Available online at www.sciencedirect.

com

ScienceDirect
Procedia Food Science 6 (2016) 339 – 343

International Conference of Sabaragamuwa University of Sri Lanka 2015 (ICSUSL 2015)

Impacts of agricultural practices on water quality in Uma Oya

Abstract

Sustainability of global food production is highly depending on the quality of the environment. In many parts of the world
increase of agricultural production heavily depend on intensive agricultural practices which are having negative impact on the
environment. The impacts of agricultural practices on surface water quality is given special attention currently since the safe and
ample supply of freshwater is fundamental to humans and for the sustainability of ecosystem function. Intensive agricultural
practices in river catchments often pose threat to the ecological integrity of river ecosystems. Uma Oya watershed in the upper
Mahaweli watershed in Sri Lanka is an intensively cultivated landscape. In most parts of the catchment previously forested lands
have been cleared and converted to agricultural lands. However, the empirical evidence on quantitative assessment of such land
use conversion impacts on stream ecological health is lacking in the context of river catchments in Sri Lanka. Therefore the
present study was aimed at evaluating the agricultural land use impacts on stream physical habitat quality, water quality and
macroinvertebrate indices in the Uma Oya catchment at different spatial scales. The relationship between catchment and site
scale % agricultural lands, water quality and macroinvertebrate indices were evaluated using univariate and multivariate
approaches. The results indicated that stream physical habitat quality, water quality parameters and macroinvertebrate indices are
significantly (p<0.05) affected by catchment scale % agricultural land cover. Among the water quality variables that were tested
NO2-N, NH3-N, PO4-P and BOD5 level in sites with higher percentage of agricultural land cover exceeded the drinking water
quality standards during dry season. PO4-P and BOD5 level in those sites exceeded the proposed ambient water quality standards
for inland waters in Sri Lanka for aquatic life and for irrigation purposes. Findings of the present study suggest that catchment
scale interventions are crucial for the management of Uma Oya watershed and for the improvement of water quality and
sustainable agricultural production.

© 2016
© 2016Published
The Authors. Published
by Elsevier by Elsevier
Ltd. This is an openLtd.
access article under the CC BY-NC-ND license
Peer-review under responsibility of scientific committee of International Conference of Sabaragamuwa University of Sri Lanka
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
2015 (ICSUSL
Peer-review under2015).
responsibility of International Conference of Sabaragamuwa University of Sri Lanka 2015 (ICSUSL 2015).

Keywords: Agricultural practices; water quality; physical habitat quality; macroinvertebrate indices

2211-601X © 2016 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of International Conference of Sabaragamuwa University of Sri Lanka 2015 (ICSUSL 2015).
doi:10.1016/j.profoo.2016.02.068
340 W.D.T.M Gunawardhana et al. / Procedia Food Science 6 (2016) 339 – 343

1. Introduction

Sustainability of global food production is highly depending on the quality of the environment. The need for
increased food production to cater to increasing human population is challenged by issues related to sustainability of
the world’s ecosystems on which food production depends on. In many parts of the world increase of agricultural
production heavily depend on intensive agricultural practices which are having negative impact on the environment.
The impacts of agricultural practices on surface water quality is given worldwide attention currently since the safe
and ample supply of freshwater is fundamental to humans and for the sustainability of ecosystem function.
The Upper Mahaweli watershed area in Sri Lanka has undergone considerable land cover changes during the last
few decades due to the anthropogenic influences. In most parts of the previously forested catchment, areas have
been continuously cleared and converted into agricultural lands. Intensive agricultural practices in steep slopes in the
catchment have used extremely high levels of pesticides (insecticides and fungicides) and fertilizers in order to
maintain high yields and profits1. Evaluation of such land conversion impacts on stream ecological integrity is
important for taking decisions in catchment management and for the adoption of sustainable agricultural practices.
The present study was conducted in Uma Oya (fig. 1) catchment in upper Mahaweli watershed in Sri Lanka with the
aim of testing the agricultural land use impacts under different spatial scales on river health in terms of physical,
chemical and biological quality.

