the need for a more harmonized standard of the product. The sensory
evaluation result also revealed that on average the judges slightly like the stingless bee honey,
which is slightly above the neutral score 5 (neither like nor dislike). The result of the
melissopalynological analysis also confirmed the presence of three predominant plant species
and important pollen types that confirm the honey botanical and its real geographical origin.

Keywords: Honey, physicochemical parameters, Quality standards, Stingless bee

299
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Stingless bees are highly social (Costa et al., 2003), and they are the smallest of the honey-
producing bees and belong to the family Apidae and subfamily Meliponinae (Anaraddi et al.,
2011). Mainly found in the tropical and subtropical areas (Martins et al., 2004). They are
considered as effective pollinators of a wide range of wild crops (Slaa et al., 2006). They are
locally called as “Tazima Nib” and are uniquely identified in Ethiopia, living by harboring
underground nests.

The composition of stingless bee honey differs from other species with some physicochemical
parameters (Özbalci et al., 2013). Honey from stingless bees is more valuable, and it has been
used for a long time to treat various diseases (Souza et al., 2006). Though the amount of
meliponinae honey is smaller than Apis mellifera honey, its medicinal importance makes it
attractive and fetches higher market values to it (Santos et al., 2015).

However, the method of harvesting honey from feral colonies in the country is traditional and
destructive, resulting in low quality of honey. Global and national honey quality standards were
established only for A. mellifera honey by following the guidelines of international standards of
the Codex Alimentarius Commission (CODEx, 2001). However, due to the scant knowledge
about the product, meliponine honey is not included in the international standards for honey
(CODEx, 2001), and it is not regulated by the food control authorities. Thus, there is no
assurance for consumers.

Since the aim of the International Honey Commissions is the establishment of quality standards
of bee products other than Apis mellifera honey, stingless bee honey must be considered.
Because of the diversity of Meliponinae species and insufficient information regarding the
characteristics of the honey produced by this bee species, determining the physical-chemical
parameters of the honey is crucial for the process of creating specific legislation for stingless bee
honey in Ethiopia. This study, therefore, aimed to evaluate and characterize the physicochemical
quality of honey produced by the stingless bees (Meliponinae) in the Amhara region.

300
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIAL AND METHODS

The study was conducted from November 2018 to February 2019. Nine districts (Sekota,
Dehana, Lasta, Tilili, Dangla, Dembecha, Banja, Bure, and Guangua) from the Amhara region
were selected based on their potential. Sample kebeles were selected purposely based on their
production potential, agro-ecology, and vegetation cover. A total of 41 stingless bee honey
(called Tazma-mar in Amharic) samples were collected directly from local beekeepers (hunters).
250g of stingless bee honey was collected from each local beekeeper and kept in the refrigerator
(8°C). The physicochemical analysis was performed in the Sekota Dryland Agricultural Research
Center bee products laboratory within a month after the samples were collected.

Determination of Moisture Content

Determination of the refractive index of the honey sample using a refractometer at a constant
temperature near 20oC was adopted by IHC. Conversion of the reading to moisture content
(percent by mass) made using a formula or a conversion table. If the determination is made
different from a temperature of 20oC, the reading converted to a standard of 20oC, according to
the temperature corrections quoted.

Determination of pH and Acidity

The pH and acidity were determined following the methodology adopted by Moraes and
Teixeira. The pH value was determined using a solution containing 10 g of honey dissolved in
75ml of distilled water, homogenized and subjected to reading in a pH meter.

Acidity was obtained by performing the neutralization of acidic solution of honey (10g of honey
dissolved in 75ml of distilled water) using a sodium hydroxide solution 0.1N and 1% of
phenolphthalein indicator solution until a pink color was obtained for 10 seconds. The reading of
the sodium hydroxide volume 0.1N required in the titration was recorded.

The result is expressed in meq kg-1 using the equation:

Acidity = V (NAOH) x PA,

Where: V (NAOH) is the volume of NAOH (ml) and PA is the sample weight (g)

301
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Determination of Electrical Conductivity

The electrical conductivity of a solution of 20 g dry matter of honey in 100 ml distilled water
was measured using an electrical conductivity cell. The determination of the electrical
conductivity is based on the measurement of the electrical resistance, of which the electrical
conductivity is the reciprocal. The method is based on the original work of Vorwohl and IHC.
The electrical conductivity was measured using a conductivity meter, and the result was
expressed in milliSiemens per centimeter (mS.cm-1) for each sample.

Determination of Hydroxymethylfurfural (HMF)

The determination of the hydroxymethylfurfural (HMF) content was based on the determination
of UV absorbance of HMF at 284 nm. The HMF content was calculated after subtraction of the
background absorbance at 336 nm. The determination of HMF was based on the readings in
different absorbance scales (284 and 336 nm wavelengths) in a spectrophotometer.

The HMF was expressed in mg kg-1 in the equation:

HMF = (A284 – A336) x 149.7 x 5 x D/W,

Where: A284 is the absorbance at 284 nm, A336 is the absorbance at 336 nm; D is the dilution
factor, if necessary and W is the weight of honey sample (g)

Determination of Specific Rotation

The specific rotation [α] 20 D is the angle of rotation of polarized light at the wavelength of the
sodium D line at 20°C of an aqueous solution of 1 dm depth and containing 1g/ml of the
substance. The method can be applied to all honey samples. In particular, most of the honeydew
samples have positive values of specific rotation whereas nectar honeys have negative values. A
clear, filtered solution of honey sample having carrez solutions, were immersed into a 2-dm
polarimeter tube and read the angular rotation by polarimeter.

302
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Determination of Ash Content

The ash content of honey means the residue which is obtained by a defined procedure and
expressed as a percentage by weight. The honey samples were entered in to the muffle furnace
and ashed at a temperature no higher than 600°C and the residue weighed and ash content was
determined by the equation:

m3 − m1
Wa = × 100
m2

Where: Wa is the ash in g/100g honey, m1 is the weight of dish, m2 is the weight of honey taken
and m3 is the weight of dish + ash

Reducing Sugar and Apparent Sucrose Determination

This method is a modification of the Lane and Eynon procedure, involving the reduction of
Soxhlet’s modification of Fehling’s solution by titration at boiling point against a solution of
reducing sugars in honey using methylene blue as an internal indicator. The difference in
concentrations of invert sugar is multiplied by 0.95 to give the apparent sucrose content. The
reducing sugar and sucrose content of the stingless bee honey samples were analyzed using this
method, which is also used in the Codex alimentarius standard procedure.

Pollen Analysis

The samples were subjected to qualitative and quantitative pollen analysis following the
methodology recommended by the International Commission for Bee Botany (ICBB) (Louveaux
et al., 1978). The pollen types recovered and identified were placed under four frequency classes
as mentioned below.

Predominant pollen type: More than 45% of the total pollen grains counted.
Secondary pollen type: Between 16 and 45% of the total pollen grains counted.
Important minor pollen type: Between 3 and 15% of the total pollen grains counted.
Minor pollen type: Less than 3% of the pollen grains counted.

Colour

The color of the honey samples was characterized using an instrument called Lovibond
comparator 2000 + instrument

303
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Physical Characterization Using Sensory Analysis

The sensory testing panel consists of 12 selected and trained people. Testing was performed in a
sensory laboratory under sunlight using thirty grams of honey sample per person. Sensor
evaluation results of each of the samples were registered on an individual basis. Each term was
evaluated using a scale ranging from one to nine (extremely dislike, strongly dislike, moderately
dislike, slightly dislike, neutral, slightly like, moderately like, strongly like, extremely like).

Statistical Analyses

Results were reported as the mean ± standard deviation of duplicate experiments. Using the
SPSS statistical package (version 23), ANOVA and post hoc multi-comparison test were used for
comparison of means (p<0.05). Several parameters were correlated using Pearson’s correlation
coefficient (r) (p<0.01).

RESULTS AND DISCUSSION

Moisture Content

The average moisture content of the stingless bee honey measured for collected honey samples
was 29.69±3.53% (Table 1). This result agrees with the moisture content (29%) of stingless bee
honey from West Shoa zone of Oromia region (Alemayehu et al., 2018). Global published
reports also indicate that moisture content of global stingless bee honey could be between 19.9
and 41.9 (Souza, 2006). The sample of honey analyzed by Nascimento et al. (2015) also showed
the average moisture ranged from 25.99% (M. bicolor) to 36.89% (M. quadrifasciata) in
Brazil.The moisture content reported in stingless bees could go from as low as 13.26 g/100 g to
as high as 45.8 g/100 g, with a mean of 28.6 g/100 g and a standard deviation of 5.7 g/100 g
(Nordin et al., 2018).

However, Vit et al. (2004) findings showed a slightly higher threshold of 30g/100g in
Guatemala, Mexico and Venezuela. According to this, the mean moisture content value of
stingless bee honey of the Amhara Region lies within the reported moisture content range. But
the moisture content exceeds the maximum permissible limit for Apis honey (20%). Stingless
bee honey has been reported to contain greater water content compared to Apis mellifera honey.

304
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

pH and Free Acidity

Honey is acidic in nature, with a pH ranging from 3.2 to 4.5 (Solayman et al., 2016). The mean
pH value of the collected honey samples was 3.38±0.19 for all samples collected from the region
(Table 1). This result agrees with the pH value (3.73) of stingless bee honey from West Shoa
zone (Alemayehu et al., 2018). Global published reports also indicate that pH of global stingless
bee honey could be between 3.15 and 4.66 (Souza, 2006). Nordin et al. (2018) also reported
overall the pH of stingless bee honey ranges from 3.15 to 6.64 with the mean of 3.85. According
to this Author also, the lowest pH value 3.15 was detected from the honey of Melipona
Scutellaris and the highest pH value was detected from the honey of Melipona quadrifasciata
from Brazil (Carvalho et al., 2009). Therefore, the mean pH value of stingless bee honey of the
study areas lies within the reported moisture content range. There was a significant difference
(p>0.05) in pH between honey samples obtained from different sampled locations (Table 1). This
variation might be due to the variations in vegetation sources and harvesting practices.

The average acidity of honey samples collected from the study area was 67.47±23.85 (Table 1).
The result revealed no significant difference (p>0.05) in free acidity between honey samples
obtained from different sampled locations. This result is higher than the free acidity value (57.3)
of stingless bee honey from the West Shoa zone (Alemayehu et al., 2018). Global published
reports indicate that free acidity of global stingless bee honey could vary from 5.9 to 109 (Souza,
2006). According to this, the mean free acidity value of stingless bee honey of the study areas
lies within the reported free acidity range. The honey analyzed by Nascimento et al. (2015) also
showed an average acidity ranging from 22.55 to 48.58 meq kg-1 in Brazil. Brazil establishes a
maximum of 50 meq kg. The threshold set for Scaptotrigona species by Vit et al. (2004), is more
tolerant at 85 meq/kg. In addition, a high value of total acidity may imply that at some point the
honey began to ferment and that the produced alcohol was transformed into organic acids.

Electrical Conductivity

The average electrical conductivity value was 0.78±0.14 (Table 1). This result is slightly higher
than the EC value (0.22) of stingless bee honey from West Shoa zone of Oromia region
(Alemayehu et al., 2018).

305
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Global published reports indicated that EC value of stingless bee honey could vary from 0.49 to
8.77 (Souza, 2006). The IHC recommended that EC should not be more than 0.8ms/cm.
According to this, the mean EC value of stingless bee honey of the Amhara Region lies within
the reported free acidity range.Honey from the different localities also show significant
difference in their EC content (p>0.05) (Table 1).This variation might be due to the variations in
vegetation sources and mineral content.

Ash Content

The average value of ash content was 0.5±0.17 (Table 1). The result is comparable with the ash
content (0.41) of stingless bee honey from West Shoa zone (Alemayehu et al., 2018). Global
published reports indicated that ash value of global stingless bee honey could vary from 0.01-
1.18 (Souza, 2006). In agreement with this result, Nordin et al. (2018) reviewed that the ash
content of stingless bee honey ranging from 0.01g/100 g (Melipona scutellaris from Brazil) to as
high as 3.1g/100 g of honey (Tetrigonamelanoleuca from Thailand) with the mean value of
0.4g/100g. Both the IHC and Vit et al. (2004) set the ash content threshold at not more than 0.5
g/100 g for a good quality honey from both Apis mellifera and Meliponinae species. Therefore,
the mean ash content of stingless bee honey lies within the reported ash content range. Honey
from the different localities also show significant differences in their ash content (p>0.05). This
variation might be due to the variations in vegetation sources and mineral content.

HMF

The average HMF contents of collected honey samples were determined to be 8.38±4.47mg/kg
(Table1). This result is much lower than the HMF content (18) of stingless bee honey from West
Shoa zone (Alemayehu et al., 2018). Though there is no Ethiopian standard for HMF value of
stingless bee honey, published reports indicate that HMF value of global stingless bee honey
could vary from 0.4 - 78.4 (Souza, 2006). HMF content has been set to be not more than 40
mg/kg in a good quality Apis mellifera honey. According to this, the mean moisture content
value of stingless bee honey of the Amhara Region lies within the reported moisture content
range.

306
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

In reality, though the naturally higher moisture content of Stingless bee honeys directly
influences their HMF content, stingless bee honeys have been reported to be more resistant to
HMF formation than Apis mellifera honey when subjected to heat treatment (Nordin et al.,
2018). Honey from the different localities also show significant differences in their HMF content
(p>0.05) (Table 1). This variation might be due to the variations in storage practices, depending
on the pH and storage temperature.

Total Reducing Sugar and Sucrose

The mean reducing sugars content of the study areas honey is 55.27±4.24% (Table 1). This
indicates that the reducing sugars content of honey of the study area is lower than the minimum
limits (58%) reported by Souza (2006) for global stingless bee honey. According to the standard
set by IHC, a good quality honey should have reduced sugar not less than 60g/100g in the case of
A.mellifera honey. Stingless bees have been reported to contain lower sugar content compared to
Apis mellifera honey (Nordin et al., 2018). The low reducing sugars content of the study areas
honey may be attributed to its high moisture contents. Honey from the different localities also
show significant differences in their total reducing sugar content (p>0.05) (Table 1). This
variation might be attributed to the variation in moisture content.

Adulteration of honey can be detected by measuring its sucrose content. The mean sucrose
content of the study areas honey is 3.17±1.5% (Table 1). Though there is no Ethiopian standard
for sucrose content of stingless bee honey, published reports indicate that Sucrose content of
global stingless bee honey could vary from 1.1 to 4.8 (Souza, 2006). According to the standards
set by the IHC, a good quality honey should have a sucrose content that is not more than 5
g/100g. Vit et al. (2004) has increased the threshold for sucrose a little higher at 6 g/100 g to
accommodate the slightly higher sucrose content reported in stingless bee honey. For Malaysian
standards, the maximum threshold of 7.5 g/100 g sucrose was set to regulate honey adulteration.
Therefore, the mean sucrose content of stingless bee honey of the Amhara Region lies within the
reported range. The amount of sucrose in the honey samples obtained from the different
localities also did not show significant difference (p>0.05) (Table 1).

307
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Specific Rotation

This property is related to the concentration of levorotatory (as fructose) and dextrorotatory (as
glucose) compounds. The average specific rotation value of the collected samples was 2.24
(Table 1), so they are predominantly dextrorotary showing that it becomes a potential
differentiation criterion from Apis mellifera of Ethiopia as these honeys always retain a
levorotatory character. From 41 honey samples analyzed for their specific rotation the majority
of the samples 25/41 (60.98%) showed a positive value. This indicates that the presence of
sugars and other compounds with rotation capacitySimilar with the value (2.3) reported by
Fuenmayor et al. (2012) for Vensuelan stingless bee honey. Honey from the different localities
also show significant differences in their specific rotation (p>0.05) (Table 1). This variation
might be attributed to the variations in vegetation sources, sugar and mineral content.

Correlation among Some Physicochemical Parameters

The correlation matrices showed significant correlations between some of the physicochemical
parameters. In stingless bee honey samples, a strong negative correlation was found between EC
and some parameters (HMF and total reducing sugar: Table 2). Also, correlation matrices of the
honey samples showed Positive correlations between EC and Ash content. Ash and electrical
conductivity values depend on the mineral content of the honey: ash gives a direct measure of
inorganic residue after carbonisation, while electric conductivity measures all ionisable organic
and inorganic substances (Sanchez et al., 2010).

The pH value had negative correlations with total reducing sugar and sucrose content (Table 2).
Strong positive correlation was established between HMF and reducing sugar while HMF had
negatively correlated with specific rotation (Table 2).

308
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 1: Mean values of physicochemical parameters result of sampling areas

Total
Electrical Moisture Specific Reducing
Location pH Conductivity Content HMF Ash Free acidity Rotation Sugar Sucrose
Sekota 3.25±0.16bc 0.76±0.53bc 25.34±0.96c 6.13±0.67b 0.59±0.11ab 55.52±11.68 0.71±2.56cd 58.58±2.04a 3.69±2.54
Dehana 3.45±0.19abc 0.7±0.27c 27.76±3.5ab 13.3±2.9a 0.62±0.14ab 55.49±15.25 -2.3±2.56d 59.87±0.44a 3.53±1.97
Lalibela 3.23±0.16c 0.67±0.12c 30.85±3.87a 14.48±3.69a 0.35±0.14cd 75.57±47.25 -1.4±5.11d 59.16±1.74a 3.86±1.29
Tillili 3.54±0.19ab 0.84±0.16abc 30.97±1.84a 4.45±1.38b 0.57±0.14abc 72.22±12.85 6.31±1.81ab 52.41±1.98bc 2.35±0.5
Dembecha 3.41±0.21abc 0.97±0.11a 30.0±2.83ab 6.28±0.21b 0.68±0.29a 72.18±7.06 9.73±0.66a 48.68±1.5d 3.23±0.7
Dangla 3.43±0.7abc 0.9±0.1ab 31.43±2.03a 6.11±1.39b 0.41±0.07bcd 69.56±3.26 5.8±2.3ab 49.99±0.73cd 3.65±0.77
Banja 3.34±0.36bc 0.78±0.06abc 32.75±4.31a 7.48±2.53b 0.56±0.21abc 71.75±11.59 4.46±4.37bc 53.8±2.05b 1.97±0.68
Bure 3.67±0.29a 0.95±0.01a 28.65±1.62ab 5.09±0.63b 0.5±0.07abcd 63.58±0.56 6.9±0.18ab 52.23±1.34bc 2.31±0.33
Guangua 3.38±0.06abc 0.69±0.09c 32.7±4.39a 6.71±1.19b 0.3±0.07d 74.9±25.4 -1.97±1.49d 51.56±1.23bc 2.38±0.11
Total 3.38±0.19 0.78±0.14 29.69±3.53 8.38±4.47 0.5±0.17 67.47±23.85 2.24±4.79 55.27±4.24 3.17±1.5
Proposed quality standard
(stingless bee honey) 3.15-4.5 0.49-8.77 19.9-41.9 0.4-78.4 0.01-1.18 5.9-109 58-75.5 1.1-4.8
National acceptable limit
(A.mellifera honey) 3.5-4.5 0.8 Max 21 Max 40 0.6 Max 40 -ve Min 65 Max 5

Table 2: Correlation analysis of physicochemical parameters

Moisture Free Specific Reducing
PH EC Content HMF ASH Acidity Rotation Sugar Sucrose
PH **
EC .229 **
Moisture Content -.142 .077 **
HMF -.256 -.503** .033 **
ASH .286 .337* -.329* -.299 **
Free Acidity -.225 .002 .492** -.111 -.019 **
Specific Rotation .251 .684** .337* -.639** .369* .333* **
Reducing Sugar -.352* -.504** -.429** .628** -.015 -.218 -.696** **
Sucrose -.490** -.119 .003 .227 -.153 .038 -.135 .349* **

309
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Color Analysis

Honey is often judged according to its colour. Colour and brightness contributes to the aesthetic
value of the honey, which is important to the consumer. Besides, colour is very important,
because this plays an important role to determine its market value. The colour of 97.6% of honey
samples range from amber to dark amber. All the colour of 100% of honey samples from the
midland was amber. Agree with Nascimento et al. (2015) who reported that the color of the
honey samples ranged from light amber (0.302 nm) to dark amber (1.225 nm) in Brazil.

Table 3: Summary of Honey colours from P-Fund readings

Agroecology and their Percentage
Colour Determined Pfund Scale Midland Highland Total
Extra white 9-17mm
White 18-34mm
Extra light amber 35-48mm
Light amber 49-83mm
Amber 84-114mm 100 94.1 97.6
Dark amber Greater than 114 5.9 2.4

Sensory Evaluation

Sensory analysis of honey may be used as a complement to physico-chemical and pollen

analyses. It is used to confirm quality, verify the absence of defects, and evaluates the conformity
to established sensory profiles of unifloral honeys and also to understand consumer preferences
(Ana et al., 2018). The percentage of fructose, glucose, amount and type of amino acids and the
organic acids vary by floral source that in turn determines the flavor of honey.

The effects of the main factors on mean acceptance of study honey varieties with respect to
color, smell thickness, mouth fill (texture), taste, sweetness, after taste and overall acceptance are
presented in Table 3. The score of the color acceptance test ranged from 4.64 (Lalibela location
of Eastern Amhara region) to 5.45 (Bure location of Western Amhara Region) on a scale of 9
with a significant difference among different locations. The judges in this sensory evaluation
determined the honeys from the Denbecha location as being darker in their color than those from
other locations (Table 3).

310
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The floral source also influenced the smell of the honey significantly among the honey samples
from different locations. In this regard honey fromBanja, Guangua, Bure and Dangla was found
to be significantly better preferred honeys in terms of their smell attributes than honey from
Lalibela. Honey from Sekota and Bure were perceived as being more intense regarding the
attributes of thickness and mouth fill (texture) when compared to Lalibela honey. The judges
determined that the honeys from the majority of locations did not significantly differ in taste,
sweetness and after taste attributes, but all are more intense than Lalibela honey (Table 2). The
lesser intensity of the taste attributes described in the sensory evaluation for honeys from
Lalibela honey compared to those from other locations appeared to be associated with their lower
pH and EC related to the higher mineral content (Table 1). The sweet taste of honey is attributed
to its higher fructose content and lower pH. Generally, in all attributes Stingless bee honey
produced from Sekota and Dehana from Eastern Amhara Region and Stingless bee honey
produced from Bure, Banja, and Dangla from Western Amhara Region perceived as more
intense in overall acceptability by the Judges.

