SUSTAINABLE TILAPIA FEED DERIVED

FROM URBAN FOOD WASTE

A Thesis
Presented to
The Academic Faculty

by

Justin G. Chaddick

In Partial Fulfillment
of the Requirements for the Degree
Master of Science in the
School of Biology

Georgia Institute of Technology

December 2015

Copyright  ©  2015  Justin  G.  Chaddick

Sustainable Tilapia Feed Derived from Urban Food Waste

Approved by:

Dr. Jeannette Yen, Advisor

School of Biology
Georgia Institute of Technology

Dr. Marc Weissburg

School of Biology
Georgia Institute of Technology

Robert Wallace
Georgia Tech Research Institute

Date Approved: December 3, 2015

 
 Acknowledgements

I am grateful for all of the support I received in completing this work. Thank you to

the ArkFab team, specifically Corwin May, Brent Lyman and Larisa Pender-Healy for

providing indispensible technical assistance. I would like to thank my advisor, Jeannette

Yen, for her unwavering confidence and enthusiasm and Marc Weissburg and Robert

Wallace for their support. Thank you to Truly Living Well for helping us connect with the

community and make this work more meaningful. Thank you also to the PNC Foundation

for making this work financially possible.

iii  
For Liam.

iv  
 Table of Contents

Acknowledgements ....................................................................................................... iii

List of Tables .................................................................................................................. vi
List of Figures ............................................................................................................... vii
Summary....................................................................................................................... viii
1. Introduction .............................................................................................................. 9
1.1. Black Soldier Fly Larvae................................................................................. 12
1.2. Duckweed ........................................................................................................ 13
1.3. Feed Conversion Efficiency and Feed Assimilation Efficiency .................. 15
2. Methods and Materials .......................................................................................... 17
2.1. Feed Conversion Efficiency Study ................................................................ 17
2.1.1. Study Setup................................................................................................. 17
2.1.2. Fish ............................................................................................................. 18
2.1.3. Weighing Fish ............................................................................................. 18
2.1.4. Feeding Sessions........................................................................................ 18
2.1.5. Feed Components ...................................................................................... 18
2.1.6. Feed Preparation ........................................................................................ 19
2.1.7. Nutritional Analysis...................................................................................... 20
2.1.8. Statistical Analysis and FCR ....................................................................... 20
2.1.9. Average Daily Gain ..................................................................................... 20
2.2. Fish Feed Assimilation Assay ....................................................................... 21
2.2.1. Study Design............................................................................................... 21
2.3. BSFL Conversion Efficiency Study ............................................................... 22
3. Results .................................................................................................................... 24
3.1. Feed Conversion Study .................................................................................. 24
3.1.1. Comparison of Means ................................................................................. 25
3.1.2. Binned Comparison of Variance ................................................................. 26
3.1. Feed Assimilation Study ................................................................................ 28
3.2. BSFL Conversion Study ................................................................................. 29
4. Discussion.............................................................................................................. 30
5. Conclusion ............................................................................................................. 33
References..................................................................................................................... 34

v  
 List of Tables

Table 1 Sorted Food Waste ..................................................................................... 22

Table 2 FCR Summary ............................................................................................. 24

Table 3 Histogram Bin Counts................................................................................ 27

Table 4 Treatment Group Standard Deviation....................................................... 27

vi  
 List of Figures

Figure 1 System Overview. ..................................................................................... 11

Figure 2 Final Means ............................................................................................... 26

Figure 3 Population Histograms............................................................................. 28

vii  
 Summary

Finding an alternative to fishmeal and fish oil in animal feeds has been a topic of

increasing interest due to the pressures being put on the ocean’s fisheries and the

increasing world demand for animal protein. An often-overlooked source of nutrients is in

the form of food waste. One third of all food produced globally ends up in landfills,

wasting a huge amount of nutrients and embodied energy that could otherwise be

redirected towards productive use. This study investigated the feasibility of feeding

Hermetia illucens, the black soldier fly larvae (BSFL), grown on urban food waste, and

Lemna minor, a species of duckweed, to tilapia in a recirculating aquaponic system as a

compound feed. The study compared the growth of two groups of 58 tilapia over 44

days; one group was fed commercial pellets and the other a compound feed composed

of BSFL and duckweed. The group fed the commercial pellets achieved heavier weight

gain than the group fed the experimental feed but both groups resulted in steady weight

gain and had similar mortality rates. Feeding the experimental feed composed of BSFL

and duckweed to tilapia in an aquaponics system is an effective method of diverting food

waste from the landfill and further research should be done to optimize this process.