Fig.1. Sampling locations in the Uma Oya catchment

2. Methodology

Ten sub catchments within Uma Oya watershed with 2nd or 3rd order tributaries draining into Uma Oya were
selected as sampling locations for the study (Fig. 1). The agricultural land percentages at catchment scale and site
scale were evaluated by quantifying % of agricultural lands in whole sub-catchments and within riparian corridors
extending 200m upstream of the sampling point (with 300 m lateral extension in each side) using Landsat TM
Satellite images.
Water samples were collected monthly from 10 sampling locations covering10 sub catchments from August
2014 to January 2015. Electrical conductivity (EC), pH, temperature, Biochemical Oxygen Demand (BOD5), Total
Dissolved Solids (TDS), Total Solids (TS), Total Suspended Solids (TSS) , Nitrite-N, Nitrate-N, Ammonia-N,
Phosphate-P, Sulphate, Fluorite Iron, Magnisium and Sodium were measured following APHA (2005) procedures.
Riparian Quality Index (RQI) and Channel Physical Habitat Quality Index (CQI) were developed for each sampling
location to evaluate the riparian and channel habitat quality in each sampling reach using the criteria of Rapid Bio
Assessment Protocols for use in streams and wadeable rivers 2.
Macroinvertebrate samples were collected from study stream reaches using a suber sampler in September (dry
period) and December (rainy period) 2014 respectively. The macroinvertebrate families were identified and biotic
indices such as family richness, % EPT taxa (Ephemeroptera: Plecoptera: Trichoptera), % Chironomids were
calculated since they are universally used metrics to track water-quality changes effectively 3.
Standard Multiple Regression analysis was conducted in order to explore the effect of agricultural land impacts
at catchment and site scale to channel physical habitat quality, water quality and macroinvertebrate indices.
Principal Component Analysis (PCA) was also conducted in order to identify patterns of variation of sites based on
stream physical habitat quality, water quality, and catchment land use at different spatial scales and in different
seasons
3. Results and discussion

Results of the multiple regression analysis are presented in the Table 1. It indicated that % agricultural land
cover was having a significant (p<0.05) negative impact on RQI, CQI and water quality. The increase of catchment
scale % agricultural lands were having significant (p<0.05) negative impact on water quality by increasing the
levels of temperature, TS, TDS, TSS, Conductivity, BOD5, NO3-N, NO2-N, NH3-N, Mg levels in stream.

Table.1 Regression coefficients (r2) for significant (p˂0.05) stream physical habitat quality index and water quality variables
(n=60) for percentage agricultural land use categories in the micro catchments
Spatial Physical Water quality Macro-
scale Environmen invertebrate
t Indices
Conductivity
Temperature

abundance
richness
NO3-N

NO2-N

NH3-N
BOD5

Total
Taxa
TDS
RQI

CQI

TSS

Mg
pH

Cd
TS

% Model (r2) 0.49 0.30 0.14 0.12 0.13 0.30 ns 0.25 ns 0.17 ns ns ns 0.17 0.50 0.41
Agricultur Catchment - - 0.41 - 0.51 0.26 0.39 - -
al lands scale 0.42 0.29 0.70 0.70
Site scale - - ns ns ns 0.02 Ns ns ns ns ns ns ns
0.49 0.23

It also indicated the catchment scale % agricultural lands are having significant (p<0.05) negative effects on
macroinvertebrate Family richness and Taxa abundance and no impact on % Chironomidae and % EPT at both
spatial scales.
342 W.D.T.M Gunawardhana et al. / Procedia Food Science 6 (2016) 339 – 343

4 % agri-Season
L dry
M dry
H dry
M wet
2 H wet
Sulphate mg/L L wet
Phosphate mg/L

pH
Temperature ( 0C) SQI
RQI
PC2 ( 18.1%)

TSS mg/L

0
Conductivity
TDS(μS/cm)
mg/L
Nitrite mg/L

BOD mg/L
Nitrate mg/L flow

-2

-4
-6 -4 -2 0 2 4 6
PC1(33.9%)

Fig. 2. PCA plot indicating site variation based on stream physical habitat and water quality (Thee letters in the legend indicate
level of % agricultural land cover and corresponding season; % agricultural lands: L=Low; M=Medium; H=High; Season: W=
Wet season; D= Dry season) (n=60)