The results in Table 3 revealed that on average the judges slightly like the stingless bee honey, as
the higher acceptance mean was 6.13, which is situated slightly above the neutral score 5 (neither
like nor dislike). In agreement with this result Vit et al. (2010) reported the neutral acceptance
mean score of the stingless honey sample with a maximum of 6.2 score of Melipona favosa
honey with the country of origin from Venezuela, mean score of 5.1 of Tetragonula carbonaria
honey with the country of origin from Australia and mean score of 5.6 scores for Melipona
scutellaris honey with the country of origin from Australia given by the Spanish consumers.

311
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 4: Sensory evaluation of stingless bee honey collected from different locations of Amhara region
District Color Smell Thickness Mouth fill Taste Sweetness After taste Overall acceptability
Sekota 5.14±2.02a 4.96±2.19abc 5.94±1.73a 5.92±1.48a 5.82±1.71a 5.91±1.8a 5.87±1.6a 6.13±1.49a
Dehana 5.16±2.07a 4.64±2.14abc 5.29±2.29ab 518±2.03ab 5.33±2.13a 5.36±2.14a 5.35±2.0ab 5.73±1.86ab
Lalibela 4.64±2.16a 3.99±1.93c 4.74±1.96b 4.32±1.92c 4.03±2.03b 4.03±2.05b 4.36±2.1c 4.32±2.1c
Tilili 4.77±1.86a 4.61±1.69abc 5.08±1.71ab 4.87±1.46bc 5.1±1.66a 5.03±1.5a 5.03±1.5abc 5.1±1.47bc
Dembecha 3.41±1.87b 4.23±1.72bc 5.32±1.7ab 4.91±1.31bc 5.09±1.54a 5.00±1.41a 4.82±1.14bc 5.05±1.09bc
Dangla 5.09±2.44a 5.00±1.82ab 5.33±1.53ab 5.42±1.68ab 5.73±1.64a 5.73±1.75a 5.58±1.68ab 5.61±1.60ab
Banja 4.73±2.12a 5.32±1.64a 5.14±1.46ab 5.32±1.52ab 5.95±1.46a 5.73±1.28a 5.45±1.68ab 5.73±1.49ab
Bure 5.45±1.65a 5.14±1.52ab 5.82±1.43a 5.82±1.68a 5.82±1.87a 5.73±1.45a 5.73±1.28ab 5.86±1.58ab
Guangua 5.27±1.6a 5.24±1.60a 5.3±1.36ab 5.52±1.37ab 5.15±1.5a 5.03±1.47a 5.21±1.62abc 5.52±1.44ab
Total 4.89±2.04 4.67±1.92 5.28±1.79 5.15±1.74 5.17±1.9 5.15±1.88 5.19±1.79 5.34±1.77

312
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Melissopalynological Analysis

Monofloral honey is where the bees have been foraging predominantly on one type of plant,
and is named according to that plant. The result of 41 studied honey samples in this work
shows that from 27 honey samples only 3 pollen types (Eucalyptus camaldulensis, Guizotia
abyssinica and Bidens pachyloma) were identified as monofloral honey with a predominant
level ranging from 57.71% (Bidens pachyloma) to 77.79% (Eucalyptus camaldulensis) (Table
5). Among the samples studied, Eucalyptus was the most common dominant pollen types
found in the majority of the honey samples (17/27), an indication that the pollen and nectar of
these plant species are important bee foods, while Bidens pachyloma was found to be less
common pollen types found in the studied samples (4/27) which were found to be dominant.

The second major pollen contribution is secondary pollen type of honey which ranges from
16 - 45% of pollen count due to this classification the studied honey samples are listed
according to secondary pollen honey in (Table 5). Secondary pollen types (>16 - 44%)
identified from the samples in order of decreasing significance were Eucalyptus
camaldulensis, Hypericum quartinianum, Gravlliea robusta, Bidens pachyloma, Guizotia
abyssinica, Echinops, Viciafaba,, Sorghum bicolor, Caesalpinia decapetala and Hypoestes
trifolia with their respective percentage shown in the table 5 below. Among these secondary
pollen types identified in the sample, Bidens pachyloma was the most common secondary
pollen types found in the honey samples (15), while Hypoestes triflora was found to be less
common secondary pollen types found in the studied samples (only in 1 sample) (Table 4).

The important minor pollen contributions are the third pollen type of honey which ranges
from 3 - 15% of pollen count due to this classification the studied honey samples are listed
according to important minor pollen in Table 6. Important minor pollen type (3 - 15% of the
count) identified from the samples in order of decreasing significance were Hypericum
quartinianum, Vicia faba, Hypoestes trifolia, Caesalpinia decapetala Schinus molle, Bersama
abyssinica, Gravillia robusta, Bidens pachyloma, Eucalyptus camaldulensis, Echinops, Zea
mays, Sorghum bicolor,and Croton macrostachyus with their respective percentage shown in
the table 5 below. Among these secondary pollen types identified in the sample, Bidens
pachyloma was the most common Important minor pollen type found in the honey samples
(9), while Hypericum quartinianum was found to be less common Important minor pollen
type found in the studied samples (only in 1 sample) (Table 4).

313
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The presence of these prominent and important pollen types in these samples confirm that the
honey samples were of botanical origin and also a true indication of their geographical origin,
while the rest pollen types were categorized as important minor and minor pollen types.
minor pollen type (Less than 3% of the count) identified from the samples were Acacia
brevispica, Acacia Senegal, Acacia seyal, Becium grandiflorum, Bersama abyssinica,
Caesalpinia decapetala, Echinops spps, Zea mays, Ocimum bacilicum, Sorghum bicolor and
Vernonia spps with their respective percentage shown in Table 7 below.

314
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 5: Identified predominant pollen types (more than 45% of the count)
Botanical sources Family name No of samples Mean pollen percentage Range Agroecology
Guizotia abyssinica Asteraceae 6 67.05% 45.53 - 91.77 Mid & highland
Bidens pachyloma Asteraceae 4 57.71% 53.2 - 61.15 Midland
Eucalyptus camaldulensis Myrtaceae 17 77.79% 59.15 - 92.98 Mid & highland
Total 27/41 (65.58%)

Table 6: Identified secondary pollen types (16 - 45% of the count)

Botanical sources Family Name No of samples Mean Pollen percentage Range Agroecology
Bidens pachyloma Asteraceae 15 29.03 16.4 - 44.83 Midland
Caesalpinia decapetala Fabaceae 2 22.93 22.8 - 23.06 Midland
Echinops Asteraceae 2 27.55 17.62 - 32 Highland
Eucalyptus camaldulensis Myrtaceae 2 31.01 30.79 - 44.83 Highland
Gravillea robusta Proteaceae 3 29.77 22.86 - 27.36 Midland
Guizotia abyssinica Asteraceae 7 28.5 16.4 - 35.93 Mid & highland
Hypericum quartinianum Guttiferae 6 30.52 17.64 - 42.1 Mid & highland
Hypoestes trifolia Acanthaceae 1 20 Highland
Sorghum bicolor Poaceae 2 27.26 16.7 - 21.21 Midland
Vicia faba Fabaceae 4 27.39 16.4 - 41.17 Midland

315
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 7: Identified minor pollen type (3 - 15% of the count)

Botanical sources Family Name No of samples Mean pollen percentage Range Agroecology
Gravillea robusta Proteaceae 5 9.15 4.13 - 157.17 Midland
Bersama abyssinica Francoaceae 6 10.07 6.8 - 13.6 Midland
Bidens pachyloma Asteraceae 9 8.72 3.76 - 15.5 Midland
Caesalpinia decapetala Fabaceae 3 11.25 8.3 - 12.85 Midland
Croton macrostachyus Euphorbiaceae 2 4.15 3.6 - 4.7 Midland
Echinops spps Asteraceae 7 7.3 3.15 - 14.65 Midland
Eucalyptus camaldulensis Myrtaceae 7 7.67 3.58 - 13.1 Mid & high
Hypericum quartinianum Guttiferae 1 13.3 Midland
Hypoestes trifolia Acanthaceae 3 12.16 5.3 Midland
Zea mays Poaceae 6 6.8 3.3 Highland
Schinus molle Anacardiaceae 5 10.71 7 - 15.47 Midland
Sorghum bicolor Poaceae 2 5.21 3.72 - 6.7 Midland
Viciafaba Papilionaceae 5 12.7 10.12 - 14.4 Mid & high

Table 8: Identified minor pollen type (less than 3% of the count)

Botanical sources Family Name No of samples Mean Pollen percentage Range Agroecology
Acacia brevispica Fabaceae 1 0.8 Highland
Acacia senegal Fabaceae 1 0.5 Highland
Acacia seyal Fabaceae 3 2.3 2.3 - 2.7 Midland
Becium grandiflorum Lamiaceae 3 2 1 - 2.7 Highland
Bersama abyssinica Francoaceae 1 2.3 Midland
Caesalpinia decapetala Fabaceae 1 2.8 Highland
Echinops spps Asteraceae 1 1.3 Midland
Zea mays Poaceae 3 1.42 1.3 - 1.65 Midland
Ocimum bacilicum Lamiaceae 2 1.58 1.25 - 1.9 Midland
Sorghum bicolor Poaceae 1 2.06 Midland
Vernonia spps Asteraceae 1 2.18 Midland

316
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

The parameters pH, electric conductivity, sucrose, ash and HMF in the honey samples
comply with the requirements of the previous global reports. The parameters of moisture
content, reducing sugar and free acidity do not comply with the limit compared to Apis
mellifera honey. This fact points to the necessity of proper honey harvesting, creating specific
legislation for stingless bee honey and justifies the need for a more harmonized standard of
this food product, which will include stingless bee honey from around the country. The study
confirmed that the stingless bee honey produced in the Amhara Region had a slightly like
acceptance score supported by the sensory panel evaluation.

This study has also led to the identification of major plants visited by stingless bees in the
study area. In this research, the possibility of producing three monofloral honey (Gizotia,
Bidens and Eucalyptus honey) from stingless bee honey in Amhara region could also provide
a major understanding of this type of honey’s excellent potential for rural household income
diversification and use it as a therapeutic agent in various health issues. However, to better
understand its beneficial effects of each type of honey as a therapeutic agent, more scientific
studies and clinical trials on human subjects need to be conducted.

REFERENCES

Alemayehu Gela, Zewdu Ararso and Deresa kebede. 2018. Production and Composition
Analysis of Stingless bees honey from West Showa zone of Oromia region, Ethiopia
Carvalho C.A.L., Sodre G.S., Fonseca A.A.O., Alves R.M.O., Souza B.A. and Clarton L.
2009. Physicochemical Characteristics and sensory profile of honey samples from
stingless bees (Apidae: Meliponinae) submitted to a dehumidification process. A Acad
Bras Cienc 81(1): 143-149
Costa, M.A., Del Lama, M.A., Melo, G.A.R. and Sheppard, W.S. 2003. Molecular phylogeny
of the stingless bees (Apidae, Apinae, Meliponini) inferred from mitochondrial 16S
rDNA sequences. Apidologie, 34, 73–84. https:// doi.org/ 10.1051/ apido: 2002051
Codex Alimentarius Commission. 2001. Alinorm 41/10: Revised standard for honey.
Alinorm 1:AC:19-26.
Fuenmayor C.A., Diaz-Moreno A.C., Zuluaga-Dominguez, Diaz-Moreno A.C. and Quicazan
M.C. 2012. Miel de angelita: Nutritional composition and physicochemical properties of
Tetragonisca angustula honey. Intercencia 37: 142-147
Issaro N., Weerakul T., Machana S., Ornnim P., Phanudulkitti C., Srijan T.,
Laiwattanaphaisal J. and Pattarapanich C. 2013. Stingless bee honey II: Qualitative and
quantitative studies on honey produced by three stingless bee species collected from a
mangosteen garden in Chantaburi province, Thailand Thai J Pharm Sci (Suppl.).

317
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Louveaux J., Maurizio A. and Vorwohl G. 1978. Methods of Melissopalynology, Bee World
59, 139– 157.
Martins, C.F., Cortopassi-Laurino, M., Koedam, D., and Imperatriz-Fonseca, V.L. 2004.
Tree Species used for Nidification by Stingless Bees. Biota Neotropica, 4(2).
Nascimento, A., Marchini, L., Carvalho, C., Araújo, D., Olinda, R. and Silveira, T. 2015.
Physical-chemical parameters of honey of stingless bees (Hymenoptera: Apidae).
American Chemical Science Journal 7(3):139-149.
Nordin A., Sainik N.Q.A.V., Chowdhury S.R., Saim A.B. and Idrus R.B.H. 2018.
Physicochemical properties of stingless bee honey from around the globe: A
comprehensive review, Journal of Food Composition and Analysis (2018),
https://doi.org/10.1016/j.jfca.2018.06.002
Özbalci, B., Boyaci, I.H., Topcu, A., Kadilar, C. and Tamer, U. 2013. Rapid analysis of
sugars in honey by processing Raman spectrum using chemometric methods and
artificial neural networks. Food Chemistry, 136(3–4): 1444–1452. https:// doi.org/
10.1016/ j.food chem.2012.09.064
Rodríguez-Malavera A.J., Rasmussen C., Gutierrez M.G., Gild F., Nieves B. and Vit P. 2009.
Properties of honey from ten ten of Peruvian stingless bees. NPC 4(9): 1221-1226
Santos, A., Marchini, L.C., Alfredo, C., Carvalho, L. De, Feliciano, D., Araújo, D. and
Antonia, T. 2015. Physical-Chemical Parameters of Honey of Stingless Bee
(Hymenoptera : Apidae), 7(3), 139–149. https://doi.org/10.9734/ACSj/2015/17547
Slaa, E.J., Sanchez Chavas, L.A., Malagodi-Braga, K.S. and Hofstede, F.E. 2006. Stingless
bees in applied pollination: practice and perspectives. Apidologie, 37, 293–315. https://
doi.org/ 10.1051/apido: 2006022
Solayman M., Islam M.A., Paul S., Ali Y., Khalil M.I., Alam N. and Gan S.H. 2016.
Physicochemical properties, minerals, trace elements, and heavy metals in honey of
different origins: A comprehensive review.Compr Rev Food Sci Food Saf. 15 :219-233
Souza, B., Roubik, D., Barth, O., Heard, T., Enríquez, E., Jerônimo, C. and Vit, P. 2006.
Composition of Stingless Bee Honey: Setting Quality Standards, 31(12), 867–875.
Vit P., Medina M. and Enriquez M.E. 2004. Quality standards for medicinal uses of
Meliponinae honey in Guatemala, Mexico and Venezuela. Bee World. 85(11): 2-5.

318
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Assessment of the Current Trophic Status of Lake Tana, Ethiopia

Adane Melaku and Alayu Yalew

Bahir Dar Fishery and Other Aquatic Life Research Center, P.O. Box 794, Bahir Dar, Ethiopia

ABSTRACT

Current assessment of trophic status of Lake Tana was conducted from 2017 to 2018. Water
transparency was measured using a Secchi disk and total phosphorus was measured as
molybdate reactive phosphate method. Chlorophyll-a (Chl-a) was extracted in acetone and
the pigment extract was centrifuged at 3000 rpm. The absorbance of the extracted was
measured with UV spectrophotometer at a wavelength of 650 and 750 nm. Calculation of
Chl-a was done by using the formula of Talling and Driver (1963). The mean value of total
phosphorus (mg L-1), Secchi depth (cm) and chlorophyll-a (Chl-a) were recorded 0.48, 51.5
and 9.92 µg L-1, respectively. The trophic state of Lake Tana was estimated and the trophic
state index value of Total phosphorus (TSITP) was 93.38 which imply that the lake was
Hypereutrophic. Trophic state index values of Secchi depth (TSISD) and Chl-a (TSIchl-a) were
50.42 and 53.1, respectively; implying that the lake is strongly mesotrophic. Carlson Trophic
State Index (TSIC) was estimated by the average of trophic state index values of Secchi depth
(TSISD), Total phosphorus (TSITP) and Chl-a (TSIchl-a). Thus, according to these three trophic
state parameters (TSIC) Lake Tana was Eutrophic Lake. This indicates the existence of high
nutrient load entering into the lake from the catchment. So, concerned bodies shall protect
and minimize the nutrient load before adverse effects on the aquatic ecosystem.

Keywords: Trophic state, Eutrophic Lake, mesotrophic, Chlorophyll-a

INTRODUCTION

Lakes are commonly classified according to their trophic state, a term that describes how
“green” the lake and measured by the amount of algae biomass in the water (Carlson, 1977).
Three trophic state categories are used to describe lakes as they grow progressively greener:
oligotrophic, mesotrophic and Eutrophic (Carlson, 1977). Watershed managers typically do
not determine trophic state by directly measuring algae biomass and hence the index number
can be calculated from any of several parameters, including Secchi disk transparency,
chlorophyll, and total phosphorus (Carlson, 1977; Kalff, 2002).

319
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Carlson (1977) established a classification of trophic state based on combined biological

parameter (chlorophyll-a), physical parameter (Secchi disk, transparency) and chemical
parameter (Total phosphorus). Phosphate concentration was used as one of the variables
because it limits algal growth in most lakes (Carlson, 1977). The trophic index of Carllson
(1977) has several advantages over previous attempts at trophic classification as they
determine how much nutrient abatement is necessary to reach a desired trophic condition.
The community structure and primary production of phytoplankton are available for many
east African Lakes (Talling and Lemoalle, 1998). Phytoplankton primary production is the
cornerstone of the entire aquatic ecosystem. Up to 70% of the atmospheric oxygen comes as a
result of the photosynthetic activity of aquatic micro and macro algae (Reynolds, 1994).

The amount of primary productivity in a lake is a better indicator of fisheries yield and total
biomass (Ryder, 1974; Downing et al., 1990) and the trophic state (Carlson, 1977). The
amount of primary production and biomass is, in turn, dependent upon several physico-
chemicals and biological factors. Physical features such as mixing, stratification driven by
wind regime and atmospheric temperature, determine the availability of nutrients to the
phytoplankton in space and time (Levinton, 2013). The extent of water transparency or
turbidity determines the depth of light penetration which, in turn, affects the rate of
photosynthesis (Dejen et al., 2017 and Schindler et al., 2019). The mixing pattern of the water
body also determines the distribution of phytoplankton species in the water column.
Planktonic species composition, density, age and physiological states are among biological
factors that determine the amount of primary production (Schindler et al., 1977;
Vollenweider, 1991).

Primary productivity also contributes to the amount of dissolved oxygen in water which is
available for community respiration including fish (Lee et al., 1996). However, increased
biological oxygen demand can lead to oxygen depletion during complete mixing or massive
nutrient enrichment which results in increased oxidation of some compounds by bacteria
(Carlson, 1977; Vollenweider, 1991; Carlson and Simpson, 1996). Nutrient dynamics of a
particular water body, thus, indirectly determines parameters like dissolved oxygen, pH,
alkalinity and conductivity which, in turn, affect the physiological state of aquatic animals
from zooplankton to fish (Kalff, 2002; Wetzel, 2001; Mays, 1996 and Lee et al., 1991).

320
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Ethiopia is a land locked country but endowed with relatively vast freshwater resources in the
region (FAO, 2016). The country is networked with twelve main river basins (FAO, 2016).
The Great East African Rift valley that bisects the country is home for diverse lakes, along
with dozens of highland and crater lakes. According to Greboval et al. (1994), the total area
of inland waters in Ethiopia is estimated to be 8,800 square kilometers, representing 0.72
percent of the total surface area of the country (Greboval et al., 1994). Studies show that
water bodies in Ethiopia are biologically productive in terms of phytoplankton, zooplankton,
benthic and fish species that make up the aquatic food chain and food webs (Tudorancea et
al, 2002). The diverse aquatic ecosystems found in the country are of greater economic
importance and provide ample opportunities for scientific studies (Zinabu, 1994). Among the
important lake ecosystems in Ethiopia are the Rift Valley Lakes and the associated Crater
Lake systems and the highland lakes.

The first series of limnological studies began in the 1960’s focusing on the morphometric and
limnological studies (Mohr, 1961; Baxter et al., 1965; Prosser et al., 1968). The second series
of studies latter continued as several authors investigated some alkalinity-salinity and
planktonic studies (Talling et al., 1973; Wood et al., 1976). Wood and Talling (1988) have
compiled some sporadic studies on diversity of phytoplankton species in Ethiopian water
bodies. Studies on zooplankton diversity for these lakes date back to expeditionary studies in
the 1930’s. However, by the time the third series of limnological studies commenced in the
1990’s, several water bodies in Ethiopia have undergone rapid changes in terms of planktonic
content, trophic state and chemical compositions (Brook, 1994; Zinabu, 1994; Elizabeth and
Amha, 1994; Elizabeth and Wiellén, 1998). Zinabu (1994) relates the changes to deleterious
anthropogenic activities in the form of industrial pollution, changes in land cover in the upper
water shades such as improper agricultural practices and deforestation. In the late 1990s,
some authors already found the extreme changes in the limnology and morphometric of some
water bodies (Zinabu, 2002) in which some lakes became non-existent as a result of
deleterious human intervention.

In general, trophic state measurements serve as benchmarks for measuring the success of a
lake management program. Therefore, the study aimed to assess the current trophic state of
Lake Tana.

321
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of Study Area

Lake Tana is the largest lake in Ethiopia, with an area of 3150 km2, located at an altitude of
1786 m (Dersseh et al., 2019; Vijverberg et al., 2009). Lake Tana is a crater lake formed two
million years ago, due to the volcanic blocking of the Blue Nile River (Vijverberg et al.,
2009). It has a mean depth of 8 m and a maximum depth 14 m (Vijverberg et al., 2009). The
catchment area of Lake Tana is about 16,000 km2 (Dersseh et al., 2019). Lake Tana is fed by
seven large permanent and 40 small seasonal rivers but four permanent rivers, namely Gilgel
Abbay, Megech River, Gumara River and Rib River account for 95% of the inflow of the
lake, while the Blue Nile is the only out flowing river (Dersseh et al., 2019; Vijverberg et al.,
2009).