viii  
 1. Introduction
The world’s food system is facing a multitude of crises. Shortages of arable land

and fresh water, climate change, biodiversity loss, ecosystem destruction, and

environmental pollution loom as growing problems with potentially devastating effects if

not addressed (FAO 2011b; FAO 2014a; Rockström et al. 2009). These problems are

compounded by the increasing demand for animal protein created by the rapidly

developing economies of large countries such as China (FAO 2003) since globally 35%

of all food produced is used as animal feed (Foley et al. 2011). In addition to terrestrial

animal protein, the global demand for seafood is growing. However, the ocean’s

fisheries have reached their harvest limits as 30% of fish stocks are fished at a

biologically unsustainable level and another 60% are fished at their maximum yield (FAO

2014b). Aquaculture is expected to meet the increase in demand (Neori et al. 2007) but

has traditionally been reliant on resources derived from the ocean, namely fishmeal and

fish oil (Olsen and Hasan 2012; Tacon 2006; R. W. Hardy and Tacon 2002). Fishmeal

has been the ideal protein source in aquaculture because of its high protein content,

balanced amino acids, vitamins and minerals, essential fatty acid content, and

historically cheap price (T. N. Nguyen, Davis, and Saoud 2009; Gatlin et al. 2007). Due

to a limited supply and increasing demand for fishmeal (FAO 2014b; Olsen and Hasan

2012) there has been a large amount of research focused on finding alternative sources

of protein in aquaculture diets (A. F. M. El-Sayed and Tacon 1997; T. N. Nguyen, Davis,

and Saoud 2009; Olsen and Hasan 2012; Tacon 2006). Using plant proteins from

sources such as soybean meal to replace fishmeal has yielded some positive results (El-

Sayed and Tacon 1997; El-Sayed 1999; Naylor et al. 2009; Hardy 2010). However,

while the use of conventional plant proteins as aquaculture feed decouples aquacultural

9
production from capture fisheries, it still relies on resource intensive farming and

competes with other uses of commodity crops (Hardy 2010).

Utilization of waste products in aquaculture feed could help tip the balance from

aquaculture being an extractive industry towards a productive one (Wohlfarth and Hulata

1987; Wu et al. 1994; Ulloa et al. 2004). An often-overlooked source of nutrients is food

waste. Globally around one third of all food produced is wasted (FAO 2011a), usually

ending up in a landfill where it becomes a source of organic pollution. In North America,

at just the consumer level, around 100 kg of food is wasted per capita every year (FAO

2011a). The US EPA recommends food that is no longer suitable for human

consumption be fed to animals (US EPA 2015). In order to utilize food waste as animal

feed it must be handled and processed into a form that is economical and nutritionally

viable for the animals being fed by it. Black soldier fly larvae (BSFL) can be used to

digest food waste to create an animal feed that is high in protein and fat (Nguyen,

Tomberlin, and Vanlaerhoven 2015) and have been successfully fed to fish (Bondari and

Sheppard 1981; St-Hilaire et al. 2007; Kroeckel et al. 2012). Another organism that has

attracted attention as an alternative fish feed is duckweed (Gaigher, Porath, and Granoth

1984; Hassan and Edwards 1992; Fasakin, Balogun, and Fasuru 1999; Leng, Stamboli,

and Bell 1995). Duckweed is efficient at absorbing nutrients from water (Hasan and

Chakrabarti 2009; Leng, Stamboli, and Bell 1995) and grows well when there is

decaying organic material in the water (Skillicorn 1993). This quality might make it

possible to use the refuse from BSFL culture as the nutrient source for a duckweed

culture. By doing so, two types of alternative fish feed could be generated from food

waste diverted from the landfill. These feeds could be then fed to tilapia in an aquaponic

system. Aquaponics is a type of recirculating aquaculture that uses edible plants to

perform biological filtration of the aquacultural water (Rakocy, Masser, and Losordo

2006). Combining all of these elements together would make it possible to create a

10  
variety of marketable yields from diverted food waste (figure 1). More information about

BSFL and duckweed is detailed in the following sections.