PCA analysis of stream physical habitat, water quality and % agricultural land use is depicted in the figure 2.
The PC axis 1 represented 33.9 % of the data variability. Vector loading to principal component axis indicated that
SQI, RQI, TDS, temperature and conductivity are highly correlated with PC1 axis suggesting the main variables
driving the separation of sites along the PC1 axis. Sites with low % agricultural land at both spatial scales were
grouped together in the positive side of the PC1 and they were associated with improved RQI and CQI. Sites with
higher % agricultural lands at both spatial scales were grouped in the negative side of the axis and they were
associated with increased temperature, conductivity and TDS.
PC2 axis represented 18.2% of the data variability and parameters such as water flow, NO 3-N, PO4-P and
Sulphate were mainly responsible for the separation of sites along PC2. The separation of sites along the PC2 axis is
mainly attributed to the seasonal effects.
Physical habitat and water quality of streams are often related to processes that occur across the terrestrial
landscape. The magnitude of these land use impacts on stream health may depend upon the spatial scale that they are
measured4. Results of the present study indicated that % agricultural lands at catchment scale negatively affect the
water quality and stream habitat quality. Among the water quality variable that were tested NO2-N, NH3-N, PO4-P
and BOD5 level in sites with higher percentage of agricultural land cover exceeded the drinking water quality
standards during dry season rather than the wet season. Further, PO4-P and BOD5 level in those sites exceeded the
proposed ambient water quality standards for inland waters in Sri Lanka for aquatic life and for irrigation purposes.
During the dry period, agricultural practices were predominantly conducted using irrigated water and runoff from
the crop lands carry high amount of sediments from land preparation activities, fertilizer, residues of pesticides,
herbicides, and ions contributing to deteriorated water quality. The low flow of the streams during the dry period
may also contribute to the concentration of nutrients and ions in those sites during dry period. During the rainy
 W.D.T.M Gunawardhana et al. / Procedia Food Science 6 (2016) 339 – 343 343

period a substantial amount of organic loads, nutrients such as NO 2-N enters the streams from the agricultural lands
and lack of riparian cover along the streams in agricultural areas may have further enhanced the deterioration of
water quality and these may lead to arising various environmental and health issues that resulted from exceeding
ambient water quality standards.
Macroinvertebrate biotic indices reflect long-term impacts to stream health hence these indices are essential to
have holistic view of the ecological integrity of streams impacted by catchment disturbances5. CQI reflects the
habitat diversity within channels and improved habitat diversity within channel support divers communities and
increased the species richness2. The present study indicated that % agricultural lands deteriorate the riparian
vegetation and channel quality as reflected in the reduced RQI and CQI. Increased sedimentation of riverbed, water
flow regulation impacts may have contributed to reduced channel quality Index and habitat diversity for biota in
study stream reaches. It is possible that lowered macroinvertebrate family richness and abundance values reported
in agricultural sites may have attributed to the degraded habitat quality and water quality associated with sites with
higher agricultural practices.
The findings of the present study indicated that increase of % agricultural lands negatively affect the stream
ecological integrity and water quality of the study catchment. Mechanisms such as better soil management practices
in agriculture, forest cover restoration projects, riparian vegetation and channel quality restoration efforts are some
of the essential steps needed to upgrade the stream ecological health in the Uma Oya catchment. Such catchment
management efforts are essential for the sustainability of agricultural production and preservation of water quality in
the study catchment.

Acknowledgements

Authors acknowledge the National Research Council of Sri Lanka Grant No.13-160 for funding the research.

References

1. Bandara. DGVL , Thiruchelvam S. Factors affecting the choice of soil conservation practices adopted by potato farmers in Nuwara Eliya
district, Sri Lanka, Tropical Agricultural Research and Extension. 2008;11:49-54.
2. Barbour. MTJ, Gerritsen BD, Snyder ,Stribling. JB, Rapid Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton,
Benthic Macroinveretebrates and Fish, Second Edition. EPA 841-B-99-002. U.S. Environmental Protection Agency; Office of Water,
Washington, D.C; 1999.
3. Wallace. JBJW, Grubaugh.,Whiles MR. Biotic indices and stream ecosystem processes: results from an experimental study. Ecological
applications. 1996; p. 140-151.
4. Sponseller RE, Benfield, Valett. H. Relationships between land use, spatial scale and stream macroinvertebrate communities. Freshwater
Biology. 2001; 46(10):1409-1424.
5. Kroll. SA, Llacer CN, de la Cruz Cano.M. & de las HerasIbáñez. J. The influence of land use on water quality and macroinvertebrate biotic
indices in rivers within Castilla-La Mancha (Spain). Limnetica. 2009; 28(2):203-214.