Lake Tana is a highly turbid lake with low biological productivity, but unique diversity of
cyprinid fish (Vijverberg et al., 2009). Environmental changes in Lake Tana and its
watershed including eutrophication, associated with various anthropogenic activities that
resulted in the destruction of wetlands, have been observed (Tewodros et al., 2014). Recently,
development of extensive stands of water hyacinth on the shore areas of this sensitive lake,
one of the most ecologically dangerous weed infestations, has been reported. Other weeds
introduced into Lake Tana include the aquatic fern Azolla species (Wassie et al., 2014).
Shore of Lake Tana is covered with swamps whose macrophyte vegetation is dominated by
papyrus (Cyperus papyrus), Hippo grass (Echinochloa stagnina), Elephant grass (Phragmites
karka), Aquatic ferns (Azolla spp.), Typha latifolia, Waterlilies (Nymphaea spp.), and
Ceratophyllum species. Especially at the Southern part of Lake Tana’s are covered by
Ceratophyllum sp. (Ayalew, 2006; Wassie et al., 2014).

Sampling Protocol

Site selection was based on the proximity of agricultural practices, presence of industrial,
domestic and urban influences and rivers permanently feeding the lake. Thus, nine sites were
selected at shore and mouth of major rivers while the two sites were selected at open water on
Lake Tana purposely (Figure 1). Thus, collection of samples and in situ measurements of
physico-chemical parameters were made from eleven representative sampling sites. Samples
were collected from September 2017 to August 2018 at Lake Tana. Coordination point and
local names of representative sites of Lake Tana were presented in Table 1.

322
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 1: Representative sampling sites map at Lake Tana

Table 1: Coordination point and Local name of sampling sites at Lake Tana
Codes of sites Local name N E
S1 Blue Nile Resort 11°33'14.0" 037°18'40.3"
S2 Mango 11°38'49.067" 037°23'21.567"
S3 Tana Hotel 11°30'23.34" 037°20'18.25"
S4 Zegie open water 11°40'32.5" 037°23'08.8"
S5 Abay river mouth 11°52'07.9" 037°07'59.2"
S6 Maremiabet 11°36'24.3" 037°24'03.5"
S7 Sekela open water 12o13'31.2" 037o18'50.9"
S8 Megech river mouth 12°14' 44.2" 037°13'5.30"
S9 Derma river mouth 12o15'40.3" 037o18'83.3"
S 10 Gedamat 12o12'50.2" 037o17'33.3"
S 11 Kuriftu Resort 11°35'55.1" 037°23'06.4"

323
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Measurements of Physico-Chemical Parameters

Water transparency was measured using a Secchi disk of 20 cm diameter. Total phosphorus
was measured as molybdate reactive phosphate (MRP) after the organically-bound
phosphorus was converted to orthophosphate through oxidative hydrolysis with potassium
persulfate (APHA, 1999). This means 50 ml sample water was added in the Erlenmeyer flask.
Phenolphthalein was added on it. Then, 5N H2SO4 and potassium persulfate (K2SO8) was
added on the solution respectively. Then, the solution was boiled gently until the final volume
of 10 ml was reached. The boiled solution was cooled and phenolphthalein and 5N NaOH
was added on it respectively. Then, the combined reagent of potassium antimonyl tatrate
solution, ammonium molybdate solution and ascorbic acid solution was added to the sample
solution. Finally, the absorbance of the solution was measured with Jenway 6300UV-VIS
spectrophotometer at wavelength of 880 nm (APHA, 1999).

Chlorophyll-a Biomass of Phytoplankton

Appropriate volume of samples of phytoplankton was filtered using glass fiber filter papers
(GF/F) and frozen immediately. Chlorophyll-a (Chl-a) was extracted in acetone and the
pigment extract was centrifuged at 3000 rpm for 10 minutes. The absorbance of the extracted
was measured with Jenway 6300 UV spectrophotometer at a wavelength of 650 and 750 nm
(Wetzel and Likens, 2000). Calculation of Chl-a concentration was done using the following
formula (Talling and Driver, 1963).

µg 13.9(𝐸665 − 𝐸750)𝑉𝑒
Chl − a( ⁄L) =
𝑉𝑓𝑋𝑍

Where: E665 and E750 are absorbance at 665 nm and 750 nm, respectively, Ve is the volume
of extract acetone in ml, Vf is the volume of sample filtered in liter and Z is the Path length
of cuvette (cm)

Trophic State of Lake Tana

The trophic state index of Lake Tana was estimated according to Carlson (1977). The
Trophic State Index (TSI) of Carlson (1977) is a trophic state classification method that is
based on total in-lake phosphorus concentration, in-lake Chlorophyll-a (Chl-a) and water
transparency (ZSD).

324
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Secchi Disk Depth TSI (TSISD)

TSISD = 60 - 14.41*ln ZSD (m): where "ln" is the natural logarithm

Chlorophyll-a TSI (TSIChl)

TSIChl = 30.6 + 9.81*ln Chl-a (µgL-1)

Total Phosphorus (TP) TSI (TSITP)

TSITP = 4.15 + 14.42*ln TP (µgL-1)

Carlson Trophic State Index (TSIC) was estimated by the average of trophic state index
values of Secchi depth (TSISD), Total phosphorus (TSITP) and Chl-a (TSIchl-a). Thus,

TSITP +TSISD +TSIChl

TSIC =
3

According to Carlson (1977); Carlson and Simpson (1996) model trophic states of lake are
classified in Table 2.

Table 2: Trophic state categories proposed for temperate waters

Trophic state Carlson and Simpson (1996)
Eutrophic Hypereutrophic >65
Strongly eutrophic 62 - 64.9
Eutrophic 58 - 61.9
Slightly eutrophic 54 - 57.9
Mesotrophic Strongly mesotrophic 49 - 53.9
Mesotrophic 43 - 48.9
Slightly mesotrophic 38 - 42.9
Oligotrophic Slightly oligotrophic 33 - 37.9
Oligotrophic 26 - 32.9
Strongly oligotrophic 20 - 25.9

Statistical Analysis

The mean variation and mean of Physico-chemical parameters and chlorophyll-a were
analyzed between the dry and rainy seasons using one way ANOVA of SPSS software
(version 22).

325
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

RESULTS AND DISCUSSION

Physico-Chemical Parameters

Relatively high mean value of total phosphorus was recorded at site S 11 in dry season and at
S 5 in rainy season (Table 3); however, there was no significant difference (p>0.05) between
the two seasons of mean value of total phosphorus (Table 4). According to Tewodros et al.
(2014), this relatively high concentration of total phosphorus might have contributed to the
presence of agricultural practices around the catchment of Lake Tana. The total mean of total
phosphorus 0.48 mgL-1 was recorded at Lake Tana in dry and rainy seasons (Table 3).
According to Adopt a wetland program, Delaware Department of Natural resources and
Environmental control (ADDNE) (2003) and James (1997) guideline, total phosphorus
greater than 0.15 mgL-1 may lead to algal blooms in surface waters. The present mean of total
phosphorus was relatively higher than the previous reported concentration of soluble reactive
phosphorus for open water of Lake Tana (<0.1 mg L-1, Ayalew et al., 2007). Thus, this
nutrient is essential for survival, reproduction and growth of phytoplankton. Extreme input of
nutrients into water ecosystems leads to an excessive algal growth and causes eutrophication
(Kalff, 2002; Zinabu, 1994). Therefore, in fresh water, reducing the input nutrients, especially
phosphorus (<0.05 mgL-1, ADDNE, 2003) and nitrogen are able to control eutrophication
(Kalff, 2002; Zinabu, 1994).

Table 3: Concentration of total phosphorus (mgL-1), secchi depth (ZSD, cm) and chlorophyll-a
(Chl-a, µgL-1) at Lake Tana in dry and rainy seasons
Season
Dry Rainy Dry Rainy Dry Rainy
Sites TP (mgL-1) Secchi depth (ZSD, cm) -1
Chl-a (µgL )
S1 0.16 0.68 80 86 8.271 7.298
S2 0.62 0.258 55 45 16.087 13.638
S3 0.52 0.89 51 30 11.996 9.468
S4 0.025 0.233 66 52 6.993 7.680
S5 0.18 1.655 20 23 12.997 2.085
S6 0.26 0.173 70 60 7.089 9.383
S7 0.24 0.408 70 75 10.808 9.700
S8 0.458 0.607 30 25 12.997 12.51
S9 0.378 0.665 42 20 12.178 11.68
S 10 0.156 0.226 60 70 6.267 5.445
S 11 0.89 1.04 - - 12.965 10.68
Mean± SD 0.34±0.27 0.62±0.45 48.6±23.77
54.4±19 48.6±23.77 10.79±3.17 9.05±3.31
Mean ± SD 0.48±0.38 51.5±21.16 9.92±3.29

326
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Relatively high value of Secchi depth (ZSD, cm) was recorded at the site of S 1 both in the dry
and rainy season. The low value of Secchi depth (ZSD, cm) was recorded at the site of S 5
during this study period (Table 3) and there is a statistical difference among the groups
(p<0.05), (Table 4). At site S 5 was entered river Abay which may carry loam soil and
organic particulates by runoff (Zinabu, 1994) due to this factor low value of Secchi depth was
recorded at S 5. The overall mean of Secchi depth (ZSD, cm) was recorded to be 51.59 (Table
3). The present study in secchi depth was higher than the previous study reported (35.4 cm,
Ayalew et al., 2007).

Table 4: P-value of Chlorophyll-a (Chl-a), Total phosphorus (TP) and Secchi depth (ZSD)
Parameters Sum of Squares Mean Square F p-value
Chlorophyll-a(Chl-a) 154.260 15.426 2.328 0.091
Total phosphorus (TP) 1.677 0.168 1.242 0.362
Secchi depth(ZSD) 7747.000 860.778 11.356 0.000*
* is significant at P= 0.05 level

Chlorophyll-a of Algae (Chl-A)

Biomass of algal was measured as Chlorophyll-a slightly varied both spatially and temporally
at different sites in Lake Tana during the study period. Chl-a concentration at site S2 was
relatively high in the dry and rainy season. The minimum Chl-a concentration was recorded
at site S10 among in the dry season. In the rainy season the minimum concentration was at S5
(Table 3) but the difference was no statically significant difference (p>0.05; Table 4). The
overall mean of Chl-a concentration of 9.92 µgL-1 was recorded at Lake Tana during both dry
and rainy seasons (Table 3). The present mean chlorophyll-a biomass result was slightly
greater than the previous reported biomass of chl-a (2.6-8.5 µgL-1) (Ayalew et al., 2007).

Estimation of Trophic State Index of Lake Tana

Trophic state index values of Lake Tana were calculated using the total mean value of Secchi
depth (m), total phosphorus (TP) and Chl-a data of all sampling sites of Lake Tana according
to Carlson (1977). The total phosphorus trophic state index (TSITP) value suggested that
Lake Tana was at a hypereutrophic state while Secchi depth trophic state index (TSISD) and
chlorophyll-a (TSIchl) was implied strongly mesotrophic (Table 5 and Table 2).

327
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Thus, a trophic state index value (TSIC) of 65.63 was obtained for Lake Tana. Lake Tana is a
eutrophic lake and found near the upper border of Eutrophic level, according to the indices of
the classification system for trophic states (Carlson, 1977). Goshu et al. (2017), predict that
the trophic state of Lake Tana has gradually changed from mesotrophic to Eutrophic in some
places due to nutrient load. Thus, the cause of eutrophication of Lake Tana might be input of
high nutrient load from the catchment by erosion. However, Lake Tana was previously
described as mesotrophic with low chlorophyll content and primary production by tropical
lakes' standard (Ayalew et al., 2007; Yirga and Hassen, 2015).

Table 5: Estimated value of trophic state index of Lake Tana

Trophic State Values estimated for the
Parameters used Index (TSI) Lake Tana Trophic state Category
Secchi depth (ZSD ) TSISD 50.42 Strongly mesotrophic
Total phosphorus (TP) TSITP 93.38 Hypereutrophic
Chlorophyll-a (Chl-a) TSIChl 53.10 Strongly mesotrophic
TSIC 65.63 Eutrophic

CONCLUSION AND RECOMMENDATION

Lake Tana could be described as strongly mesotrophic by using both chlorophyll-a and
Secchi depth trophic state index. However, the total phosphorus trophic state index was
indicated that Lake Tana was Hypereutrophic Lake. In general, according to the average of
three measured trophic state indexes, Lake Tana could be classified under Eutrophic
category; this indicated that the lake might be polluted and the water quality might be
dropped. The different development activities being carried out in the proximity of its shores
and in the catchment area of the lake might be the possible cause for increased total
phosphorus which leads the Lake Tana to be a eutrophic lake. Therefore, to minimize freely
released high nutrient especially phosphorus from domestic, urban effluents and agricultural
practices to the lake, feasible mitigation and control measures should be taken by the
concerned bodies in collaboration with the local community. Hence, it is recommended to
conduct continuous limnological studies to closely monitor the water quality of the lake so
that some remedial actions could be undertaken.

328
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENT

We would like to thank the Amhara Agricultural Research Institute (ARARI) and Ethiopia
Institute of Agricultural Research (EIAR) for financing the research project. We also thank
our staff members of Bahir Dar Fishery and other Aquatic Research Center especially
researchers and research assistants for their assistance during the research time.

REFERENCES

Adopt a wetland program, Delaware department of natural resources and environmental

control (ADDNE), 2003. Background information for interpreting water quality
monitoring results. Smyrna, pp. 735-8665.
APHA (American Public Health Association), American Water Works Association, AWWA
and Water Environment Federation, WEF, 1999. Standard methods for the examination
of water and Wastewater 20th Ed. New York, pp. 1-360.
Ayalew Wondie, 2006. Dynamics of the Major Phytoplankton and Zooplankton Communities
and Its Role in the Food Web of Lake Tana, Ethiopia. Ph.D. Thesis, Addis Ababa
University, Addis Ababa, pp.164.
Ayalew Wondie, Seyom Mengistu, Vijverberg, J., and Eshete Dejen, 2007. Seasonal
variation in primary production of a large high altitude tropical lake (Lake Tana,
Ethiopia): Effects of nutrient availability and water transparency. Aquatic Ecology,
41:195-207.
Baxter, R.M., Prossor, M.V., Talling, J.F. and Wood, R. B. 1965. Stratification in tropical
African lakes at moderate altitudes (1500-2000). Limnol.Oceanogr, 10:510- 520.
Brook Lemma, 1994. Changes in the limnological behavior of a tropical Africanexplosion
crater lake: L. Hora-Kilole, Ethiopia. Limnologica 24 (1):57-70.
Carlson, R. 1977. The trophic state index for lakes. Limnol. Oceanogr. 22: 361-369.
Carlson, R.E. and Simpson, J. 1996. A coordinator’s guide to Volunteer Lake monitoring
methods. North American Lake management society. Canada, 96pp.
Dejen Eshete, Anteneh Wassi and Vijverberg, J. 2017. The decline of the Lake Tana
(Ethiopia) fisheries: causes and possible solutions. Land Degrad. Develop, 28: 1842–
1851.
Dersseh Minychl G., Kibret Aron A., Tilahun Seifu A., Worqlul Abeyou W., Moges
Mamaru A., Dagnew Dessalegn C., Abebe Wubneh B. and Melesse Assefa M. 2019.
Potential of Water Hyacinth Infestation on Lake Tana, Ethiopia: A Prediction Using a
GIS-Based Multi-Criteria Technique. Water, doi: 10.
Downing, J.A., Plante, C. and Lalond, F. 1990. Fish production correlated with primary
productivity. Can. J. Fish. Aquat. Sci., 47:1929-1936.
Elizabeth Kebede and Amha Belay, 1994. Species composition and plankton biomass in a
tropical African lake (Lake Awassa, Ethiopia). Hydrobiology, 288:13-32.
Elizabeth Kebede and Willén, E. 1998. Phytoplankton in a salinity-alkalinity series of lakes
in the Ethiopian Rift Valley.Algological studies, 89: 63-96.

329
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Goshu, G., Koelmans, A.A. and Klein, J.J.de. 2017. Water quality of Lake Tana basin, upper
Blue Nile, Ethiopia: A review of available data. In: K.Stave, G.Goshu and S. Aynalem
(Eds.), Social and Ecological system Dynamics, pp.127-141.
James, E.K. 1997. Black Earth Creek and Limnology Minifacts & Analysis sheet 2:
phosphorus amounts. Spring Harbor, pp. 4.
Kalff, J. 2002. Limnology: Inland water Ecosystems. Prentice-Hall, Inc, NJ, pp. 1-592.
Lee, J.H.W., Wu, R.S.S. and Cheung, Y.K. 1991. Forecasting of dissolved oxygen in Marine
Fish Culture Zone. J. Env. Engrg.117:816-836.
Lee, Z. P., K. L. Carder, J. Marra, R. G. Steward and Perry M. J. 1996. Estimating primary
production at depth from remote sensing. Appl.Opt., 35: 463–474.
Levinton, J.S. 2013. Marine biology: function, biodiversity, ecology. Oxford university press,
New York, pp. 173- 193.
Li, N., Shi, K., Zhang, Y., Gong. Z., Peng, K., Zhang, Y. and Zha, Y. 2019. Decline in
Transparency of Lake Hongze from Long-Term MODIS Observations: Possible Causes
and Potential Significance. Remote Sens., 11: 177.
Mays, L. W. 1996. Water Resources Handbook. McGraw-Hill. New York, pp: 92.
Mohr, P.A. 1961. The geology, structure and origin of the Bishoftu explosion crater lakes,
Shoa, Ethiopia. Bull. Geophys. Obs., Addis Ababa, 2:65-101.
Prosser, M.V., Wood, R.B. and Baxter, R.M. 1968. The five Bishoftu crater lakes. SINET:
Ethiop. J. Sci. 8: 9-29.
Reynolds, C.S. 2006. The ecology of phytoplankton. Cambridge University Press,
Cambridge, pp. 1-300.
Schindle, D.W. 1974. Eutrophication and recovery in experimental lakes. 184(4139):897-9.
Schindler, D.W. 1977. Evolution of phosphorus limitation in lakes. Science, 195: 260-262.
Reynolds, C.E. 1987. Cyanobacterial water blooms. In: Callow, J. (Ed.). Advances in
Botanical Research, Vol. 13, Academic Press, London, pp. 67-143.
Reynolds, C.S. 1994. The long, the short and the stalled: on the attributes of phytoplankton
selected by physical mixing in lakes and rivers. Hydrobiologia, 289: 9-21.
Reynolds, J.W. 1994. Earthworms of the world. Global Biodiversity, 4(1): 11-16.
Ryder. 1974. Primary production: The magnitude and dynamics of phytoplankton.
Talling, J. F. and Driver, D. 1963. Some problems in the estimation of chlorophyll-a in
phytoplankton. In: Primary Productivity, Marine and Freshwater, pp. 142 - 146 (Doty,
M., ed.), UD Atomic Energy Commission, Washington, DC.
Talling, J.F. and Lemoalle, J. 1998. Ecological dynamics of tropical inland waters.
Cambridge University Press, Cambridge, pp. 1-350.
Talling, J.F., Wood, R.B., Prosser, M.V. and Baxter, R.M. 1973. The upper limit of
photosynthetic productivity by phytoplankton: evidence from Ethiopian soda lakes.
Freshwater. Biol., 3: 53 -76.
Tewodros Taffese, Seifu Tilahun, Tammo Steenhuis, 2014. Phosphorus modeling in Lake
Tana Basin, Ethiopia. J. Environ. Human, 2: 47 - 55.
Tudorancea, C. and Taylor W. and Leiden, W. 2002. Ethiopian Rift valley lakes: Inland
water.

330
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Vijverberg, J., Sibbing, F. A. and Dejen, E. 2009. Lake Tana: Source of the Blue Nile. In H.
J. Dumont, The Nile: Origin, Environments, Limnology and Human Use (p. 818).
Dordrecht: Springer Netherlands.
Vollenweider, A. 1991. A Manual on Methods for Measuring Primary Production in Aquatic
Environments. Int. Biol Book of Roses.
Wassie Anteneh, Minwyelet Mingist, Ayalew Wondie, Dereje Tewabe, Woldegebrael
W/Kidan, Addisalem Assefa and Wondie Engida, 2014. Water hyacinth coverage survey
report on Lake Tana. Technical Report Series 1, BDU, ORDA, BOEPLAU, ARARI,
Bahir Dar, pp. 1- 29.
Wetzel, R.G. 2001. Limnology: Lake and River Ecosystems. 3rd ed. Academic Press. N.Y.,
pp. 1-1006.
Wetzel, R.G. and Likens, G.E. 2000. Limnological Analyses.3rd Edn. Verlag, Inc. N.Y., New
York, 360pp.
Wood, R.B. and Talling, J.F. 1988. Chemical and algal relationships in salinity series of
Ethiopian inland waters. Hydrobiologia 15: 29– 67.
Wood, R.B., Prosser, M.V. and Baxter, R.M. 1976. The seasonal pattern of thermal
characteristic of four Bishoftu crater lakes, Ethiopia. Fresh water. Biol. 6: 519-530.
Yirga Kebede Wondim and Hassen Muhabaw Mosa, 2015. Spatial variation of sediment
physico-chemical characteristics of Lake Tana, Ethiopia. J. Environ. Earth Sci., 5: 95-
109.
Zinabu Gebre-Mariam, 1994. Long–term changes in indices of chemical and productivity
status of a group of tropical Ethiopian lakes, with differing exposure to human influence
.Arch. Hydrobiologia. 132(1):115-125.
Zinabu Gebre-Mariam, Elizabeth Kebede, and Zerihun Desta, 2002. Long-term changes in
chemical features of waters of Ethiopian rift-valley lakes. Hydrobiologia.477: 81–91.

331
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Recent Trends in the Physico-Chemical Characteristics of Lake Tana, Ethiopia

Adane Melaku, Tizazu Yitayew, Abrham Amare, Berhan Mohammed, Alayu Yalew,
Benyam Hailu, Haymanot Yenesew and Wondie Zelalem

Bahir Dar Fishery and Other Aquatic Life Research Center, P.O. Box 794, Bahir Dar, Ethiopia

ABSTRACT

A recent trend in the physico-chemical features of Lake Tana was assessed from 2012 to
2018. Phosphate-P, nitrate, nitrite, total hardness, alkalinity, silica, ammonia-N and hydrogen
sulfide was done immediately after collection with a water analysis kit of Wagtech
International, Palintest transmittance display photometer 5000, Palintest Ltd., UK. Dissolved
oxygen (DO), conductivity, pH, temperature, salinity and total dissolved solids (TDS) were
measured in situ using YSI 556 multi-probe system. The trend of electrical conductance, total
dissolved solids (TDS), nitrate, nitrite, soluble reactive phosphate, total hardness, alkalinity,
silica, and hydrogen sulfide was increasing from time to time. Dissolved oxygen of Lake
Tana the surface decreased slightly year to year. According to the depths of sampling sites of
Lake Tana indicated that the depth of the lake declined from 2012 to 2018 except 2017.
However, the trends of surface water temperature, pH, salinity, turbidity and Secchi depth
readings of the lake varied in time slightly. In this study nitrate-N, phosphorus and silica level
was high in Lake Tana. Thus, these nutrients indicated that high nutrients might enter from
the catchment of the lake and urban wastes. The decline of depths of representative sites at
Lake Tana indicated that the depth level of Lake Tana water was declining due to the
existence of loss of water by different mechanisms. Therefore, it is necessary to understand
the current changes of Lake Tana to take action to protect the aquatic and terrestrial life as
well as the risk of the lake.