Figure 1 System Overview- Overview of system for diverting food waste from landfills towards
productive yields. Food waste is fed to BSFL, which provide nutrients for duckweed culture. BSFL
and duckweed are fed to fish in an aquaponic system, which also produces edible plants. The
refuse from the BSFL could be used to feed vermiculture creating additional marketable yields of
earthworms and earthworm castings.

11  
 The specific aim of this project was to evaluate the viability of a compound feed

composed of black soldier fly larvae and duckweed by comparing its performance to

conventional feed containing fishmeal in a recirculating aquaponic system. Three studies

were conducted. The first study compared the feed conversion efficiencies of fish fed the

duckweed and BSFL feed and the commercial feed containing fishmeal. The second

study compared the feed assimilation efficiencies of the two feeds. The third study

investigated the feed conversion efficiency BSFL to ascertain an approximate idea of

how effective the BSFL cultures were at consuming waste and producing larvae. More

broadly, this research seeks to broaden the arena of knowledge regarding closed loops

systems of production in order to achieve maximum material and energetic yields from a

set of resources.

1.1. Black Soldier Fly Larvae

The larvae of Hermetia illucens, the Black Soldier Fly (BSFL), posses many qualities

that make them attractive for converting urban food waste into animal feed and other

valuable products. They have been successfully fed to a variety of livestock animals

including pigs (G. L. Newton et al. 1977), (Hale 1973), tilapia, and catfish (Bondari and

Sheppard 1981).

BSFL have been shown to effectively digest a wide range of organic waste products

such as offal, kitchen waste, and fruit and vegetable waste (Nguyen, Tomberlin, and

Vanlaerhoven 2015). BSFL are also useful for managing manures and when grown on

chicken manure were able to reduce the amount of manure by 50% and the larvae had a

resulting composition of 42% protein and 35% fat (Sheppard et al. 1994). The

conversion of low value manure, or other organic waste streams, to bulk proteins and

lipids creates many opportunities for developing valuable products. The ability of the

12  
BSFL to create such high proportions of lipids has even attracted attention to how their

culture might be applied to biofuel systems (Li et al. 2011).

Despite their fondness of manure, BSFL are thought of as a non-pest species

(Sheppard et al. 1994) and have been shown to reduce E. coli levels in chicken and

dairy manures (Erickson et al. 2004; Liu et al. 2008) as well as compete with house fly

larvae, thereby decreasing their population (Bradley and Sheppard 1984).

The ease of harvesting the BSFL is also of immense value. The last stage of the

BSFL’s life cycle before pupation is a migratory stage where the larvae develops a large

fat store, empties its gut, and then seeks a place away from the waste to pupate

(Sheppard et al. 1994; G. L. Newton et al. 1977). This makes collecting the larvae as

simple as providing a singular high point that falls into a collection bucket. Being able to

harvest the larvae passively makes scaling systems much simpler and much more cost

effective than more intensive systems.

After the food waste has been processed by the BSFL, the leftover biomass is

nutrient rich, low-odor, humus that can be added to soils as an amendment to increase

organic matter and fertility (Diener, Zurbrügg, and Tockner 2009).

One drawback of BSFL when compared to fishmeal is the presence of chitin that

forms the BSFL outer membrane. Chitin is largely indigestible by fish and it should also

be noted that the nitrogen content of chitin is 6.89%, which should be taken into account

when formulating feed as it significantly changes the amount of digestible crude protein

(Diener, Zurbrügg, and Tockner 2009).

1.2. Duckweed

Duckweed is an extremely fast growing aquatic macrophyte that floats on the surface

of the water. It is distributed around the world in tropical, subtropical, and temperate

climates and grows in still, fresh, or brackish water (Skillicorn 1993; Rusoff, Blakeney,

13  
and Culley 1980). There are 37 identified species of duckweed that span five genera of

the Lemnaceae family: Wolffia, Wolfiella, Spirodela, Landolita, and Lemna (Appenroth,

Borisjuk, and Lam 2013). Of those species, Lemna minor was the only one cultured in

this study. Under ideal conditions, such as growing in the presence of decaying organic

matter (Skillicorn 1993), duckweed is capable of forming dense mats of colonies that

grow over each other (Hillman 1961; Rusoff, Blakeney, and Culley 1980) and can double

its mass in less than 48 hours (Skillicorn 1993). Duckweeds grows fastest in warm

waters with a pH between 6.5-7.5 and full sun exposure but are tolerant of water

temperatures between 6 and 33 oC and a large range of light intensities (Hillman 1961;

Leng, Stamboli, and Bell 1995). Duckweeds are highly effective at concentrating

nutrients such as phosphorus and nitrogen in their tissues and are capable of lowering

the concentrations of nutrients in eutrophic waters down to trace levels (Hasan and

Chakrabarti 2009; Leng, Stamboli, and Bell 1995).