Keywords: physical and chemical characteristics, trends

332
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Ethiopia is a landlocked country but endowed with relatively vast freshwater resources in the
regions (Greboval et al., 1994; FAO, 2016 and FDREMWR, 2002). The country is networked
with twelve main river basins (FAO, 2016). According to Greboval et al. (1994), the total area of
inland waters of major lakes and reservoirs in Ethiopia is represented by 8800 km 2. Studies show
that water bodies are biologically productive in terms of phytoplankton, zooplankton, benthic and
fish species that make up the aquatic food chain and food webs (Tudorancea et al., 2002). The
diverse aquatic ecosystems found in the country are of greater economic importance and provide
ample opportunities for scientific studies (Zinabu, 1998).

Rapid population growth accompanied by increased water demand has led to diminishing
quantity and deteriorating quality. In particular, natural water bodies such as lakes and rivers are
increasingly becoming under human pressure (Mamaru and Desta, 2016). Anthropogenic
activities impact aquatic ecosystems both directly and indirectly while the ecosystems in some
cases responded to influences that surpassed their resilience capacity. Direct influences include
water abstraction or water flow via diversion of streams and rivers which result in a complete
change in the bathymetry, chemistry, and the ecosystem structure (Strayer and Findlay, 2010).
Indirect influences refer to terrestrial activities, such as agriculture and industries which can result
in nutrient load and chemical pollution through surface runoff and or ground percolation
(Reynolds, 1984; Talling and Wood, 1988). The resulting nutrient enrichment, especially in
closed basin water bodies can result in increased biomass production by phytoplankton and
macrophyte, which gradually leads to elevated biological oxygen demand (BOD). This is also
accompanied by remarkable change in physico-chemical parameters, such as pH, alkalinity,
salinity, dissolved oxygen (DO), light attenuation and nutrients (Kalff, 2002; Wetzel, 2001;
Mays, 1996). These will, in turn, impact the zooplankton community and the rest up the food
chain including fish (Lee et al., 1991).

Limnological studies for some of the Ethiopian rift valley and highland lakes began in the 1960’s.
The studies mainly include salinity-alkalinity series, planktonic composition and biomass,
nutrient dynamics and stratification and lake’s morphometry. Those studies were carried out in
three distinct periods, 1960’s (Mohr, 1961; Baxter and Wood, 1965; Prosser et al., 1968), 1970’s
(Talling et al., 1973; Wood et al., 1976) and 1980 - 90’s (Wood and Talling, 1988; Brook, 1994;
Zinabu, 1994; Elizabeth Kebede, 1994; Elizabeth and Wiellén, 1998; Zinabu, 1998) each of
which were characterized by the dynamicity in physico-chemical and biological changes in major
inland water bodies and explanations given to the changes by the respective authors.

333
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The studies have clearly indicated the danger faced by many of Ethiopia’s valuable aquatic
ecosystems in which two notable water bodies underwent significant changes. One of these
lakes, Lake Haramaya disappeared due to unbalanced water abstraction and eutrophication
(Zinabu, 1998; Brook, 2002; Zinabu, 2002). The other lake, Lake Kilote (Kilole), once saline
soda lake home to migrating Flamingos and Arthrospira blue green algae, turned into an
almost freshwater lake (Zinabu 1994, 1998; Brook, 2002). The latter was due to diversion of
a small river into it with an intention to use the lake as a reservoir for irrigation scheme.
Many other water bodies underwent changes in their limnological features due to multi-
factorial reasons but ultimately related to direct or indirect human influences.

However, there is no holistic data for the major inland water bodies over the last decade, a
fourth period as it is continued from the past studies even though anthropogenic influences
and interventions have continued to act together with ever increasing climate change impact.
Indirect impacts such as changes in land cover, runoff and sediment loading have also gained
momentum but their impacts on the physico-chemical features over the last ten years or so
are unknown. This study, therefore, intended to close determine current physico-chemical
properties of Lake Tana and their present and future implication on the biological
productivity and water quality of Lake Tana

MATERIALS AND METHODS

Study Area Description

Lake Tana is the largest lake in Ethiopia, with lake area of 3150 km2 and catchment’s area of
16,000 km2, located at an altitude of 1786 masl (Dersseh et al., 2019; Vijverberg et al., 2009).
Lake Tana is a crater lake formed two million years ago, due to the volcanic blocking of the
Blue Nile River and it has a mean depth of 8 m and a maximum depth 14 m (Vijverberg et
al., 2009).

Lake Tana is a highly turbid lake with low biological productivity, but a unique diversity of
cyprinid fish. Environmental Changes in Lake Tana and its watershed including
eutrophication, associated with various anthropogenic activities that resulted in the
destruction of wetlands, have been observed (Tewodros et al., 2014). Recently, development
of extensive stands of water hyacinth on the shore areas of this sensitive lake, one of the most
ecologically dangerous weed infestations, has been reported. Other weeds introduced into
Lake Tana include the aquatic fern Azolla species (Wassie et al., 2014).

334
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The inappropriate discharge of waste water into Lake Tana and Abay (Blue Nile) River and
the associated health risk should be given due attention by the health inspectors and the local
inhabitants. Lack of facility for waste water collection and treatment seems to be the most
important problem. Furthermore, lack of awareness and carelessness on the part of both the
public and sanitation workers and the weak control or absences of penalty for offenders seem
to have contributed to the degradation of the water environment around the fast-growing
cities like Bahir Dar (Fesseha, 1988).

Description of Sampling Sites

A total of 12 sampling sites (S) were selected as representative of the whole Lake Tana
(Figure 1). The selection of the sampling sites was based on agricultural, domestic, urban
effluents, vegetation cover and depth purposely. Collection of samples and in situ
measurements of physico-chemical parameters were made at the 8 sampling sites from 2012
to 2016 at about 2 months interval. In 2017 and 2018 another 4 sampling sites were added on
the previous 8 sampling sites which are presented in Table 1.

S2 and S5 represent open parts of the lake at north and south of Lake Tana, respectively,
which were less impacted from human activities. S12 was found at the north part of the Lake
Tana at Delgi and urban waste is entered into the lake at this site and agricultural practices
are conducted commonly in the area. While S1, S3, S4 are found in the north of Lake Tana
around Gorgora. Sites like S8, S9, S10 and S11 are found in the southern gulf of Lake Tana,
and different urban effluents and domestic waste were entered on these sites from Bahir Dar
city. S7 and S6 are found eastern and western parts of Lake Tana which are located at
Gumera river mouth and Abay river mouth respectively. Lists of sites and its GPS
coordination point was described in table 1.

335
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 1: A map showing representative sampling sites of Lake Tana

Table 1: GPS coordination point and depth of sampling sites at Lake Tana from 2012 to 2018
Site Average
code Local name Depth (m) N (degree ) E (degree) Elevation (m)
S1 Dirma river moth 2.08 12.26119 37.32314 1791
S2 Sekela open 8.24 12.22533 37.31414 1790
S3 Gedamat 2.61 12.21394 37.20925 1791
S4 Gorgora fishery sub-center 1.60 12.23228 37.30008 1793
S5 Zegie open 10.93 11.67569 37.38578 1793
S6 Abay river moth 1.29 11.86886 37.13311 1785
S7 Gumera river mouth 2.27 11.89458 37.49047 1789
S8 Gerima 3.48 11.61467 37.38214 1790
S9 Debremariam 3.61 11.61608 37.40058 1792
S10 Maremiabet 3.17 11.60675 37.40097 1788
S11 Kuriftu resort 2.72 11.59864 37.38511 1794
S12 Delgi 1.93 12.1843 37.05947 1782

336
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Measurements of Physico-Chemical Parameters

Water transparency was measured using a Secchi disk of 20 cm diameter. Dissolved oxygen
(DO), conductivity, pH, temperature, salinity and total dissolved solids (TDS) were measured
in situ using YSI 556 multi-probe system. Turbidity was determined photo-electrically using
Wagtech International, Palintest transmittance display photometer 5000, Palintest Ltd., UK.

Determination of Chemical Parameters

The concentrations of the following inorganic nutrients were determined using a chemical
analysis kit, Palintest transmittance display photometer 5000 (Wagtech International).
Ammonia-nitrogen (NH3_N) was measured using filtered composite samples by the
indophenol method. Nitrate and nitrite-nitrogen (NO3 and NO2) were analyzed by the
Palintest nitrate test and Palintest Nitricol methods, respectively. Hydrogen Sulphide was
determined using a reagent containing diethyl-p-phenylenediamine (DPD) and potassium
dichromate with Palintest transmittance display photometer 5000. Phosphate (PO4) was
analyzed by the Palintest Phosphate LR method, while total hardness and alkalinity was done
by colorimetric method. Silica (SiO2) was determined by the molybdo silicate method.

Statistical Analysis

Physico-chemical parameters were assessed using Descriptive Statistics and the variation of
mean between years was analysed by one way ANOVA by SPSS statistical software (version
22). Graph presentations were done using Sigma plot 10 Software.

RESULTS

Physico-Chemical Parameters of Lake Tana

A trend of surface water temperature ranged from 22.6 to 24oC. The minimum value of
absolute temperature was recorded in 2018 while the maximum value was recorded in 2014
(Table 2).

337
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Mean temperature and salinity trends of Lake Tana surface water properties from
2012 to 2018
Year Temperature (°C) Salinity (ppt)
2012 22.7742±1.077012 0.05908±0.0193
2013 22.8286±1.620148 0.06929±0.0047
2014 23.9066±1.327303 0.05906±0.0117
2015 23.3613±1.852891 0.05872±0.0047
2016 23.5500±1.502164 0.0667±0.0056
2017 23.3180±1.030083 0.0580±0.0045
2018 22.5574±1.110889 0.0672±0.0094

The difference in surface water temperature was statistically insignificant (p>0.05). However,
specific conductance and TDS were statically significant (p<0.05) (Table 3).

Table 3: P-value and mean square of physico- chemical characteristics in one way ANOVA
Parameters F- test P-value Mean square Minimum Maximum
Secchi depth 1.30 0.258 1382.02 0.00 200
depth 0.81 0.56 9.11 0.50 12.80
Temperature 2.12 0.054 4.50 19.38 29.63
pH 10.9 0.00* 3.06 6.05 9.26
DO 1.87 0.089 1.064 3.75 7.82
Sp. conductance 5.39 0.00* 1682.40 0.00 175.00
TDS 6.95 0.00* 1196.22 0.00 142.00
Salinity 7.16 0.00* 0.001 0.00 0.08
Phosphate (PO4) 6.50 0.00* 1.26 0.00 3.75
Nitrate (NO3) 2.71 0.016* 4.55 0.00 13.8
Nitrite (NO2) 0.89 0.50 0.011 0.00 1.36
Ammonia (NH3-N) 5.58 0.00* 0.065 0.00 0.88
Total hardness 2.99 0.009* 3573.48 0.00 425.00
Silica (SiO2) 26.60 0.00* 781.53 0.00 32.6
Sulphate 8.19 0.00* 56.12 0.00 18
Alkalinity 2.35 0.033* 605.48 00.00 163.00
Turbidity 1.62 0.15 7945.06 0.00 550.00
Hydrogen sulfide (H2S) 40.23 0.00* 0.018 0.00 0.14
DO = Dissolved oxygen, Sp. conductance = Specific conductance, TDS = Total dissolved
solids * is significant at the 0.05 level

Specific conductance (K25, µS cm-1) ranged from 122 to 150.7. Among this, the minimum
was recorded in 2017 while the maximum value was recorded in 2018. The total dissolved
solids ranged from 81.2 to 126.9 mg L-1. The minimum value of TDS was recorded in 2017
while the maximum value was recorded in 2014 (Figure 2).

338
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 2: Specific conductivity and TDS trends of Lake Tana water properties from 2012-
2018.

The pH value was in the range of 7.2 to 8.0. The lowest pH value was reported in 2016 and
the maximum pH recorded in 2017 (Figure 3). The lowest DO concentration was observed in
2018 with the value of 5.9 mg L-1 while the highest concentration mg L-1 (6.8) was in 2012.
The salinity (ppt) of the Lake Tana was low (<0.1) from 2012 to 2018 (Figure 3).

Figure 3: pH and dissolved oxygen (DO) Trends of Lake Tana water characteristic from
2012-2018

The mean value (mg L-1) of Nitrate (NO3) varied between years and lower value was
observed in 2017 (0.71). The maximum mean concentration of Nitrate was recorded in 2013.
The minimum value of soluble reactive phosphorus (SRP mg L-1) of Lake Tana was 0.09 in
2017 while the maximum value was 0.65 in 2014. The mean value (mg L-1) of ammonia
(NH3-N) ranged from 0.01 to 0.21. The maximum mean value of ammonia-N was recorded in
2012, while the minimum mean value was recorded in 2014 at Lake Tana. The mean value
(mg L-1) of nitrite (NO2) ranged from 0.007 to 0.06 (Figure 4).

339
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 4: Ammonia-N, soluble reactive phosphorus, nitrate and nitrite trends of Lake Tana
water properties form 2012-2018

The mean value of total hardness (mg L-1 CaCO3) and alkalinity of Lake Tana ranged from
47.29 to 84.3 and 35.4 to 71.92, respectively. The maximum value of total hardness was
recorded in 2014 and the minimum value was in 2013. Both total hardness and alkalinity
showed increased values in 2018 (Figure 5).

Figure 5: Total hardness and alkalinity trends of Lake Tana water properties for 2012-2018

During the study period, the hydrogen sulphide (H2S) concentration (mg L-1) ranged from
0.03 to 0.098 at Lake Tana. The concentration of H2S increased from 2015 to 2019 radically
except the missing of 2017 (Table 4).

340
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 4: Mean of hydrogen sulfide, silica and depth trends of Lake Tana characteristics from
2012 - 2018
Year Hydrogen sulfide (mg L-1 H2S) Silica (mg L-1 SiO2) Depth (m)
2012 - 14.03±0.5.039 5.48±4.066
2013 - 8.21±0.7.642 4.96±3.741
2014 0.029±0.0104 0.35±0.949 4.25±3.610
2015 0.026± 0.0147 2.90±7.820 3.77±3.317
2016 0.063± 0.0270 5.46±5.755 3.43±3.054
2017 - 2.88±1.919 5.03±4.148
2018 0.098±0.0164 13.97±3.440 3.91±2.846

The silica concentration (SiO2) varied in time within a range of 0.3 to 15.23 mg L-1 (Table 4).
The maximum value of silica concentration was recorded in 2012 and the lowest silica
concentration was in 2014.

The mean depth (m) of the sampling site on Lake Tana varied between years and ranged from
3.43 to 5.48 m. The mean depth of the sampling site on lake declined from 2012 to 2018 with
the exception in 2017 (Table 4). This result was indicated that the depth level of Lake Tana
was declined and fluctuated timely due to different factors.

The mean of Secchi depth (cm) and turbidity (NTU) varied from 36.4 to 62.3 and 25.2 to
71.83, respectively. Maximum Secchi depth reading and turbidity were recorded both in 2016
while the minimum values were recorded in 2017 and 2014, respectively (Figure 6). The
trend of mean turbidity and Secchi depth was shown irreversible on Lake Tana in the years of
2012 to 2018.

Figure 6: Turbidity and Secchi depth trends of Lake Tana during 2012 to 2018.

341
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

DISCUSSION

Physico-Chemical Characteristics of Lake Tana

The depth of the sampling site on Lake Tana was decreased timely according to Tamar and
Ilia (2011) water levels trend of lakes fluctuated naturally due to hydrological and climatic
forcing. Thus, the depth of sampling site trend indicates that depth was declined on Lake
Tana. The decline of depth of sampling sites on Lake Tana might come from decreased inputs
of water entering through the Feeder Rivers of Lake Tana and increased loss of water through
evaporation of climatic condition and water use for hydroelectric power generation (Ayalew
et al., 2007).

The trend of absolute mean surface temperature was observed from 22.56 - 23.91°C on Lake
Tana. Absolute surface water temperature in all sampling years (2012 - 2018) was within the
range of variation reported for most tropical water bodies’ 20 - 30°C (Talling and Lemoalle,
1998). The result was also similar to those reported previously on Lake Tana (Ayalew et al.,
2007; Akoma, 2010). The variation of absolute mean surface water temperatures in 2012 to
2018 might be attributable to the effect of macrophytes on heat transfer from the near-surface
and region to deeper parts (Wetzel, 2001) and heat transported through inflows of rivers. The
level of mean salinity was lower than 0.1ppt at Lake Tana. According to WHO (2011), the
present study indicates that level of salinity is under drinking water scale and safe to aquatic
life.

The trend of Secchi depth and turbidity of water fluctuate up and down on Lake Tana within
the year of 2012 to 2018. According to Shang et al. (2015) the trend of Secchi depth and
turbidity of water was shown inversely related between them from 2012 to 2018 on Lake
Tana. High Secchi depth was recorded in 2016, due to this might be the presence of
macrophytes, which reduce wave action thereby reducing the level of total suspended solids
and the soil erosion might be minimized with different factors like drought. The minimum
value Secchi depth was recorded in 2014, the cause of this high turbidity of Lake Tana
(Tewodros et al., 2014) and the factor mentioned above may be the reverse is true (Shang et
al., 2015; Wetzel and Likens, 2000).

342
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The trend of specific conductivity (K25, µS cm-1) of water was recorded from 122 to 150.7 on
Lake Tana. The high conductivity recorded may be associated with the influx of dissolved
solutes from urban wastes and inputs from agricultural practices. However, the observed
conductivity of Lake Tana was laid on the human drinking water range (25 - 250 µS cm-1,
WHO, 1996). According to WHO (2011), the palatability of the water of the present
sampling sites, whose mean total dissolved solids (TDS) were less than 600 mg L-1, is
considered to be good. The pH level was varied year to year between near neutral to slightly
alkaline due to the low buffering capacity of the lake water (Wood and Talling, 1988).
However, the actual recorded mean pH level was between the recommended value of WHO,
2011 (6.5 - 8.5) safe to drinking water.

The trend of mean dissolved oxygen (mg L-1) of water was in the range of 5.9 to 6.8 during
2012 to 2018 on Lake Tana. According to Tarekgne et al. (2012) dissolved oxygen level was
varied from 4.7 - 10.4 mg L-1 on Lake Tana at Welala and Shesher wetland. Thus, the current
trend of dissolved oxygen level was found between previously reported results like Tarekgne
et al. (2012). The lowest level of dissolved oxygen may have resulted from reduced solubility
of the gas associated with the high temperature and increasing turbidity due to oxidative
decomposition of organic matter during 2012 to 2018 (Welch, and Jacoby, 2004). According
to WHO (2011) dissolved oxygen greater than 3 mg L-1 is enough to fish in the aquatic
ecosystem.

The present trend of high mean nitrate level (mg L-1) in water varied 0.71 to 1.83 at Lake
Tana. Ayalew et al., (2007) reported relatively lower nitrate levels that range from 0.2 to 0.6
mg L-1 at open water of Lake Tana. Akoma (2010) reported a higher level of nitrate (1.92 -
2.61 mg L-1) than in the present observed trend level of nitrate. However, the nitrate
concentrations of the present study are below the maximum level tolerable by fish and other
aquatic organisms (<5 mg L-1, Chattopadhyay and Banerjee 2007) and (<50 mg L-1, WHO,
1996). According to Talling and Lemoalle (1998), the concentration of nitrite is always much
lower than nitrate and ammonia-N. Thus, the present result of nitrite level (0.007 - 0.06 mg L-
1
) was confirmed those facts, and the maximum value was lower than guideline set of WHO
(2011) (max 5 mg L-1). The mean value of ammonia-N level was observed 0.01 to 0.21 mg L-
1
at Lake Tana. This value was low due to well- oxygen shallow water oxidation of nitrate
and uptake of phytoplankton and macrophyte (Eppley et al. 1969; Liao and Lean, 1978).

343
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The comparison of the present result of ammonia-N level with previously reported, the
previous result is found in the range of the present result (0.175 - 0.15 mg L-1, Wondie, 2013)
at Lake Tana. The present trend of soluble reactive phosphate (mean mg L-1) value was
ranged from 0.09 - 0.65 at Lake Tana. This result contrasts with Ayalew et al. 2007 (<0.1 mg
L-1) reported on open water at Lake Tana and strongly agreed with Wondie, 2013 (0 - 0.65
mg L-1) reported on non-infested area of water hyacinth at Lake Tana. The Present study
range of total hardness and alkalinity (mean, mg L-1 CaCO3) was 47.3 - 84.31 and 35.4 - 71.9
at Lake Tana respectively. According to WHO (2011) and APHA, 1999, this result indicates
that Lake Tana is grouped under fresh water scale and specially alkalinity indicates that the
lake has a weak buffer zone, however, the lake water is safe to aquatic life.

The mean of silica level was 0.35 - 14.03 mg L-1 at Lake Tana. This high level of silica might
come from the greater mobility of silica in soil (Talling, 1965, Talling, 1992). Wondie,
(2013) has reported silica levels (mg L-1), which was uncommonly low (0.04 - 3.6) for
tropical lakes (Talling, 1965, Talling, 1992). The current mean of hydrogen sulfide was 0.026
to 0.098 mg L-1 at Lake Tana during the study period. According to WHO (2011) the value
greater than 0.05 mg L-1 was not good for Aquatic life and human drinking water. This high
level of hydrogen sulfide might come from domestic, urban effluents and high decomposition
of macrophytes, especially invasive weeds like water hyacinth.

Thus, the high levels of nutrients in the lake might result in the attribution to alteration of
aquatic ecosystems (Reynolds, 1994; Talling and Wood, 1988). The source of high
macronutrients especially phosphorus and nitrate might be from the terrestrial environment
where by runoff water conveys fertilizers and other inputs applied from agricultural lands.
The domestic influent and industrial waste waters might cause an internal loading and mixed
by wind driven waves (Levinton, 2013; Welch and Jacoby, 2004). According to Welch and
Jacoby (2004) high nutrient changes to the aquatic ecosystem food webs lead to appear
harmful algal blooms, including toxin-producing species.

344
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

The trend of physico-chemical characteristics of Lake Tana was varied temporally. The trend
of pH at Lake Tana indicated that the lake water fluctuated into neutral and slightly alkaline
pH level during the study period. Hydrogen sulfide (H2S) concentration increased from year
to year and the representative sites of the Lake Tana depth indicated that the depth declined
from time to time. The decline of depth of the lake might be the negative impact of human
activities and climatic conditions might change by different factors like drought. The trend of
nutrients indicated that high levels of nutrients like NO3, NO2, SiO2, and soluble reactive
phosphors (SRP) entered into the lake year to year. The high levels of nutrients indicated the
presence of a weak buffer zone in Lake Tana. These high levels might lead to enhancement
of the eutrophication process and affects higher trophic organisms including fish. Therefore,
the physico-chemical characteristics of the lake were apparently in continuous change.

Thus, high level of nutrient in Lake Tana clearly reflect environmental degradation and
exposed to anthropogenic effects, therefore it’s is necessary to ensure the protection of
aquatic and terrestrial life in the lake as well as the risk of the lake by controlling nutrient
input and protect the buffer zone of the lake. To protect the health of Lake Tana and
sustainable use of its resources, it is necessary to take measures on the application of
agricultural inputs of nutrients by erosion and effluents entering to the lake by delineating and
developing the buffer zone of Lake Tana. To reduce the impact of human activities on the
aquatic ecosystem, Lake Tana leads to reduce the input of nutrients. Limnological data will
be important like heavy metals and pesticide for monitoring and management of the Lake
Tana aquatic ecosystem.

REFERENCES

American Public Health Association (APHA), 1999. Standard methods for the examination of
water and Wastewater 20th edn. New York, pp. 1-360.
Ayalew Wondie, Seyom Mengistu, Vijverberg, J. and Eshete Dejen, 2007. Seasonal variation in
primary production of a large high altitude tropical lake (Lake Tana, Ethiopia): Effects of
nutrient availability and water transparency. Aquatic Ecology, 41:195-207.
Baxter R.M., Prossor M.V., Talling J.F. and Wood R. B. 1965. Stratification in tropical African
lakes at moderate altitudes (1500-2000).Limnol. Oceanogr., 10: 510- 520.
Brook Lemma, 1994. Changes in the limnological behavior of a tropical African explosion crater
lake: L. Hora-Kilole, Ethiopia. Limnologica 24 (1):57-70.
Chattopadhyay C. and Banerjee T.C. 2007. Temporal changes in environmental characteristics
and diversity of net phytoplankton in a freshwater lake. Turk J Bot., 31: 287–296.

345
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Dersseh Minychl G., Kibret Aron A., Tilahun Seifu A., Worqlul Abeyou W., Moges Mamaru A.,
Dagnew Dessalegn C., Abebe Wubneh B. and Melesse Assefa M. 2019. Potential of Water
Hyacinth Infestation on Lake Tana, Ethiopia: A Prediction Using a GIS-Based Multi-Criteria
Technique. Water, doi: 10.
Elizabeth Kebede and Amha Belay, 1994. Species composition and plankton biomass in a
tropical African lake (Lake Awassa, Ethiopia). Hydrobiology, 288:13-32.
Elizabeth Kebede and Willén, E. 1998. Phytoplankton in a salinity-alkalinity series of lakes in the
Ethiopian Rift Valley. Algological studies, 89: 63-96.
Eppley R.W., Rogers J.N. and McCarthy J.J. 1969. Half-saturation constants for uptake of nitrate
and ammonium by marine phytoplankton. Limnol. Ocanogr., 14: 912-920.
Food and Agriculture Organization of the United Nations (FAO), 2016. Water resources of
Ethiopia.
Federal Democratic Republic of Ethiopia Ministry of Water Resources (FDREMWR), 2002.
Water sector development program, main report volume i, pp. 193.
Fesseha Hailu Mekonnen, 1988. Liquid waste management: The case of Bahir Dar, Ethiopia.
J.Prof. Geogr., 40: 1-17.
Greboval D., Bellemans M. and Fryd M. 1994. Fisheries Characteristics of the Hared Lakes of
the East African Rift. CIFA Technical Paper No. 24. Rome: FAO.
Kalff J. 2002. Limnology: Inland water Ecosystems. Prentice-Hall, Inc, NJ, pp.1-592.
Lee J.H.W., Wu R.S.S. and Cheung Y.K. 1991. Forecasting of dissolved oxygen in the Marine
Fish Culture Zone. J. Env. Engrg.117:816-836.
Levinton J.S. 2013. Marine biology: function, biodiversity, ecology. Oxford university press,
New York, pp. 173- 193.
Liao, C.F.H. and Lean, D. R.S. 1978. Nitrogen transformations within the trophogenic zone of
lakes. J. Fish. Res. Board Can., 35:1102-1108.
Mamaru Brook, Desta Hayal, 2016. Review of the natural conditions and anthropogenic threats to
the Ethiopian Rift Valley. J. Lakes & Reservoirs Research & Management, 21: 133-151.
Mays L. W. 1996. Water Resources Handbook. McGraw-Hill. New York, pp: 92.
Mohr P.A. 1961. The geology, structure and origin of the Bishoftu explosion crater lakes, Shoa,
Ethiopia. Bull. Geophys. Obs., Addis Ababa 2:65-101.
Prosser, M.V., Wood, R.B. and Baxter, R.M. 1968. The five Bishoftu crater lakes. SINET:
Ethiop. J. Sci. 8: 9-29.
Reynolds C.E. 1987. Cyanobacterial water blooms. In: Callow, J. (Ed.). Advances in Botanical
Research, Vol. 13, Academic Press, London, pp. 67-143.
Reynolds C.S. 1994. The long, the short and the stalled: on the attributes of phytoplankton
selected by physical mixing in lakes and rivers. Hydrobiologia, 289: 9-21.
Schindle D.W. 1974. Eutrophication and recovery in experimental lakes. 184 (4139):897-9.
Vollenweider A. 1991. A Manual on Methods for Measuring Primary Production in Aquatic
Environments. Int. Biol Book Of Roses.
Shang S., Hu C., Du K., Weidemann A., Hou W., Lin J. and Lin G. 2015. Secchi disk depth: A
new theory and mechanistic model for underwater visibility. Remote Sen. of Env. 169: 139-
149.
Strayer D.L. and Findlay S.E.G. 2010. Ecology of freshwater shore zones. Aqu. Sci. 72:127–163.
Talling J.F. 1992. Environmental regulation in African shallow lakes and wetlands. Rev.
Hydrobiol. Trop., 25: 87-144.

346
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Talling J.F. and Talling I.B. 1965. The chemical composition of African lake waters. Int. Rev.
Gesamten. Hydrobiol. 50: 421–463.
Talling J.F. and Lemoalle J. 1998. Ecological dynamics of tropical inland waters. Cambridge
University Press, Cambridge, pp. 1-350.
Talling J.F., Wood R.B., Prosser M.V. and Baxter R.M. 1973. The upper limit of photosynthetic
productivity by phytoplankton: evidence from Ethiopian Soda lakes. Freshwater Biol., 3:53 -
76.
Tamar Z. and Ilia O. 2011. Ecological impacts of excessive water level fluctuations in stratified
freshwater lakes, Inland Waters, 1: 47-59.
Tarekegn Wondmagegne, Ayalew Wondie, Minwyelet Mingist, and Jacobus V. 2012. Seasonality
in Abundance, Biomass and Production of the Phytoplankton of Welala and Shesher
Wetlands, Lake Tana Sub-Basin (Ethiopia). J. Wat. Resour. Prot., 4: 877-884.
Tewodros Taffese, Seifu Tilahun, Tammo Steenhuis, 2014. Phosphorus modeling, in Lake Tana
Basin, Ethiopia. J. Environ. Human, 2: 47 - 55.
Tudorancea C. and Taylor W., Leiden W. 2002. Ethiopian Rift valley lakes: Inland water.
Vijverberg J., Sibbing F. A. and Dejen, E. 2009. Lake Tana: Source of the Blue Nile. In H. J.
Dumont, The Nile: Origin, Environments, Limnology and Human Use (p. 818). Dordrecht:
Springer Netherlands.
Wassie Anteneh, Minwyelet Mingist, Ayalew Wondie, Dereje Tewabe, Woldegebriel W/Kidan,
Addisalem Assefa and Wondie Engida, 2014. Water hyacinth coverage survey report on
Lake Tana. Technical Report Series 1.
Welch E.B. and Jacoby J.M. 2004. Pollutants effects in fresh water applied limnology. 3rd ed.,
Spon press, Wagshington, 515pp.
Wetzel R.G. and Likens G.E. 2000. Limnological Analyses. 3rd ed.Verlag, Inc. New York, 360pp.
Wetzel R.G. 2001. Limnology: Lake and River Ecosystems. 3rd ed. Academic Press. N.Y.
Wondie Zelalem, 2013. Assessment of Water Hyacinth (Eichhornia crassipes (Mart) Solms) in
Relation to Water Quality, Composition and Abundance of Plankton and Macro-
invertebrates in the north-eastern part of Lake Tana. M.Sc. Thesis, Addis Ababa University,
Addis Ababa, 105 pp.
Wood R.B. and Talling J.F. 1988. Chemical and algal relationships in salinity series of Ethiopian
inland waters. Hydrobiologia 15: 29– 67.
Wood R.B., Prosser M.V. and Baxter R.M. 1976. The seasonal pattern of thermal characteristic
of four Bishoftu crater lakes, Ethiopia. Fresh water. Biol. 6: 519-530.
World Health Organization (WHO), 2011. Guidelines for drinking water quality. 4th edition,
Gutenberg publisher, Malta, 564pp.
World Health Organization (WHO), 1996. Guidelines for drinking water quality. 2nd edition,
volume 2. WHO Geneva.
Zinabu Gebre-Mariam, 1994. Long–term changes in indices of chemical and productivity status
of a group of tropical Ethiopian lakes, with differing exposure to human influence.
Arch.Hydrobiologia. 132(1):115-125.
Zinabu Gebre-Mariam, 1998. Science in Africa - Emerging water management problems. In: J.
Schoneboom (ed.), Human interactions and water quality in the Horn of Africa, pp 47-61.
Zinabu Gebre-Mariam, Elizabeth Kebede and Zerihun Desta, 2000. Long-term changes in
chemical features of waters of seven Ethiopian rift-valley lakes. Hydrobiologia.477: 81–91.

347
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Frame and Catch Assessment Survey of Lake Tana Fisheries, Ethiopia

Dereje Tewabe and Erkie Asmare

Bahir Dar Fisheries and Other Aquatic Life Research Center, P.O. Box 794, Bahir Dar, Ethiopia

ABSTRACT

Although fisheries of Lake Tana is accredited for its multi-functionality in improving the
livelihood of various actors across the fish value chain, the amount of catch from this Lake is
highly deteriorating from year to year. Therefore, this study was initiated to collect reliable
data on fishing activities and to conduct catch assessment surveys. To meet these objectives,
the data were collected in 2016 using a census survey. In this study, about 5403 fishers were
identified from all bordering Woredas of Lake Tana. Among these fishers, 2480 of them were
full time participants whereas 2923 fishers had been engaged on a part time and seasonal
basis. The main fishing net used in the area was a monofilament gillnet with a total length of
2,958,500 meter. On the other hand, the two mainly used fishing boat types were reed and
motorized boats and this can be quantified as 4,957 and 196 boats, respectively. In this Lake,
the catch per unit of effort (CPUE) of fishers from reed-boat was about 4.4kg/trip. On the
other hand, the CPUE from the motorized boat was about 19.97 kg/trip. Interestingly, the
average number of fishing days per annum is estimated to be 286 days/year. Therefore, the
aggregated annual catch from Lake Tana fishery was estimated to be 7,357.33 tons/year.

Keywords: commercial fishery, CPUE, fishers, monofilament gillnet, reed boat, and survey

INTRODUCTION

Ethiopia has abundant fresh water bodies with a potential of 51,500 tons of fish per year
(Agumassie, 2018). Based on the resource endowment, Ethiopia has the potential to be self-
sufficient in fish production for both home consumption and export. Among the contributing
lakes, Lake Tana has 28 fish species with a predominant cyprinids fish family. Besides the
biodiversity roles, six thousand tones, which valued 65 million Ethiopian birr earned from
products of Lake Tana commercial fishery (ANRSLRDPA, 2011). Traditionally, Lake Tana
fisheries consisted of mainly artisanal papyrus reed boats (Tankwa).

348
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Since 1986, a motorized commercial gillnet fishery has developed in Lake Tana, targeting
commercial catches of large barbs and tilapia. Following this, the catch per unit of effort
(CPUE) of Labeobarbus species was 63 kg/trip in 1991 - 1993 (Brehan et al., 2011).
Gradually, the CPUE dropped from 28 to 6 kg/trip in 2001 and 2010 respectively (Brehan et
al., 2011). The main reason for such sharp decline in annual catch is due to targeting of
spawning aggregation at river mouths and migratory routes during fish migration for
spawning (de Graaf et al., 2004). In the last decade, commercial catch of riverine spawning of
Labeobarbus species declined by 75%. In addition, populations of juvenile Labeobarbus in
the littoral (length range: 5 - 18 cm) decreased even by more than 85% (de Graaf et al.,
2006). The major reason for the collapse of these fish species is due to the destructive way of
fishing during their spawning season and ground. Due to this, there is a tendency of drastic
reduction in fish catches from Lake Tana.

Therefore, there is a need to devise a scientifically sound management plan for the fishery of
Lake Tana. In this regard, consistent time series data on the amount of catch, effort and gears
are essential for rational management decisions. However, to devise such a strategy there is
one challenging issue that fisheries of Lake Tana have taken place with a little systematic
recording. This creates another obstacle in estimating the current trends of fishing effort,
fishing activities and fish catches over time.

Hence, regular frame surveys are the tools that can provide data on the state of the fishery at
any particular time. In addition, this frame survey is one of the methods to get the
aforementioned basic data. Therefore, this study was conducted to meet the following
objectives.

• To collect reliable and time bounded census of fishing activities

• To estimate the annual actual catch, CPUE, and MSY of the Lake Tana fisheries
• To set ground for the sustainable fishery management of Lake Tana based on the MSY
and CPUE

349
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of the Study Area

Lake Tana, the source of the Blue Nile, is Ethiopia’s largest Lake and it covers an area of
about 3150 km2. It also has an average depth of 8 m, with a maximum of 14 m (Demeke,
1986). This Lake is situated at an altitude of 1830 meter above sea level and can be
characterized as oligo-mesotrophic Lake (Demeke, 1986). Several large and small rivers
including Gumara, Ribb, Megech, Gilgel Abay, Arno Garno and Dirma enter the lake, and
the Blue Nile River is its only outflow (Figure 1).

The climate around Lake Tana can be characterized by four seasons: 1) a main rainy season
with heavy rains during June - August, 2) a dry season during December - February, 3) a pre-
rainy season during March - May and 4) a post-rainy season during September - November.
After the main rainy season, there are two cropping seasons. The mean annual rainfall of the
lake region varies from 74.6 mm to 163.3 mm with a mean value of 166.6 mm (Erkie et al.,
2016). The long-term rainfall distribution data (from 1982 to 2014) showed that most of the
rain occurs starting from June. Data collected from 1961 to 2012 showed the mean annual air
temperature of the Lake area varied between 16.96oC and 21.24oC.

Figure 1: Bordering Woreda of Lake Tana

350
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

As depicted in Figure 1 and Table 1, the data for this study were collected from 3 zones and
10 Woredas, which encompass Lake Tana. Therefore, fishers from Bahir Dar Zuria, Bahir
Dar town, Dera, Fogera, Libo Kemkem, Gondar Zuria, Dembia, Alefa, Takusa and Semein
Achefer Woredas and their respective Kebeles were the target population.

Table 1: List of Woredas and their respective Kebeles1 involving in Lake Tana fishery
List of woredas Potential Kebeles engaged in fishing
Fogera Nabegageorgis and Wagetera
Libo Kemkem Tezamba, Agidkirigna and Kab
Gondar Zuria Mitreha Abawerka, Sheha Gomengie, Lemba and Firka Dangurie
Dembia TanaWoina, Jarjar, Achera, Abrjeha, Fentay, Debirzuria, Serabadablo,
Wawa, Mangie, Gurandie and Tezeba
Alefa Ahcha Mangur, Esey Debir and Dengel Ber
Bahir Dar Zuria Dek, Derbanta, Lijoymie, Robit, Sekelet, Woonjeta and Gobat
Bahir Dar town Zegie, Oura, Shimbit, Zenzelma, Yiganda, Woramit and Sefeneselam
Dera Mitseli, Mirafemariam and Korata
Takusa Delgy, Mekonta, Achera, Chemera, Chachinaalwa and Chankie
North Achefer Chinba,Legdia, Kunzila, Wenbera eyesus and Estumit

Data Types, Sources and Methods of Data Collection

For this study, both quantitative and qualitative data types were collected using primary and
secondary data sources. The primary data were collected from fishermen using semi-
structured questionnaires. On the other hand, secondary data were collected from research
articles, books, proceedings, working papers and institutional reports. Before the data
collection, Semi-structured questionnaire was developed and adequate training was provided
for each Woreda’s livestock and fishery resource development process owners and fishery
experts to supervise the data collection process.

In this regard, two researchers from Bahir Dar Fisheries and Other Aquatic Life Research
Centre were routinely involved from the beginning to the end. In addition, a total of 83 data
collectors and supervisors with the regular supervision of the aforementioned researchers
were involved for frame survey data collection processes. Therefore, the study was conducted
from 2016 by targeting all the landing sites of Lake Tana.

1
Kebele is the lowest administrative unit of the government system in Ethiopia. It refers to
peasant associations and may contain several villages

351
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sampling Methods

In the fisheries of Lake Tana, many stakeholders are involved in various stages of the fishery
process. To collect holistic and reliable information on the fishing practices from the 10
bordering Woredas of the lake, a total census method was employed. This census survey was
made at 10 Woredas and 51 Kebeles in which fishing is potentially practiced in Lake Tana.
As a result, sample frame data for this study was collected from 5403 fishers in 2016.

Data Analysis

Based on the format developed during the inception periods of this study, both frame survey
and catch assessment survey have been implemented. Descriptive statistics such as mean, and
percentage were computed to explain fishing practice and methods, amount of fishing gear
and type, fishing frequency etc. As a result, catch per unit of effort (CPUE) by fishing unit
type, mean total number of fishing days per year, mean length of caught fish by species and
estimation of total yield of the Lake per annum had been analyzed. Based on the information
obtained from census survey, catch assessment survey was estimated using the following
formula:

CFUx = CPUFFUx * BAFUx * DFFUx * NDUx

Where: CFUx is the total catch of fishing unit, CPUFFUx is the catch per unit of effort, BAFUx is
the boat activity, DFFUx is day of fishing per year (day of the year minus non-fishing day) and
NDUx is the total number of operational fishing unit.

RESULT AND DISCUSSIONS

Fishing Practice and Type in the Fisheries of Lake Tana

Based on the prepared data collection format, the total number of fishers and their fishing
practice were identified at each potential Woreda. Hence, in the fisheries of Lake Tana, there
were about 10 Woredas, which have the potential fishery resources. Specifically, 51 Kebeles
from the 10 Woredas are participating in fishing activities on a full, part and seasonal basis
(Tables 1 and 2). Therefore, this study found about 5403 individual fishers who were
involved in fishing practices in a variety of ways. In this regard, of the total fishers, 2480
(49.2%) fishers were engaged in the fishing process on a full time basis, meaning that their
life is highly dependent on fishing activities, whereas, about 2923 (50.8%) fishers had been
involved in the fisheries of Lake Tana as part time and seasonal fishers (Table 2).

352
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

For the part time and seasonal fishers, fishing was considered as a supplement to their off-
farm income generating activity. It is also evident that crop production in line with animal
rearing is a predominantly practiced agricultural activity in the study area. However, it does
not mean that the contribution of fishing to the annual households’ income is negligible.
Rather, it is the main source of immediate cash income or “bank in the water” for the
households in the study area. All this shows how fishing plays a vital role in livelihood
diversification and means of risk reduction during agricultural crop failure.

Table 2: Number of fishers, motorized and reed boats, length of fishing gears in each Woreda
Fishing No Total length Boat type Type of engagement
Woredas Kebeles Fishers monofilament (m) Reed Motor Full Part & Seasonal
Bahir Dar Zuria 7 477 23, 8500 477 43 434
Bahir Dar town 7 190 190000 86 52 190 -
North Achefer 5 146 73000 146 146 -
Alefa 3 129 64500 129 39 90
Dembiya 11 481 240500 481 81 400
Takusa 6 270 297000 269 1 160 110
Gonder Zuria 4 658 329000 656 2 149 509
Libo Kemkem 3 2284 1142000 2281 3 1230 1054
Fogera 2 532 266000 352 60 210 322
Dera 3 236 118000 80 78 232 4
Total 51 5,403 2,958,500 4,957 196 2,480 2,923

Fishing Gear Type and Fishing License

In Lake Tana fishery, the main fishing gear type used for commercial and subsistence fishing
was monofilament gill net and hook, respectively. This destructive and nonselective fishing
gear (monofilament) is brought from Egypt through Sudanese via black market. Rumors
indicate that this gear is being sold in every shop along the Lake Tana fishery value chains
starting from Sudan to Lake Tana Islands like Zegie and Dek. According to this study, the
total estimated length of monofilament gillnet used in the fisheries of Lake Tana was about
2,958,500 meter (Table 2). This implies that about 32 lines could be set every night from
north to south directions of Lake Tana.

On the other hand, the fishing boats, which are predominantly used were reed and motorized
steel boats. The total number of motorized boats used in the lake was estimated to be 196,
whereas about 4957 reed boats had been deployed for fishing activities.

353
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Up to the time of data collection, from the ten Woredas, which encompass Lake Tana, only
Gondar Zuria and North Achefer had given fishing licenses for about 256 fishers. These
licensed fishers are mainly members of Lake Tana Fishers’ cooperative. Even with this
license, there is a doubt that these fishers merely take the fishing license for the sake of
formality without respecting the fishing regulations.