Besides their profound ability of nutrient removal from water, duckweed also has

an impressive nutritional profile. Efficiently managed duckweed cultures typically are

composed of approximately 40% crude protein, 10% fiber, and 5% polyunsaturated fat

(Hillman and Culley 1978; Leng, Stamboli, and Bell 1995). The high protein content of

dried duckweed make it a good source of amino acids and protein concentrate (Rusoff,

Blakeney, and Culley 1980). Additionally, the protein-rich leaves of duckweed have

around one-tenth the amount of fiber as other common feed plants such as maize and

soy (Leng, Stamboli, and Bell 1995; Skillicorn 1993), which further increases its potential

as a highly digestible source of feed protein.

Due to its favorable nutrition, duckweed has been successfully used as a protein

source for a variety of animals including fish, poultry, and even humans (Hillman and

Culley 1978; Skillicorn 1993; Leng, Stamboli, and Bell 1995; Hassan and Edwards

1992). Lemna minor, the species cultured for this study, has been shown to be a

14  
complete protein that contains all of the essential amino acids (Rusoff, Blakeney, and

Culley 1980; Muztar, Slinger, and Burton 1978). Fresh duckweed of the Lemna genus

performed well as feed when fed to Nile tilapia that were able to digest duckweed and

efficiently assimilate its high protein content, with the ratio of duckweed offered to fish

biomass gained equaling 3.4 (Hassan and Edwards 1992).

The fast growth rate of duckweed allows for large yields. It has been

demonstrated that duckweeds can achieve yields of 10-20 tons dry matter/ha/year in

non-laboratory settings (Leng, Stamboli, and Bell 1995). Duckweed, like many aquatic

plants, is 91-95% water (Hasan and Chakrabarti 2009) so drying it out is important if the

protein is to be concentrated or if there is a large transportation distance between where

it grows and where it is consumed. For this reason, decentralized culture and use of

duckweeds generally enhances their economic viability (Hillman and Culley 1978).

1.3. Feed Conversion Efficiency and Feed Assimilation

Efficiency

Feed conversion efficiency (FCE), also referred to as feed conversion ratio

(FCR), is the amount of feed given to animal divided by the amount of animal biomass

harvested as expressed below:

Feed _ Input
FCE =
Animal _ Biomass _ Output

Different animals have very different FCEs and the exact conversion factor is

dependent on a€range of variable such as species, feed type, and age. According

to a recent study, the FCE was around 14 for beef cattle and 2 for chickens

(Peters et al. 2014). Most of the fish grown in aquaculture have a FCE between

1.5 and 2 (Olsen and Hasan 2012).

15  
 Feed assimilation efficiency refers to the amount of feed eaten that is used

for metabolism or turned into additional biomass. In fish, the assimilation

efficiency is typically measured by tracking the amount of feed eaten and the

amount of feces produced (Belal 2005). Individual nutrients can also be tracked

by knowing the amount eaten versus the amount excreted (Amirkolaie et al.

2005).

16  
 2. Methods and Materials
2.1. Feed Conversion Efficiency Study

2.1.1. Study Setup

The first study was designed to compare the feed conversion efficiency of the

commercial feed to the experimental feed composed of BSFL and duckweed in order to

asses its viability as a substitute for commercial feed made of grain and fishmeal. The

fish used in the study were split into two treatment groups with one fed commercially

purchased feed and the other fed the BSFL and duckweed feed. Both groups were fed

known amounts and their weight was recorded at the beginning and end of the study.

The study was conducted in a 900ft2 greenhouse in downtown Atlanta, Georgia.

The greenhouse houses an aquaponic system that recirculates water in a closed loop.

The system has one main fish tank that is a 3000-gallon raceway and is plumbed in

parallel to 5 biofilter grow beds measuring 4’x24’x8”. The system is continually

circulated with a 3000gph pond pump and aerated with an air pump at 80L/min.