Annual Catch and Catch per Unit of Effort

The number of days that fishers engaged in fishing was estimated to be 5 - 6 days per week.
Accordingly, the average number of days in which fishers undertake fishing was about 286
days per year. Within these fishing days, there was a significant variation in catch per unit of
effort (CPUE) between the two dominantly used boat types. Thus, the estimated catch per
unit of effort of fishers who use a reed boat was about 4.4kg/trip. Whereas, the CPUE was
estimated to be 19.97kg/trip for those who used motorized fishing boats. Accordingly, the
aggregated amount of catch from Lake Tana fisheries was about 7357.33 tons/year.

Length Weight Relationship and Fishing Pressure

The finding of this study reveals that the highest frequency by length of different species
caught by the commercial fishery falls in the size range of 20 to 26cm for both C. gariepinus
and Labeobarbus species. On the other hand, from the daily fish catch composition of fishers,
the size of O. niloticus ranged from 14 to 20cm (Figure 2). This finding is also inconsistent
with the finding of de Graaf et al. (2006) that the length range of juvenile Labeobarbus falls
from 5 to 18 cm. This shows that the size of juvenile Labeobarbus is decreased by more than
85%.

The average length at which 50% of the L. intermedius males reached maturity was 25.9 cm
FL while the length at which 50% of the females attained sexual maturity was 35.7 cm FL.
The average length at which 50% of the O. niloticus males reached maturity was 23.4 cm TL
while the length at which 50% of the females attain sexual maturity was 21.2 cm TL. The
average length at which 50% of the C. gariepinus males reached maturity was 43.2 cm TL
while the length at which 50% of the females attain sexual maturity was 57.7 cm TL. On top
of this, the sizes at 50% maturity of the Labeobarbus, O. niloticus and C. gariepinus of
females were much higher than reported by other investigators of Lake Tana (Tesfaye
Wudneh, 1998; De Graaf, 2003) which was about 31 cm, 17.5 cm and 30 cm, respectively.

354
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 2: Commercial fish catch length frequency by Species

Generally, size at maturity is negatively correlated to the degree of fishing mortality; hence,
the size of caught fish in commercial fishery indicates that there is high fishing pressure.
Lake Tana fish production and fishing effort have been increasing steadily in the last few
years from 756 tons (2000) to 9980 tons (2013). The number of fishermen also increased
from 1700 in 2000 to 3514 in 2013 and 5,403 in 2017.

In addition, there were about 4,957 reed boats and 194 motorized boats, which are involved
in fishing activities. All this clearly shows the high level of pressure exerted on the fisheries
of Lake Tana. This is also exacerbated by the nature of open-access fisheries and increasing
fishers with no license at all. The reason is that all fishers strive to catch as much as they can
before their competitors do so. This process leads to changes in fishing strategies, increased
fishing effort and changes in the type and number of fishing gears. Finally, the fishery of
Lake Tana exposes over exploitation and imminent extinction of some fish species.

355
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSION AND RECOMMENDATIONS

In the fisheries of Lake Tana, annual fish production has drastically declined after 2013,
which was about 9,980 tons per annum with 3514 fishers. The present study shows that
fishing effort is continuously increasing from 3,514 in 2013 to 5,403 in 2017, whereas the
amount of fish produced severely declined from 9,980 to 7,357 tons per year. This implies
that high fishing effort in the fishery of Lake Tana becomes the cause for the deterioration of
the amount of catch. This is also evident that the number of fishers in Lake Tana increased by
1,889 in the last five years. Sadly, the fishing gear type used by the commercial fishers is also
the most destructive and nonselective monofilament gillnet. The consequence of this illegal
gear is also aggravated by open-access and non-licensed ways of fishing.

Therefore, control measures are necessary because open-access fisheries may be by a race in
which all participants strive to catch as much as they can before their competitors do so. This
open-access fishery also leads to changes in fishing strategies, increased fishing effort and
changes in the type and number of fishing gears. Decision-making in fisheries management
should be, therefore, based on the scientific evaluation of fishing effort and the stock. In
addition, there must be an endeavor to ensure the sustainability of this resource. Finally and
most importantly, the regional government should work hard in rescuing the imminent
extinction of the Lake’s fishery resource by implementing a fishing license, close area and
season and by banning the non-selective and destructive fishing gears.

REFERENCES

Agumassie T. 2018. Review of current problems of Ethiopian fishery : Incase of human and
natural associated impacts on water bodies. International Journal of Fisheries and
Aquatic Studies, 6(2), 94–99.
ANRSLRDPA, 2011. Lake Tana fisheries management plan and processing manual, Bahir
Dar, Ethiopia.
Brehan M., Martin de G., Leo N., Wassie A. and Minwyelet M. 2011. Lake Tana’s (Ethiopia)
endemic Labeobarbus spp. Flock: An uncertain future threatened by exploitation, land
use and water resources developments. In L. Brook & G. Abebe (Eds.), Impacts of
climate change and population on tropical aquatic resources, proceedings of the Third
International Conference of the Ethiopian Fisheries and Aquatic Sciences Association
(EFASA) (pp. 285–297). Addis Ababa: AAU Printing Press.

356
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Demeke Admasu, 1986. Report on Limnological Studies on Lake Tana Addis Ababa,
Ethiopia: Addis Ababa University”. Dept. of Biology, internal report: 1-44.
de Graaf M., van Zwieten P.A.M., Machiels M.A.M., Lemma E., Tesfaye Wudneh, Eshete
Dejen and Sibbing F.A. 2006. Vulnerability to a small-scale commercial fishery of Lake
Tana’s (Ethiopia) endemic Labeobarbus compared with African catfish and Nile tilapia:
An example of recruitment overfishing? Fisheries Research 82: 304-318
Erkie Asmare, Sewmehon Demissie, Dereje Tewabe and Mihret Endalew, 2016. Impact of
climate change and anthropogenic activities on livelihood of fishing community around
Lake Tana, Ethiopia”. EC Agriculture, 3 (1): 548-557.

357
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Controlling Overpopulation of Nile Tilapia (Oreocromis Niloticus) through Polyculture

System

Mohammed Oumer, Dereje Tewabe, Benyam Hailu and Adane Melaku

Bahir Dar Fisheries and Other Aquatic Life Research Center, P.O. Box 794, Bahir Dar, Ethiopia

ABSTRACT

The experiment was conducted at the North Achefer Woreda, Euhiri Kelteafa Kebele, Klti
nursery site to assess the efficiency of African catfish (Clarias gariepinus) in controlling
unwanted Nile tilapia (Oreochromis niloticus) reproduction in grow out and to evaluate the
performance of tilapia in polyculture system with catfish under low-input production system
from July to December 2018. Mixed-sex Nile tilapia fry (25g size) were stocked at rate of 2
fish/m2 in all treatments. Stocking catfish fingerlings (30 g size) were introduced into ponds
at stocking 10% of Nile tilapia. All ponds were fertilized with a similar amount of dung at a
rate of 10 - 15 kg/ 100m2 per week. For every 50 Nile tilapia fingerlings stocked, 5 catfish
fingerlings were stocked. There was no significant difference (p<0.05) seen in physical and
chemical parameters among the treatments. The result at the end of the experimental period
indicated that catfish significantly (P<0.05) reduced by 34% in the production of Nile tilapia.
Tilapia production was significantly (P<0.05) higher in the case of tilapia only treatment
compared to the polyculture. Overall, the present experiment showed that under such
stocking rate and production system, introduction of catfish at the rate of 10% of total tilapia
stocked has not only eliminated 34% of total tilapia recruitment but also enhanced total pond
production of marketable size fish in the polyculture treatment.

Keywords: oreochromis niloticus, overpopulation, polyculture, stocking rate

INTRODUCTION

Overcrowding in pond culture caused by the prolific reproduction of Nile tilapia

(Oreochromis niloticus) results in competition for food and consequent yields composed
mainly of small fish of low market value (Lin, 1996). Nile tilapia is currently the most widely
cultivated finfish species in Africa. However, a major disadvantage of this species and tilapia
in general is their excessive reproduction and at harvest up to 23% of the biomass may
consist of fingerlings (Kaewpaitoon, 1992). The main problem with the existence of the
fingerlings is that they compete for and consume the feed provided for the adult tilapia and
consequently the growth rate of the adults is reduced (Gertjan et al., 1996).

358
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Real stunting takes place at low feeding levels but can be avoided by higher feeding levels or
by elimination of recruitment through polyculture with either the African catfish (GAFRD,
2005). Tilapia is frequently cultured with other species to take advantage of many natural
foods available in ponds and to produce a secondary crop, or to control tilapia recruitment
(GAFRD, 2005). Polyculture uses a combination of species that have different feeding niches
to increase overall production without a corresponding increase in the quantity of
supplemental feed (Gertjan et al., 1996).

Large catfish (6.8 - 130 g) were able to control the recruitment of Nile tilapia completely,
with less than 0.15% of fingerlings (as % of total harvested biomass) remaining at a stocking
density of 8 300 catfish/ha (Gertjan et al., 1996). However, small catfish (less than 3.6 g)
were not able to completely control the recruitment of Nile tilapia (a fingerling percentage of
3.7% persisting at the end of the culture period). The major reason for this was that small-
sized catfish have a specific food preference for zooplankton and probably shift to a more
piscivorous behavior once they reach a weight of 7 - 8 g (Gertjan et al., 1996).

The main problem facing tilapia producers is the early sexual maturity and fry production
before fish reach marketable size, which lead to overpopulation in production ponds and
produce smaller fish at harvest (Guerrero, 1980). Of the major limitations for the use of
predators’ population control methods, is the difficulty in obtaining stock of desirable size. On
the other hand, the new development of catfish spawning methodology which makes attaining
catfish fingerlings more practical to be stocked in a polyculture system with tilapia (El Gamal,
2007).

Therefore, there is a need for focusing on alternative methods for controlling tilapia
reproduction in ponds. This experiment was designed to determine the suitability of African
catfish as a predator to control unwanted tilapia offspring in a polyculture system and to
determine the effect of polyculture on tilapia growth under a low input production system in
geomembrane ponds.

Objectives

• To evaluate the growth performance of Nile tilapia in polyculture systems

• To assess the effects of monoculture and polyculture systems on water quality

359
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

The experiment was conducted at North Achefer Woreda, Euhiri Kelteafa Kebele at Klti
nursery site located 11°28’14.614”N and 36°57’24.585”E with an elevation of 1909.6 masl in
four geomembrane ponds of similar size (25m2) each, from July to December 2018.

Experimental Design

Ponds were assigned into two treatments replicates each, as follows:

• Treatment 1 (T1): Pond stocked with Nile tilapia and catfish

• Treatment 2 (T2): Pond stocked with Nile tilapia only (T. Only)

Experimental Fish

Nile tilapia (O. niloticus) fry were obtained from Bahir Dar Fishery and other Aquatic Life
Research Center. Catfish fingerlings (C. gariepinus) were obtained from different water
bodies. Mixed-sex Nile tilapia fry (25 g size) were stocked at rate of 2 fish/m2 for all
treatments, due to the production system was under low input. Stocking catfish fingerlings
(30 g size) were introduced into ponds at stocking 10% of Nile tilapia. All ponds were
fertilized with a similar amount of dung at a rate of 10 - 15 kg/100m2 per week.

For every 50 Nile tilapia fingerlings stocked, 5 catfish fingerlings were stocked and 30gm of
catfish size and 25gm of Nile tilapia were stocked. About 50% (100) of the stocked fish and
water quality data were sampled monthly to monitor the growth of fish. Mortality of fish was
monitored and recorded continuously throughout the experiment. At the end of the
experiment, the volume of the water was reduced and all fish were harvested using beach
seine.

Growth performances of fish were determined in terms of final individual weight (g). Growth
parameters were calculated following standard equations given below (Adebayo et al., 2004).

Relative growth rate (RGR) (Busacker et al., 1990):

RGR = [(W2 –W1) / (W1 x T)] x 100 (% / day)

Where: W1 is the initial weight at the start of the studied period (g), W2 is the final weight at
the end of the studied period (g), T is the time of the studied.

360
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Daily growth rate DGR (g/day) = [Final weight (g) – Initial weight (g)]/culture period

Weight gain (g) = Final weight (g) – Initial weight (g)

Survival rate (%) = (Number of fish harvested/Number of fish stocked) x100

Data Analysis

Data were analyzed by analysis of variance (ANOVA) using SPSS (version 8.0) statistical
software package (SPSS, Inc., Chicago, Illinois, USA). Differences were considered
significant at an alpha level of 0.05.

RESULTS

Growth Performances

Among the growth parameters of O. niloticus final weight, weight gain, final length, daily
growth and specific growth showed a significant difference (P<0.05) between treatments
(Table 1).

Production of Nile tilapia had the highest mean weight on polyculture (catfish and Nile
tilapia) compared with monoculture (only Nile tilapia). Tilapia fry production decreased
significantly with increasing stocking rate of catfish (T. + 10% CF) where total number of
tilapia reproduction at harvest was 34% of fish yield, respectively (P<0.05).

Nile tilapia survival was not significantly affected (p>0.05) with introduction of catfish into
grow out ponds with catfish and tilapia only 96.5 to 96.2 %, respectively.

Table 1: Daily and relative growth rate, mean weight, survival rate and tilapia fry biomass for
different treatments (sample size N = 25, for each pond)
T1 (Nile Tilapia and T2 (Nile Tilapia
Parameters Catfish) only) P-Value
Initial weight (g) 25.96±0.95 25.76±0.34 P>0.05
Final weight (g) 55.2±1.76 41±1.41 P<0.05
Weight gain (g) 29.24±0.81 15.24±1.07 P<0.05
Daily growth rate (g/day) 0.19±0.01 0.13±0.01 P<0.05
Relative growth rate (RGR) 0.53±0.06 0.41±0.04 P<0.05
Survival rate (%) 96.5 95.2 P>0.05
Tilapia fry Biomass 359 1045 P<0.05

361
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Physico-Chemical Parameters

The results of physico-chemical parameters during current study were in acceptable range for
the growth of Nile tilapia. There was no significant difference (p>0.05) seen in physical
parameters among the treatments (Table 2). The values of nitrates ranged from 70 - 80 mg/L,
phosphorus 1.3 - 1.4 mg/L and alkalinity 68 - 78 mg/L among the treatments and were not
significantly different (P>0.05) among treatments (Table 2).

Table 2: Recorded water quality parameters (mean±SE) throughout the experiment

Treatments
*
Parameters N T1 (Nile Tilapia and Catfish) T2 (Nile Tilapia only) P-Value
Temperature (oC) 16 23.75±2.20 22.88±2.36 p>0.05
Conductivity 16 107.5±14.5 108.3±12.34 p>0.05
Total dissolved solids 16 0.05±0.02 0.06±0.01 P>0.05
Salinity 16 0.05±.010 0.04±.006 p>0.05
Dissolved Oxygen 16 6.46±1.061 6.5±1.035 p>0.05
pH 15 7.82±1.165 7.79±1.188 p>0.05
Alkalinity (mg L-1) 6 77.5±3.54 68±11.31 p>0.05
-1
Nitrite (mg L ) 6 1.4±0.57 0.78±1.03 p>0.05
Nitrate (mg L-1) 6 84±8.49 70±14.14 p>0.05
-1
Phosphorus (mg L ) 6 1.3±0.42 1.4±0.14 p>0.05
*
N = Number of samples

DISCUSSION

The present study showed that polyculture of catfish with mixed sex tilapia in geomembrane
ponds under low input system (fertilization only) would impact tilapia reproduction and
significantly reduce tilapia fingerlings. Tilapia fry production decreased significantly with
stocking rate of catfish (T. +10% CF) than Nile tilapia only. Similarly, EI Gamal (2007)
indicated that polyculture of catfish with mixed sex tilapia in earthen ponds under low input
system (fertilization only) would impact tilapia reproduction and significantly reduce tilapia
fry biomass. Tilapia fry production decreases significantly with increasing stocking rate of
catfish (T.only, T. +7% CF, and T. +13% CF) where biomass of tilapia reproduction at
harvest is 8, 14.9 and 26.6% of fish yield, respectively (P<0.05) (EI Gamal, 2007).

The introduction of catfish at the rate of 13% of total tilapia stocked has not only eliminated
70% of total tilapia recruitment but also enhanced total pond production of marketable size
fish in both of the polyculture treatments (EI Gamal, 2007). In this study the total number of
tilapia reproductions controlled with 10% catfish was 34% of fish yield.

362
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

The efficiency of catfish and Nile perch in controlling tilapia reproduction was investigated
by El-Gamal et al. (1998) who reported that in earthen ponds when artificial feed was offered
to fish, catfish were consuming artificial feed and attributed the less predator performance of
catfish

The obtained result of this experiment among the growth parameters of O. niloticus final
weight, weight gain, final length, daily growth and specific growth showed a significant
difference (P<0.05). Similar work reported that, there were significant differences in all
growth yield, tilapia production, tilapia fry biomass, tilapia fry biomass as percentage of fish
yield, market size tilapia yield, market size catfish quantity and mean weight of catfish
marketable size among treatments (P<0.05) (EI Gamal, 2007).

Tilapia production was significantly affected (P<0.05) by introduction of catfish into tilapia
ponds either at 7 or 13%, (T2 or T3, respectively) with no significant difference between
treatment T2 and T3 (EI Gamal, 2007). In the present study tilapia production was
significantly affected by introduction of catfish into tilapia ponds at T. + 10% CF. Lin (1996)
reported similar results and concluded that polyculture of tilapia with African catfish would
reduce tilapia yield than in monoculture of tilapia. The result of this experiment disagrees
with that obtained by Ngugi et al. (2006), who found that final weight and yield of market
size tilapia was higher in ponds stocked with tilapia: catfish at 2:1 than those ponds stocked at
6:1 and 19:1.

Tilapia survival and daily weight were not affected with introduction of catfish into grow out
ponds and ranged from 95.2 to 96.5% and 0.13 to 0.19, respectively. Similar works on tilapia
survival were not affected with introduction of catfish into grow out ponds and ranged from
87.8 to 90.2% (EI Gamal, 2007). Also daily weight gain of tilapia was not significantly
different (P<0.05) and ranged between 0.89 to 0.97 g/day/fish.

There was no significant difference seen in physical parameters among the treatments.
Similarly, EI Gamal (2007) reported a non-significant difference (P>0.05) resulting in water
quality parameters such as dissolved oxygen, alkalinity, NH3, nitrate and chlorophyll-a. The
value of pH in all treatments has more or less similar reading ranging between 7.1 and 8.6
(Solomon and Boro, 2010). In the present study, the physico-chemical parameters were
observed in acceptable range for the growth of Nile tilapia.

363
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

CONCLUSIONS AND RECOMMENDATIONS

In conclusion, this study has clearly demonstrated that introduction of catfish at the rate of
10% of total tilapia stocked has not only eliminated 34% of total tilapia recruitment but also
enhanced total pond production of marketable size fish in both of the polyculture treatments.
As recommendation further research should be focused on the following areas:

• The assessment of the effects of increased variation in tilapia and catfish stocking ratios,
• The bio economics of the polyculture versus the monoculture of catfish and Tilapia and
only male tilapia should be used for the experiment.

REFERENCES

Adebayo O.T., Fagbenro O.A. and Jegede T. 2004. Evaluation of Cassia fistula meal as a
replacement of soya bean meal in practical diets of Oreochromis niloticus fingerlings.
Aquaculture. Nutr. 9: 99-104.
Busacker G.P, I.R. Adelman and E.M.Goolish. 1990. Fish growth. In: C.B. Shereck and P. Moyle
editors. Methods for fish Biology. American Fisheries Society, Maryland.pp:363-387.
El Gamal A.A., Abdel-Halim A.E. and Soliman A. 1998. Biological studies on Nile perch Lates
niloticus (L.) and African catfish Clarias gariepinus (T.) in reference to their food habits and
predation patterns in culture ponds. Egypt. J. Agric. Res., 76 (1): 335-349.
El Gamal Naggar, 2007. Efficiency of African catfish Clarias gariepinus in controlling unwanted
reproduction of Nile Tilapia (Oreochromis niloticus) in low input production system, Egypt.
J. Aquat. Biol. & Fish., Vol. 11, No.3:105: 113, ISSN 1110 – 6131
GAFRD (General Authority for Fishery Resources Development), 2005. Statistics of Fish
Production. GAFRD, Ministry of Agriculture and Land Reclamation.
Gertjan de Graaf, Felix G & Bernared B. 1996. Recruitment control of Nile tilapia, Oreochromis
niloticus, by the African catfish, Clarias gariepinus (Burchell 1822), and the African
snakehead, Ophiocephalus obscuris. I. A biological analysis. Aquaculture, 146(1-2): 85-100.
Guerrero R.D. 1980. Control of tilapia reproduction. In: R.S.V. Pullin and R.H. Lowe-McConnell
(Editors), the Biology and Culture of Tilapia, ICLARM Conference Proceedings 7, Manila,
Philippines, pp. 309-317.
Kaewpaitoon K. 1992. Utilization of septage-raised tilapia (Oreochromis niloticus) as feed for
snakehead. PhD Dissertation; Asian Institute of Technology, Bangkok, Thailand, 212 pp.
Lin C.K. 1996. Clarias and Tilapia interaction in Polyculture. CRSP research reports: Department
Of Biological Sciences Faculty of Science, University Of Abuja, Nigeria (+234), 94pp.
Solomon J.R. and Boro S.G. 2010. Survival rate in polyculture of catfish Heteroclarias /Tilapia
(Oreochromis niloticus), fed 2% body weight. New York, Science Journal 2010; 3(9).
Ngugi C., Nyanchiri E., Rasowo J. and Bowman J. 2006. The effect of different stocking ratios
on yields of tilapia (Oreochromis niloticus) and African catfish (Clarias gariepinus) in
fertilized earthen ponds. AQUA 2006 International Conference and Exhibition, Firenze
(Florence), Italy, May 9-13, 2006.

364
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Diversity, Abundance of Fishes and some Physico-Chemical Characteristics in Lake

Tirba, Awi Zone, Ethiopia

Benyam Hailu1, Minwyelet Mingist2, Wassie Anteneh3 and Dereje Tewabe1

1
Bahir Dar Fisheries and Other Aquatic Life Research Center, P.O. Box 794, Bahir Dar, Ethiopia
2
School of Fisheries and Wildlife, Bahir Dar University, P.O. Box 5501, Bahir Dar, Ethiopia
3
Food and Agriculture Organization, P.O. Box 5536, Addis Ababa, Ethiopia

ABSTRACT

This study was aimed at assessing the diversity, relative abundance and some physico-
chemical parameters of Lake Tirba. The study was conducted from September 2016 to
August 2017. Four sampling sites were selected based on habitat of the lake, interferences,
and vegetation and substrate type of the sediments. Fish specimen collection was carried out
at four sampling sites twice at wet season (between October and August) and twice in dry
season (between November and March) in sampling sites. The physico-chemical parameters
were measured using portable (YSI-54) instruments and the fish were sampled at each site
using gillnet with mesh size of 6, 8, 10 and 12 cm. Measuring board and sensitive balance
was used to measure length (cm) and weight (gm), respectively. Index of relative importance
(IRI) and Shannon diversity index (H') were used to evaluate abundance and diversity of fish
species. A total of 1074 fish specimens were collected from the Lake in both wet and dry
seasons. Three families, 5 genera and 5 species of fishes were caught in this study. The
Shannon diversity index value (H' = 0.47) was higher in wet season than dry season (H' =
0.46) in most of the sampling sites. Clarias gariepinus, Oreochromis niloticus and Garra
dembecha were the dominant and abundant species in most of the sampling sites. There was a
significant difference (P<0.05) in % IRI of O. niloticus and C. gariepinus between sites and
seasons. The length-weight relationship of the dominant species O. niloticus and C.
gariepinus were curvilinear and the relation was statistically significant (P<0.05).

Keywords: Biology, Diversity, Fish, Lake Tirba, Relative abundance

365
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Ethiopia is uniquely rich and diverse in water resources, endowed with a number of lakes and
rivers with substantial quantities of fish stocks. Based on the estimation of FAO (2003) the
surface area of major lakes and reservoirs is 7,334 km2 and the length of rivers is 7,185 km.
According to the report of the Ministry of Agriculture and Rural Development (MoARD,
2000: FAO, 2003) the total fish catch on the shore from lakes, rivers, and dams was about
85%, 9% and 6%, respectively for the year 2000. Therefore, the contribution of the lakes is
very much significant. Moreover, it is this small-scale fishery sub-sector that provides the
bulk of fish production and consumption (Tadesse, 2006). Though the diversity of the
Ethiopian fish fauna is not fully researched, these water bodies give a refuge for more than
200 species in 29 families of which around 40 of them are endemic to Ethiopia (Abebe, 2005;
Abebe et al., 2008; Redeat, 2012; Abebe, 2017).

The Amhara region has high fish production potential from the major lakes (Tana, Logo, and
Ardibo) and other small water bodies. Moreover, the region has also a considerable fish
potential from other lakes and rivers (Abebe et al., 2008; Dereje, 2008). Relatively major
lakes and rivers in the Amhara region were investigated for their fish diversity. However,
some small water bodies are not well studied. Tirba Lake is one among the crater lakes which
probably can support fish and other biodiversity in the region. Fishes are commercially
important biota that can improve the livelihood of the people around the Lake. However,
there are no previous studies conducted about the fish diversity, fisheries and their biology in
this Lake. Therefore, this study was aimed to address the following objectives.

Objectives

• To assess the fish species composition, diversity, relative abundances and some biology
of fishes and
• To provide relevant information for the management of the fisheries to circumvent the
risk of fish stock collapse.

366
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of the Study Area

Lake Tirba is found in Ankesha Guagusa Woreda, situated at an altitude of 2350 meters
above sea level and located at 10°49´42ʺN and 036°50´39.3ʺE. It is one of the 10 Woreda
found in Awi administrative zone of Amhara regional state in the Northwestern part of
Ethiopia. Agew Gimjabet is its capital town placed 450 km far from Addis Ababa, the capital
of Ethiopia, and 145 km from Bahir Dar city (MCT, 2009). The Woreda has a total of 32
Kebeles (19 from Ankesha Kebeles and 13 from Guagusa Kebeles) and the estimated total
population living in this Woreda was 217,941. Lake Tirba is a crater lake that is found in
Bohona or Trinity Kebele. It is a deep crater lake formed by volcanic explosion and collapse.
The lake has a maximum depth of 198 m, diameter of 2.5 km and total area of 54 ha (The
measurements were taken during the study period).

Description of Sampling Sites

Site selection for sampling was done by considering nature and the habitat of the lake,
considering the interferences of humans and other farm animals, substrate type of the
sediments and undisturbed areas and in grass or vegetated area. The samples were collected
both in wet and dry seasons. At each sampling site, samples were taken twice at wet season
(October and August) and twice in dry season (November and March) at all sampling sites in
the lake.

Sampling site 1

Entrance (Megibiyaw): It is a human impacted area as it is the entry place for a lot of
livestock and local communities. They use the lake for swimming and bathing in this area.
The coordinate of the sampling site is 10°49′38.4″N and 036°50′50.6″E direction with an
elevation of 2206 m.a.s.l.

Sampling site 2

Cave (Jib Washa): It is covered with permanent densely populated vegetation cover and
protected area and also there was a cave or palace for a lot of Hyena in this area. The
coordinate of the sampling site was 10°49′42.4″N and 036°50′39.3″E direction with an
elevation of 2206 m.a.s.l.

367
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sampling site 3

Holy water (Tsebelu): It is covered with highly densely populated vegetation and covered
with grass. In addition in the area there was inlet of the groundwater to the lake and spring
Holy water used for the community spiritual activity. The coordinate of the sampling site was
10°49′55.4″N and 036°51′03.4″E direction with an elevation of 2206 m.a.s.l.

Sampling site 4

Open water (pelagic area): It is the fourth sampling area which is open deep and clear water.
The coordinate of the sampling site was 10°49′38.6″N and 036°50′47.5″E direction with an
elevation of 2206 m.a.s.l (Figure 1 and Table 1).

Table 1: Description of sampling sites

Sampling Sites Lat Long X Y
Entrance (Megibiyaw) 10.82733 36.84739 264653 1197720
Cave (Jib Washa) 10.82844 36.84425 264310 1197840
Holy water (Tsebelu) 10.83206 36.85094 265045 1198240
Open water (pelagic area) 10.82739 36.84653 264559 1197730

Figure 1: Map of the study sites.

368
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Sampling Methods and Data Collection

Fish Sampling

For the fish specimen collection, gillnet of 6, 8, 10, 12 and 14 cm stretched mesh sizes were
set overnight at each sampling site. In addition, monofilaments gillnet with mesh size of 0.5 –
0.55 cm was set on the lake for one and half hours to sample small sized fishes. The fish
specimens were sorted into different species and sizes; and a picture of the specimen was
taken for each species. Finally, total and standard length and weight of all specimens of fish
were measured to the nearest 1 cm and 0.1 gm, respectively.

Diversity and Relative Abundance

Shannon diversity index (H') was used to evaluate species diversity of sampling sites and
season in the lake. Shannon diversity index explains both variety and the relative abundance
of fish species (Naesje et al., 2004). The species composition of gillnet and monofilament
catches both in wet and dry season ranked based on the index of relative importance (IRI) for
different sampling sites was given in (Table 2). IRI and H' was used to evaluate relative
abundance and diversity of fish, respectively. IRI is a measure of relative abundance or
commonness of the species based on number and weight of individuals in catches as well as
their frequency of occurrence (Kolding, 1989). IRI gives a better replacement of the
ecologically important species rather than the weight, number or frequency alone (Kolding,
1989).

(%Wi + % Ni) %fi

IRI = ∗ 100
∑(Wj + Nj) ∗ % Fj

Where: % Wi and % Ni are percentage weight and number of each species of total catch,
respectively. % Fi is a percentage frequency occurrence of each species in total number of
settings, % Wj and Nj are percentage weight and number of total species in total catch.

Shannon Diversity Index (H'): is a measure of species weighed by the relative abundance
(Begon et al., 1990). H' was used to indicate diversity of fishes at different sampling sites. H'
was calculated using the formula below.

H' = -∑ Pi * lnpi

Where: pi is the proportion of individuals in the ith species.

369
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Length-Weight Relationship: The relationship between total length and total weight of most
dominant fish will be calculated using power function as in Bagenal and Tesch (1978).

TW = aTLb

Where: TW is the total weight (gm), TL is the total length (cm), a is the intercept of
regression line, b is the slope of regression line.

Statistical Analysis

All the data were analyzed using SPSS (version 20) to determine if there exist changes in
composition and abundance between sites and seasons. Excel of window 2007, ANOVA,
regression, and some descriptive statistics were used to present the analyzed values. Values
were considered significant at 5% level.

RESULT

Fish Species Composition

In this study, a total of 1056 fish specimens were collected from the Lake in both seasons.
The fish species composition of Lake Tirba were from three family represented by five
species namely, from Cyprinidae family Cyprinous Carpio, Varichorinous beso, Garra
dembecha and from Cichlidae family Oreochromis niloticus from Clariidae family Clarias
gariepinus. Clarias gariepinus, O. niloticus and G. dembecha were found in almost all sites
except in open water sites during both sampling seasons. The fish species composition was
dominated by the family Cichlidae (O. niloticus) followed by Clarias gariepinus.

Diversity and Relative Abundances

A total number of 1056 specimens and 425.6 kg of fishes were caught from the study sites in
wet and dry seasons. The Shannon diversity index (H') varied spatio-temporally (Table 2).
The entrance and the Holy water site of the lake had relatively high H' values in both seasons.
The test statistics showed that there was a significant difference in H′ and number of species
(N) between seasons and sites (P<0.05). Generally, the lake had less species diversity than
rift valley crater lakes in the country.

370
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 2: Shannon diversity index and number of fish species during wet and dry seasons
Dry season Wet season
Site Number H′ P value Number H' P value
Entrances 4 0.16 0.139 5 0.16 0.139
Cave 2 0.14 0.182 3 0.15 0.182
Holy water 4 0.16 0.139 5 0.16 0.139
Open water 0 0.00 0.391 0 0.00 0.391
Total 10 0.46 13 0.47
Note: Computed using alpha (P<0.05).

In Lake Tirba, all the sites showed similar dominance trends with order of O. niloticus, C.
gariepinus, and G. dembecha (Table 3). However, the IRI values of fishes vary in each site
between seasons. Generally, the Cichlidae and Clariidae family were the dominant family in
the entire sampling site and Cyprinidae family was the second dominant family in the entire
sampling site in the total catch.

Table 3: Percentage IRI of fishes in Lake Tirba

Site Fish Species N N% W W% F %F IRI %IRI
Entrance O. niloticus 211 51.1 59 37.8 4 100 8890.0 45.6
C. gariepinus 179 43.3 80 51.3 4 100 9461.0 47.9
G. dembecha 23 5.6 17.042 10.9 2 50 824.5 6.5
Total 413 156 4 7510.8
Cave O. niloticus 122 51.5 54 50.4 4 100 10190.6 53.3
C. gariepinus 103 43.5 53 49.5 4 100 9295.5 45.4
G. dembecha 12 5.1 0.081 0.1 4 50 256.9 1.3
Total 237 107.1 4 7897.1
Holy water O. niloticus 224 55.2 73 44.9 4 100 10008.4 52.1
C. gariepinus 161 39.7 74 45.5 4 100 8518.2 42.3
G. dembecha 21 5.2 15.543 9.6 4 50 736.7 5.6
Total 406 162.5 4 7575.3
Open water O. niloticus 0 0 0 0 0 0 0 0
C. gariepinus 0 0 0 0 0 0 0 0
G. dembecha 0 0 0 0 0 0 0 0
Total 0 0 0 0 0 0 0 0
Note: W is weight of the fish in kg, F is frequency of sampling, and IRI is Index of relative
importance.

371
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Length Frequency of the Dominant Species

In Lake Tirba, O. niloticus was the abundant species that had a total length ranging from 12.6
cm to 35.3 cm in wet and from 15.0 cm to 35.2 cm in dry season. The second abundant fish
species, C. gariepinus, had a total length ranging from 15.6 cm to 66.1 cm in wet and from
15.1 cm to 67.5 cm in dry season. The third dominant species, G. dembecha had a total length
ranging from 3.5 cm to 17.1 cm in wet and from 3.1 cm to 16.5 cm in dry season. The overall
length frequency distribution of the two most abundant species was shown in Figure 2.

120
Frequancey (Number)

100

80

60

40

20

0
14.9 15.0-17.9 18.0-20.9 21.0-23.9 24.0-26.9 27.0-29.9 30 - 32.9 33

Total Length (cm)

120
Frequancey (Number)

100

80

60

40

20

Total Length (cm)

Figure 2: Length frequency distribution of O. niloticus (N = 327, top) and C. gariepinus (N =
380, bottom) in Lake Tirba.

372
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Length-Weight Relationship

The relationship between total length and total weight for the dominant fish species, O.
niloticus and C. gariepinus was curvilinear and the line fitted to the data was described by the
regression equation. The length-weight relationship of O. niloticus and C. gariepinus in Lake
Tirba showed positive allometric growth (Figures 3).

800 TW = 0.041TL2.72
700 R² = 0.92
Totla weight (gm)

N = 174
600 P < 0.05
500
400
300
200
100
0
0 5 10 15 20 25 30 35 40
Total length (cm)

1800
1600 TW = 0.038TL2.49
R² = 0.85
1400
Total weight (gm)

N = 181
1200 P < 0.05
1000
800
600
400
200
0
0 20 40 60 80

Total length (cm)

Figure 3: Length-weight relationships of O. niloticus (top) and C. gariepinus (bottom) in

Lake Tirba.

373
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

DISCUSSION

In agreement with studies on Lake Babogaya and Lake Hayq (Lemma, 2007), O. niloticus
showed nearly allometric growth like that of Lake Tirba but dissimilar result (isometric
growth) of O. niloticus fish reported by Zenebe (1997) in Lake Tana. Clarias gariepinus in
Lake Tirba showed positive allometric growth, similar with the results reported on previous
studies for Lakes Tana (Abebe, 2004) and Babogaya (Lemma, 2007) and also similar result
found nearly positive allometric growth of C. gariepinus in Lake Naivasha Kenya (Anyanwu
et al., 2007). Garra dembecha in Lake Tirba showed similar positive allometric growth like
that of Lake Tana (Akwake, 2007) but dissimilar with the report of Borkena Rivers (Assefa,
2010).

CONCLUSION AND RECOMMENDATION

From the current study, five fish species: Clarias gariepinus, O. niloticus, G. dembecha, C.
Carpio and V. beso were found in Lake Tirba. Oreochromis niloticus and C. gariepinus were
the dominant and abundant fish species in sampling sites and seasons. The Shannon diversity
index indicates that the entrance site and Holy water site had higher fish diversity than the
cave site. In the study time in terms of % IRI of wet season C. gariepinus, O. niloticus and G.
dembecha were abundant species at Entrance site; O. niloticus, C. gariepinus and G.
dembecha were dominant species at Cave site; and at Holy water site the same species in
different proportion were abundant. The length-weight relationship for the dominant fish
species, O. niloticus and C. gariepinus and G. dembecha was curvilinear. With the above
conclusions, the following recommendations were amended.

• Further studies on the food feeding, reproductive biology of fish species and macro-
invertebrates shall be undertaken and
• Stock assessment, age and growth, for sustainable utilization of the resources closing
season, appropriate (legal type) of fishing gear and protection of breeding area and season
will be implemented in the lake.

374
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENT

We would like to thank Amhara Agricultural Research Institute for funding this project. Mr.
Getnet Temsgen (Research Assistant) and Sebat Molla (driver) were acknowledged for their
unreserved support during data collection. We are grateful to Dr Koos (J.) from Vijverberg,
Netherlands Institute of Ecology (NIOO-KNAW) for his valuable advice and comments on
the study.

REFERENCES

Abebe Getahun. 2005. An overview of the diversity and conservation status of the Ethiopian
freshwater fish fauna. In: proceeding of the Pan-African Fish and Fisheries Society,
Cotonoue, Benin, Nov. 2003.
Abebe Ameha. 2004. The effect of Birbira, Milletia ferruginea (Hochst.) Baker on some
Barbus spp. (Cyprinidae, Teleostei) in Gumara River (Lake Tana), Ethiopia Pp.24-25.
Abebe Getahun. 2017. Freshwater fishes of Ethiopia: Diversity and Utilization. View
Graphics and Printing PLC, Addis Ababa. Pp. 1-349.
Abebe Getahun, Esthete Dejene and Wassie Anteneh. 2008. Fishery studies of Rib River,
Lake Tana Basin, Ethiopia Final Report Vol. 2. Presented to the World Bank-financed
Ethiopian-Nile Irrigation and Drainage Project Coordination Office, Ministry of Water
Resources: January 2008 Addis Ababa, Ethiopia.
Akwake Geremew. 2007. Taxonomic revision, relative abundance and aspects of the biology
of some species of the genus Garra, Hamilton 1922 (Pisces: Cyprinid) in Lake Tana,
Ethiopia. MSc. Thesis, Addis Ababa University.
Anyanwu P.E., Okoro B.C., Anyanwu A.O., Matanmi M.A., Ebonwu B.I. and Ayabu-
Cookey I.K. 2007. Length–Weight relationship, condition factor and sex ratio of African
mud catfish (Clarias gariepinus) reared in indoor water recirculation system tanks,
Research Journal of Biological Sciences. 2007; 2(7):780-783.
Assefa Tessema. 2010. Diversity, Relative Abundance and Biology of Fishes in Borkena and
Mille Rivers, A wash Basin, Ethiopia. MSc. thesis, Bahir Dar University.
Bagenal T.B. and Tesh F.W. 1978. Age and growth. IN Bagenal.T.B. B (ed.). Methods for
assessment of fish production in freshwater. Black well Oxford, New York, 1o1-136 Pp.
Begon M., Harper J.L. and Townsend C.R. 1990. Ecology: individuals, Population and
communities. 2nd edt Black well, scientific publication, 945pp. 2001, University Saints
Malaysia, Penang, Malaysia. 6 pp.
Dereje Tewabe. 2008. Diversity Relative abundance and Biology of fishes in Gendwuha,
Guang, Shinfa and Ayima Rivers, North Gondar Ethiopia. M.sc Thesis, Addis Ababa
University, Ethiopia.
FAO. 2003. Information on fisheries management in the Federal Democratic Republic of
Ethiopia, http://www.fao.org/fi/fcp/en/ETH/ body.htm.
Kolding J. 1989. The fish resource of Lake Turkana and their environment. Thesis.
University of Bergen, Norway, 262, Pp.

375
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Lemma Abera. 2007. Length-weight relationship, sex ratio, length at maturity and condition
factor of African catfish Clarias gariepinus in Lake Babogaya, Ethiopia. MSc Thesis
Addis Ababa University, Ethiopia.
Ministry of Culture and Tourism, 2009. Tourism Development Policy. Addis Ababa,
Ethiopia: Birana Printed Enterprise.
Naesje. T.F., Hay C.J.J., Nickansor N., Koekemoer J.H., Strand R. and Thorstad E.B. 2004.
Fish population, gillnet catch and gillnet selectivity in the Kwando River. Namibia.
Norwegian Institute for Nature research Tungasletta 2. No-7485.
Redeat Habteselassie. 2012. Fishes of Ethiopia. Annotated checklist with pictorial
identification guide. Ethiopian Fisheries and Aquatic Science Association, Addis Ababa.
Pp.1-250.
Ricker W.E. 1975. Computation of interpretation of biological statistics of fish population.
Bull. Fish Board of Canada, 191: 382.
Tadesse Shado. 2006. Challenges of Sustainable Livelihood: The Case of Fishing
Communities Around Lake Chamo, Addis Ababa University School of Graduate Studies
the Degree of Masters of Arts in Development Studies.
Zenebe Tadesse. 1997. Breeding season, fecundity, Length-weight relationship and condition
factor of Oreochromis niloticus L. (Pisces: Cichlidae) in Lake Tana, Ethiopia. SINET:
Ethiop. J. Sci. 20(1): 31-47.

376
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Participatory Evaluation of some Fish Drier Methods in the Surrounding of Tekeze

Reservoir, Wag-Himra Zone, Ethiopia

Wubet Biset, Melese Mamo, Alachew Adino and Ademe Mihiretu

Sekota Dryland Agricultural Research Center, P.O. Box 62, Sekota, Ethiopia

ABSTRACT

This study aimed to evaluate and select the best technology from different fish drying
methods through fishers’ participation. Data were collected from the observation and training
program done for 14 cooperatives in 3 districts from April 2018 to March 2019. A total of 28
fishers and 26 sector participants involved in the participatory approach. Three treatments:
solar tent drying with salting, open-air drying with salting and open-air drying were
evaluated. The collected primary data was summarized and analyzed by descriptive statistics.
The temperature record inside the solar tent was in the range of 22.6 to 52.9oC and the
environmental air temperature was in the range of 25.5 to 42.9oC, whereas the average
humidity was under 40% that was “Dry”. The moisture content of the Cyprinidae family in a
solar tent dryer significantly differed from the other treatments, but it was not significant for
the Cichlidae family. Fisher's perception on solar tent drier (T1) was acceptable on most
statements like short duration for drying, low moisture content, good quality (taste, color, and
odor), long shelf life, and easy reduction post-harvest loss. However, fishers had less
confidence to use open-air drying with salting (T2) and open-air drying (T3) due to the
vulnerability of these methods for animals, insects, birds, dust, rainfall and short shelf life
with low quality. Therefore, a solar tent dryer in the form of a large scale like a greenhouse
on a multi-rack dome is the best option to reduce and avoid the discarding and dumping of
the Cyprinidae family.

Keywords: dry salting, fishers, moisture, perception, solar tent

377
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

INTRODUCTION

Fish drying is presumably the oldest method of fish preservation using heat from the sun and
atmospheric air, although it has been limited to certain climatic areas and seasons (Ogali and
Eyo, 1996). However, sun drying may be exposed to problems like contamination by dust
and insect infestation because the fishes are dried on mats spread on bare ground. To arrest
these problems, many designs of solar dryers have been developed for the preservation of fish
(Olokor and Omojowo, 2009). Solar drying refers to methods of using solar energy for
drying. A solar dryer is an enclosed unit, to keep the food safe from damages, birds, insects
and unexpected rainfall (Balasuadhakar et al., 2016). Improved solar tent drying has been
suggested for an alternative to the traditional open-air sun drying method. The solar tent
works by concentrating solar radiation resulting in increased temperature in the tent and turn,
lower humidity. With a solar tent, the drying rate can be increased and lower moisture
content can be attained with higher quality product (Assefa et al., 2008).