Two identical black 70-gallon stock tanks filled to approximately 50 gallons were

used to house the two treatment populations. The tanks were plumbed into the existing

aquaponic system so that the water leaving one bed of the plant biofilter first entered the

experimental tanks and then drained into the main raceway before being recirculated

back into the plant biofilters. This configuration was chosen to minimize water quality

differences between the two tanks and ensure adequate nutrient removal without water

exchange. Both tanks were covered with opaque corrugated PVC roofing with

approximately 2” of space left on each side to allow some light to enter the tanks but

prevent algae growth and keep fish from jumping out.

17  
 2.1.2. Fish

Oreochromis aureus (blue tilapia) were divided into two treatment populations

with each initially containing 58 individuals. Juvenile fingerlings, fish that are about a

finger long, were purchased from White Brook Tilapia FarmTM and grown out for 4

months in the main raceway on commercial feed before being used in the study.

2.1.3. Weighing Fish

Prior to beginning the study, the fish were caught in an improvised net cage and

then individually weighed on a scale out of the water before being added to the

experimental tanks. The study ran for 44 days and then fish were individually weighed

again.

2.1.4. Feeding Sessions

The two populations were fed known amounts of feed until apparent satiation

twice daily in 5 minute feeding sessions. A small portion of feed was added to the tank

and replaced at a rate matching the feeding rate until either the allotted 5-minute session

was over or the feeder observed highly reduced feeding activity for 30 seconds.

Substantially decreased feeding usually coincided with the end of the 5-minute sessions.

The fish consumed all of the food given as confirmed by visual observation of the feeder

and the amount of feed eaten by the population was recorded at the end of each

session.

2.1.5. Feed Components

The Lemna minor (duckweed) was obtained from the Georgia Tech Research

Institute from a project cultivating duckweed for chickenfeed. The duckweed was grown

18  
in shallow, fertilized ponds and was periodically harvested and dried in a solar

dehydrator. The dried duckweed was then stored in airtight plastic bags and frozen until

used in the feed preparation.

The Hermetia illucens (BSFL) was grown on site in specially designed

ProtapodTM bins purchased from an online retailer that are specifically designed for

BSFL culture (ProtaPodTM 2015). The bins are designed to provide a singular high point

for the wandering larvae to crawl to for collection. Each bin was fitted with a vinyl shoot

leading into a sealed 5-gallon bucket to collect the mature BSFL. The bins were covered

with opaque corrugated PVC roofing to prevent rain from entering the bins while allowing

adult flies to enter and exit the bins, facilitating oviposition. A hole was drilled in the

bottom of each bin to allow leachate to drain into 5-gallon buckets. The BSFL bins were

fed fruit and vegetable waste acquired from a local smoothie shop. The BSFL were

harvested and weighed immediately after the food waste additions and then frozen until

use in the feed preparation. New food waste was weighed and added on a weekly basis.

The amount of leachate and BSFL produced were recorded each time new food was

added allowing for an approximation of the BSFL food waste to BSFL biomass

conversion efficiency to be made.

2.1.6. Feed Preparation

The experimental population was fed a prepared feed of 40% dried duckweed

and 60% BSFL by estimated dry mass to achieve estimated protein content of 36%

matching the commercial pellets. To prepare the experimental feed, 40g of dried

duckweed was combined with 225g of thawed BSFL and 503g of filtered tap water. The

mixture was then blended in a high-speed blender until a homogeneous paste with

uniform consistency and no visible particulates was achieved. The paste was then

spread onto parchment paper on four 11”x17” baking sheets and dehydrated in the oven

19  
at 75oC for approximately 2 hours to create dehydrated sheets. The sheets were

removed from the oven, allowed to cool, and cut into approximately ¼” square pieces.

The pieces were stored in an airtight container and refrigerated until use.

The control group was fed Premium Fish FoodTM Tilapia Ultimate Growout

Pellets, a commercial feed comprised of fish meal, dehulled soybean meal, ground corn,

wheat middlings, brewers dried yeast, and vitamin mix, with a crude protein content of

36% (min), 6% crude fat (min), and 5% crude fiber (max). The pellets were also kept in

an airtight container and refrigerated until use.

2.1.7. Nutritional Analysis

The control and experimental feeds were sent to the University of Missouri

Agricultural Experiment Station Chemical Laboratories (ESCL) for analysis. Complete

amino acid profiles and crude protein, crude fat, crude fiber, ash, moisture, acid

detergent fiber (ADF), neutral detergent fiber (NDF), and omega-3 content, were

analyzed. Carbohydrates by difference and gross calories were calculated from the

analyses.