Solar tent fish dryers in Ethiopia were performed in different study areas mainly practiced in
at pre-scaling and demonstration activities. From these more activities are performed around
Northern and North Western part of Lake Tana (Dembia, Alefa and Gondar Zuria district)
(Erkie et al., 2015). According to his findings, the solar tent fish dryer was prepared from
readily available simple materials such as wood, white and black plastic, rope, nail and mesh
wire. From the study the sample was practiced on Labeobarbus intermedius fish species and
drying fish by solar tent fish dryer enables to produce hygienic, high quality, organoleptically
good dried fish with low cost. By drying quickly it is possible to reduce post-harvest losses,
thereby (in that case) ensuring continuous availability of cheap animal protein.

In the same manner, another study from the district of Lake Tana area, which was selected
Labeo Barbus, Clarias gariepinus and Oreochromis niloticus fish species as a sample,
reported a high quality dried fish product with long shelf life by using a solar tend dryer
(Assefa et al., 2008). However, the fishing practice in Ethiopia depends on some selected
species and others discarded (as a leftover) such as the Cyprinidae family. In addition, poor
quality and unsafe handling practices are common which exposes fish and fish products for
high post-harvest losses. Currently, this problem is pronounced at Tekeze Reservoir fisheries.
In view of this, this investigation aims to evaluate and select the best technology from the
drying methods and assess fishers’ preference towards the drying methods.

378
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

MATERIALS AND METHODS

Description of the Study Area

The study was conducted in the Eastern and Southern parts of the Tekeze Hydropower
Reservoir. The Reservoir has a maximum length and width of 75 km and 6 km, respectively,
about 160.4 km2 areas and an average depth of 58 m (Goraw et al., 2009; Tsegay et al.,
2015). It is located at an altitude of 1145 m.a.s.l and an annual average rainfall of 150 - 700
mm and temperature of 15 - 40oC (Hussein et al., 2016). It is bordered by five districts,
namely Abergele, Ziquala, Sahila and Telemt in the Amhara region and Tanqua-Abergele in
the Tigray region. The main purpose of constructing the reservoir was to produce electricity,
but fisheries were later recognized as of significant socio-economic importance (Tsegay et
al., 2015).

Two landing sites were selected based on appropriateness for the training program in the way
of clustered together (from the nearest landing), i.e. 5 kebeles from Kirchifen landing site (4
from Abergele and 1 from Ziquala district), and 6 kebeles from Anteneh-Giba landing site (4
from Sehala and 2 from Ziqaula district). Fishers in the Ziquala district have not their own
landing site; they are conducting the fish product from the nearest two landing sites. The
preliminary studies were held from the Kirchifen landing site only while the training
programs were conducted from the two landings.

Kirchifen landing site is found in the Eastern part of the reservoir and is the major/potential
landing site in Abergele district. It is located at 38o50’25.67'' N and 13o03’58.26” E (Figure
1); the average temperature was 31.25oC with an elevation of 1152m.a.s.l. The landing site is
located about 92 km from Sekota town and it is one of the drylands from Wag-Himra Zone.
About 1994 fishers were deployed around this landing site.

Anteneh-Giba landing site is a temporary landing site in the Sehala district from the Southern
part of the reservoir. The landing site is located at 38o34’45.25” N and 12o56’15.79” E with
the distances about 130 km from Sekota Town. The average temperature was 35oC with an
elevation of 1175 m.a.s.l. It is also one of the drylands of the Wag-Himra Zone. About 409
fishers were deployed around this landing site.

379
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 1: Map of the target area in the Tekeze Reservoir

Data Collection

The study was conducted from April 2018 to March 2019; data were collected during
personal observation (preliminary study) and training program. Out of 14 cooperatives of the
three districts; 11 cooperatives participated in the training program; 2 - 3 fishers represented
their cooperatives. A Total of 28 fishers (27 Males and 1 Female) participated and arranged
in 7 groups (3 - 6 fishers in one group) on the training program. Theoretical and practical
training was provided to all groups for 3 days in each landing site to make them aware of
preparing the tent from readily available materials.

Two field days prepared for 26 participants from Zone, Woredas and Sekota Drylland
Agricultural Research Center. In addition to this a lot of volunteer fishers participated side by
side with their fishing activities. A structured and semi-structured questionnaire was
delivered through oral interviews and discussions. Group discussion and personal
observations were also taken in the surroundings of the study area and marketing places that
had been evaluated and noted on fish drying practices. Temperature data recorded on a
prepared datasheet by using SH-117 Digital Thermo-Hygrometer (Yuyao Shuanghe Electron
Instrument Co., Ltd., China) and also pictures and GPS data were collected from the training
site. Secondary data were collected in collaboration with the three districts’ Livestock and
Fish Resource Development Office.

380
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Samples of the fish species targeted at Cichlidae family (like Oreochromis niloticus) and
Cyprinidae family (like Labeobarbus intermedius) were gutted and filleted using a knife,
weighed, and soaked in salt solution with 50 - 60 gram salt dissolved in one liter of water for
2 kg gutted and filleted fish (Assefa et al., 2008 and preliminary study). Weight was taken by
manual balance (before drying) and sensitive balance (after drying) to obtain the calculated
weight losses.

Treatment Design: Three treatments of fish drying methods were performed during
preliminary study and training programs as follows:

T1 = Solar tent drying + Salting

T2 = Open-air drying + Salting
T3 = Open-air drying (control)

First of all, these treatments were conducted in a preliminary study; and then training
programs have performed in a participatory approach for cooperatives (Kebeles) based on
clustering (the nearest landing). Six fish drier tents were prepared, five from Kirchifen and
one from Anteneh-Giba landing sites. The two fish families (Cichlidae and Cyprinidae) were
applied on three treatments duplicated in six replications for each family.

Figure 2: Solar tent dryer constructed at Anteneh-Giba landing site

381
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Temperature Record

The temperatures were recorded in June and December during the preliminary study and the
training programs. During the first observation in June, maximum and minimum
temperatures inside the solar tent were 50°C and 22.6°C, respectively from the Kirchifen
landing site. During the training program in December, the temperature rose to 52.9°C from
the Kirchifen landing site. The maximum and minimum temperatures were 47.4°C and 32°C,
respectively from the Anteneh-Giba landing site at the beginning of January (Table 1).

2
The highest and lowest humidity recorded in all sites was 69 (comfort) and 20 (dry),
respectively, whereas the average humidity was under 40% that was “Dry” (Table 1).

Table 1: Temperature record inside the tent drier

Landing Temperature Humidity in %
Months Sites Record time N Max Min Mean Max Min Mean
June Kirchifen 6:00 am - 6:00 pm 7 50 22.6 33.3 69 20 38.4
December Kirchifen 7:20 am - 5:50 pm 20 52.9 22.9 38.2 49 21 36.4
Anteneh-
January Giba 10:35 am - 5:00 pm 5 47.4 32 40.5 30 20 27.5

During the first observation in June, maximum and minimum temperatures of the
environment were 42.9 and 25.5°C, respectively from the Kirchifen landing site. Besides this
during the training program in December, the temperature reduced to 41.1°C from the
Kirchifen landing site. The maximum and minimum temperatures were 47.4 and 32°C,
respectively from the Anteneh-Giba landing site at the beginning of January (Table 2). The
number of temperature records varied depending upon the time of the day and season; and
also it was not constant record time.

Table 2: Environmental temperature record (outside the tent)

Landing Temperature Humidity in %
Months Sites Record time N Max Min Mean Max Min Mean
June Kirchifen 6:00 am - 6:00 pm 7 42.9 25.5 31 69 20 38.36
December Kirchifen 7:20 am - 5:50 pm 20 41.1 25.6 30.8 49 22 36.6
Anteneh-
January Giba 10:35 am - 5:00 pm 5 37.1 31.4 34.6 30 20 22.4

2
The humidity operated by the description of SH-117 Digital Thermo-Hygrometer as follows: When
the humidity lower than 40%, it displays “dry”; When the humidity higher than 70%, it displays
“wet”; and When the humidity between 40% to 70%, and the indoor temperature between 20 oC to
25oC, it displays “comfort” (Yuyao Shuanghe Electron Instrument Co., Ltd., China).

382
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Data Analysis

The moisture content of fish samples dried with three treatment samples was computed after
oven drying (Reeb and Milota, 1999; and Demeke and Alemu, 2015).

Weight of wet sample − weight of dried sample

Moisture content = 𝑥 100
weight of wet sample

Perception and attitude of fishers were analyzed by a Likert scale scoring (Likert, 1932).

The sum score

Average score =
Sample size

5
The sum score = ∑ Strongly agree, Agree, neutral, Disagree, Strongly disagree
𝑖=1

To know fishers level of agreement on different aspects of the technology scores were given
(Erkie et al., 2015): strongly agree = 5; agree = 4; neutral = 3; disagree = 2; strongly disagree
= 1. A sum score was calculated by using a formula = F. Strongly agree*5 + F. agree*4 + F.
neutral*3 + F. Disagree*2 + F. Strongly disagree*1; where F is the frequency, * is multiple
and + is plus.

Additionally, primary social data like perception and attitude of fishers towards the
technologies were summarized and analyzed by descriptive statistics using SPSS (version 20)
software. Additionally, tables and graphs were done by using Microsoft Excel (2007).

RESULT AND DISCUSION

Duration of Fish Drying

The duration of fish drying from the Kirchifen landing site was 2 - 2.5, 2.5 and 3.5 days for
treatment 1, 2 and 3, respectively. While, from Anteneh-Giba landing site was 3, 2.5 and 4
days for treatment 1, 2 and 3, respectively (Table 3). From the result, drying by solar tent
dryer had a short duration compared to the other methods from the Kirchifen landing site i.e.,
it took place only 2 days at the beginning of December, but 2.5 days in June. The difference
might have been an occasional rainfall occurred in June. However, long-duration for solar
tent and short for dry salting was recorded from the Anteneh-Giba landing site due to a
variety reasons like occasional rainfall, dusty and frost season since the experiment was
conducted at the beginning of January. This implies that the duration of fish drying varies and
is affected by different seasons (Assefa et al., 2008).

383
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 3: Duration of fish drying for each treatment

Duration for drying in day
Treatments Kirchifen landing site Anteneh-Giba landing site
Solar tent drying + salting 2 - 2.5* 3
Open air drying + salting 2.5 2.5
Open-air drying 3.5 4
Note: * 2 days in December and 2.5 days in June

Moisture Content

The moisture content was taken from the sample weight before and after the drying. The
moisture content of the Cichlidae family showed non-significant difference between the
treatments (Table 4). It might be a limitation between measurements by manual balance
during initial raw fish measuring and sensitive balance in the final weight measuring. The
moisture content of the Cyprinidae family showed a significant difference between the
treatments (Table 4).

Table 4: Moisture content between treatments for each sample of fish families
The moisture content of the The moisture content of the
Treatments Cichlidae family (%) Cyprinidae family (%)
Solar tent drying + salting 75.91 71.14b
Open air drying + salting 77.293 78.57a
Open-air drying 76.913 78.64a
LSD 3.77
CV (%) 2.86 3.85

Perceptions of the Fishers

Perceptions of fishers on the availability of inputs (Table 5) to construct the solar tent dryer
were most probably agreed that the mean score was above 3.5, but some participants did not
respond on five statements (i.e., neutral). However, fishers did not yet start to apply the
technology practically.

384
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 5: Fisher’s level of agreement on the availability of inputs to make the solar tent
3
Strongly Strongly Sum Mean
Perception agree Agree Neutral Disagree disagree scores scores
Available 14.3 85.7 - - - 29 4.14
constructing inputs
Not labor intensive 28.6 42.9 28.6 - - 28 4.00
Not expensive to - 57.1 42.9 - - 25 3.57
construct
Easy to implement 28.6 57.1 14.3 - - 29 4.14
Easy to manage 28.6 42.9 28.6 - - 28 4.00
Fish holding capacity - 85.7 14.3 - - 27 3.86
is good
Average 3.95
Cronbach’s alpha coefficient 0.87
Note: the values of scales are in percentage point

The fisher’s perception of treatment 1 (solar tent drying + salting) was acceptable on most of
the statements. However, it did not agree on the demand for marketability and vulnerable for
animals, insects, birds, dust and rain. Even though, some participants were marginal respond
to the high price that predicts the future (Table 6). Likewise, the respondent's and
stakeholder’s attitudes and perceptions towards solar tent dryer were good and acceptable in
most aspects (Erkie et al., 2015). Fisher's perception of the open-air drying with salting was
presented in Table 7. The fisher’s attitude on the open-air drying was highly agreed on
vulnerable for animals, dust and rain; and also agreed on quick drying (Table 8). This
perception seems to contradict the duration of fish drying in treatment 3, but they respond in
the way of their own trends that related to the traditional drying methods. Furthermost, the
other level of agreement was not acceptable that the mean score under 2.50 (50%) from this
treatment (Table 8).

3
The system usability scale on a five-point scale of agreement: mean score greater than 70%
(>3.51): acceptable; between 50 and 70% (2.51-3.50): marginal (neither of the two) and less
than 50% (<2.50): unacceptable (Tullis, 2013).

385
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 6: Fisher’s level of agreement on the solar tent drying + salting

2
Strongly Strongly Sum Mean
Perception agree Agree Neutral Disagree disagree scores scores
Hygienically high 57.1 42.9 - - - 32 4.57
Drying fish quickly 28.6 71.4 - - - 30 4.29
Good taste, texture, odor 42.9 57.1 - - - 31 4.43
and color
The shelf life of the 71.4 28.6 - - - 33 4.71
dried fish is longer
Post-harvest loss 57.1 42.9 - - - 32 4.57
reduced
Demand of - - - 85.7 14.3 13 1.86
marketability
High price 14.3 28.6 14.3 42.0 - 22 3.14
Easy to adopt 14.3 57.1 28.6 - - 27 3.86
technology for the future
Vulnerable for animal, - - - 14.3 85.7 8 1.14
insect, bird, dust and rain
Average 3.65
Cronbach’s alpha coefficient 0.73
Note: the values of scales are in percentage point; and 2 the second footnote

Table 7: Fisher’s level of agreement on the open-air drying + salting

2
Strongly Strongly Sum Mean
Perception agree Agree Neutral Disagree disagree scores scores
Hygienically high - 14.3 14.3 71.4 - 17 2.43
Dried fish quickly - 42.9 57.1 - - 31 4.43
Good taste, texture, - 28.6 71.4 - - 18 2.57
odor and color
The shelf life of the - 42.9 - 42.9 14.3 19 2.71
dried fish is longer
Post-harvest loss - 57.1 - 42.9 - 22 3.14
reduced
Demand of - - - 42.9 57.1 10 1.43
marketability
Vulnerable for 57.1 42.9 - - - 32 4.57
animal, insects,
birds, dust and rain
Average 3.04
Cronbach’s alpha coefficient 0.83
Note: the values of scales are in percentage point; and 2 the second footnote

386
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Table 8: Fisher’s level of agreement on the open-air drying

2
Strongly Strongly Sum Mean
Perception agree Agree Neutral Disagree disagree scores scores
Hygienically high - - 14.3 28.1 57.1 11 1.57
Dried fish quickly - 85.7 - 14.3 - 26 3.71
Good taste, texture, - - - 28.6 71.4 9 1.29
odor and color
The shelf life of the - - - 57.1 42.9 11 1.57
dried fish is longer
Post-harvest loss - - - 42.9 57.1 10 1.43
reduced
Demand of - - - 42.9 57.1 10 1.43
marketability
Vulnerable for animal, 57.1 42.9 - - - 32 4.57
insects, birds, dust and
rain
Average 2.22
Cronbach’s alpha coefficient 0.71
Note: the values of scales are in percentage point; and 2 the second footnote

The reliability and consistency of fisher’s perception on all levels of agreement (Table 5, 6, 7
and 8); the statistical tables showed that the coefficient of alpha, α>0.7 which illustrates the
statements from all categories (items) were consistent and reliable (4).

Experience of Traditional Fish Drying in the Study Areas

According to personal observation, group discussions and individual interviews; fishers

practiced on the open-air drying from the reservoir especially using the Cyprinidae family in
traditional methods (Figure 3). However, different challenges observed like unexpected
rainfall, extreme dusty wind blow, earthworms, clearwings (a type of insect), and market and
transport problems. Besides this problem, some fishers sold their dried fish products from
Kirchifen and Anteneh-Giba landing sites. The price of dried fish from the Anteneh-Giba
landing site was better than the Kirchifen landing site. The wholesalers were collect the dried
fish from the fishers and then resells to ultimate customers i.e., to East Belesa district (Central
Gondar Zone) and Wolkait district (North Gondar Zone)

4 Cronbach’s alpha reliability analysis carried out for internal consistency among Likert scale items.
The reliability coefficient normally ranges between 0 and 1, while a high value for Cronbach’s alpha
indicates good internal consistency of the items in the scale (Gliem and Gliem, 2003). The acceptable
values of alpha coefficient are in the range of 0.7 to 0.95, but the maximum alpha value of 0.90 has
the best. If alpha coefficient is closer to 1.00, as the result gets greater internal consistency of the
items in the scale and alpha coefficients above 0.70 are considered acceptable (Tavakol and Dennick,
2011; and Taherdoost, 2016).

387
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Figure 3: Plate showing traditional methods of fish drying (Cyprinidae (left and middle) and
Cichlidae (right) families)

CONCLUSION AND RECOMMENDATION

In general, the two landing sites had a high ambient temperature, lowest humidity, hottest and
driest land in most months of the year. In the average, drying by solar tent dryer took a short
duration compared to the other methods. The moisture content of the Cyprinidae family
showed that a significant difference between the treatments i.e., solar tent drying with salting
(T1) was significantly different from the other treatment. Also, solar tent dryers are more
acceptable by fishers due to different reasons; it takes a short duration to dry, low moisture
content, good quality, and good handling, long shelf life, and easily reduce post-harvest loss
compared to the other methods. However, some limitations are observed like easily burst
(rupture) of the white plastic during a high blowing of wind and rainfall. Besides this, open-
air drying with salting and open-air drying had also agreed on quick drying. However, many
limitations were observed like vulnerable to the extreme dusty wind blow, rainfall, animals,
birds, clearwings or insect infestation, long duration to dry; and low quality (bad tastes and
odor). Likewise, traditional fish drying methods of the fishers were also exposed to product
loss.

Therefore, to reduce the loss of the Cyprinidae family in the surrounding of the reservoir;
solar tent driers should be demonstrated in the form of a large scale like greenhouses (multi-
rack dome) to deploy a lot of fishers together. In addition to this, to protect the rupture or
burst, hard plastic should be used to construct the tent drier from each landing site. For fishers
of lower capital to purchase the inputs for solar tent, they can optionally use open-air drying
with salting.

388
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

ACKNOWLEDGMENT

Thanks to Abergele, Ziquala and Sihala Woreda Agricultural Offices for their good
participation in the training.

REFERENCES

Assefa Tessema, Sewmohon Demssie, Goraw Goshu, Behailu Bekele, Admasu Fantahun and
Belay Bezabeih. 2008. Evaluation of solar tent and drying rack methods for the
production of quality dried fish in the Lake Tana area. In: Solomon Gizaw, Fisseha
Moges and Halima Hassen (Eds.). 2008. Proceedings of the 3rd Annual Regional
Conference on Completed Livestock Research Activities 01 – 08, September, 2008,
Amhara Agricultural Research Institute, Bahir Dar, Ethiopia.
Balasuadhakar A., Fisseha T., Atenafu A. and Bino B. 2016. A review of passive solar dryers
for agricultural products. International Journal for Innovative Research in Science and
Technology, 3 (01), 64–70.
Demeke Teklu and Alemu Lema. 2015. Optimization of Time and Temperature for Smoking
of Nile Tilapia for a Better Preservation of Protein and Gross Energy Value. J Nutr Food
Sci 5: 341. doi:10.4172/2155-9600.1000341
Erkie Asmare, Dereje Tewabe, Birhan Mohammed and Beniyam Hailu. 2015. Pre-Scaling Up
of Solar Tent Fish Drier in the Northern and North-Western Part of Lake Tana, Ethiopia.
International Journal of Aquaculture and Fishery Sciences, 1, 48–53. https:// doi.org
/10.17352/2455-8400.000009
Gliem J.A. and Gliem R.R. 2003. Calculating, interpreting, and reporting Cronbach’s Alpha
Reliability Coefficient for Likert-type Scales.
Goraw Goshu, Dereje Tewabe and Challachew Aragaw. 2009. Survey of a newly
Constructed Reservoir, Tekeze Hydropower Dam, Ethiopia. In: Proceedings of the First
Annual Conference of EFASA Held at Zwai Fisheries Resources Research Center Zwai,
Ethiopia (pp. 149–165).
Hussien Abegaz, Dereje Tewabe, Challachew Aragaw and Tesfa Gife. 2016. Field
assessment of fish production and marketing of Tekeze Reservoir.
Likert R. 1932. A technique for the measurement of attitudes. archives of psychology, New
York University (pp. 5-55).
Ogali E.L. and Eyo A.A. 1996. Evaluation of the Effectiveness of box-type Solar dryers in
drying freshwater fish in the Kainji Lake area. National Institute of Fresh Fisheries
Research, New Bussa.
Olokor J.O. and Omojowo F.S. 2009. Adaptation and Improvement of a Simple Solar Tent
Dryer to Enhance Fish Drying. Nature and Science, 7 (10), 18–24.
Reeb J. and Milota M. 1999. MOISTURE CONTENT BY THE OVEN-DRY METHOD
FOR INDUSTRIAL TESTING. Oregon State University (pp. 66–74).
Taherdoost H. 2016. Validity and reliability of the research instrument; How to Test the
Validation of a Questionnaire / Survey in a Research. International Journal of Academic
Research in Management, 5 (3), 28–36. https://doi.org/10.2139/ssrn.3205040

389
Proceedings of the 12th Annual Regional Conference on Completed Livestock Research Activities 2020

Tavakol M. and Dennick R. 2011. Making sense of Cronbach’s alpha. International Journal
of Medical Education, 2, 53–55. https://doi.org/10.5116/ijme.4dfb.8dfd
Tsegay Teame, Natarajan P. and Zelalem Tesfay. 2015. Assessment of fishery activities for
enhanced management and improved fish production in the Tekeze reservoir, Ethiopia.
International Journal of Fauna and Biological Studies, 3 (1), 105–113.
Tullis T. 2013. Measuring the user experience collecting, analyzing, and presenting usability
metrics (2nd edition).

390