2.1.8. Statistical Analysis and FCR

The feed conversion ratio (FCR) of the tilapia was determined by dividing the amount

of feed eaten by the weight gain of the fish over the study period. The means of the final

weights of each treatment group were compared using a student’s t-test.

2.1.9. Average Daily Gain

The average daily weight gain of the fish was calculated by subtracting the

initial average weight from the final average weight and dividing by the number of

days the experiment ran as expressed below:

20  
 Mean _ Final _Weight − Mean _ Initial _Weight
Experiment _ Length _(Days)

2.2. Fish Feed Assimilation Assay

€
2.2.1. Study Design

A feed assimilation study was conducted to assess how well the fish were able to

convert the given feed into biomass. The assimilation efficiency of the fish feed into

biomass was evaluated by feeding the fish a known amount of feed, collecting and

weighing their feces, and determining the percentage absorbed or metabolized. Ten fish

from each of the feed treatments used in the feed conversion efficiency study were

individually placed into 5-gallon buckets filled with tank water filtered through a 5µm filter

and constantly aerated. Seven fish from each treatment group received either the

commercial feed or the BSFL and duckweed feed. To account for scales, mucous, or

other unknown particulate shed by the fish, the other two individuals from each treatment

group were controls and not fed. Following Heng et. al in 2007, all of the fish were

starved for 26 hours prior to beginning the study (Heng, Ong, and Hassan 2007). The

fish receiving feed in each treatment group were fed to apparent satiation and the

amount of feed was recorded for each individual. The fish were monitored after feeding

for gut evacuation of feces. Upon detection of feces, the fish were removed from their

bucket and the feces was collected by filtering the water through a 11µm filter using

vacuum flask. It was assumed that the fish completely evacuated their gut upon an initial

observation of feces. The water from the control groups was filtered after all of the fed

fish had evacuated their guts. The filter pads were placed in petri dishes and then stored

in sealed plastic bags in a cooler of ice water to prevent bacterial activity during the

collection period. Following the collection period, the filter pads were dried in an oven at

21  
68oC until constant weight and weighed. The feed assimilation was calculated as the

weight of the feces divided by the weight of the feed multiplied by 100.

2.3. BSFL Conversion Efficiency Study

A conversion efficiency study was done to determine how much BSFL mass could be

produced from fruit and vegetable wastes from a nearby smoothie shop. The BSFL were

cultured in four bins and fed all of the fruit and vegetable scraps produced from the

smoothie shop every week. Additions of fruit and vegetable waste were made once per

week. In order to approximate the composition of the fruit and vegetable scraps, two

samples of 32-gallon trash bins were sorted and weighed on three separate occasions.

The composition of the fruit and vegetable waste was broken into nine major categories

and the relative contribution of each type was computed (Table 1).

Type of Food Waste % of Total

Avocado 8.5
Bannana 20.3
Citrus 7.4
Coffee 1.2
Mango & Papaya 17.4
Melons 17.7
Pineapple 12.1
Misc 10.3
Juice 5.1

Table 1 Sorted Food Waste- Sorted food waste

components listed as percent of total.

To evaluate the productivity of the BSFL culture and approximate how much BSFL

could be generated per culture, the conversion of food waste to BSFL was quantified.

The BSFL bins were treated as “black boxes” with inputs of food waste and outputs of

22  
BSFL. The amount of residual matter left over from the BSFL was not quantified, so

rather than a true feed conversion ratio, the conversion efficiency was calculated to

provide a practical metric of productivity of the BSFL culture. The following parameters

were measured to determine the conversion efficiency of the BSFL cultures: the weight

of food waste added to the BSFL bins, the weight of BSFL harvested from each bin, and

the weight of liquid leached from the bins. The conversion efficiency was computed by

dividing the weight of BSFL produced by food waste added minus the weight of leached

liquids, then multiplied by 100. The equation used is shown below:

BSFL(lbs)
× 100%
(Feed _ Added(lbs) − Leachate(lbs)

23  
 3. Results

3.1. Feed Conversion Study

The commercial feed treatment experienced higher weight gain than the fish fed the

BSFL and duckweed diet (Table 2). The two groups had similar mortality (5 in the

commercial treatment and 4 in the BSFL and duckweed treatment). The FCR of the

commercial feed group was 1.7 while the BSFL and duckweed group had a much less

efficient FCR of 4.4. The average daily gain of the commercial treatment was about

twice as much as the BSFL and duckweed treatment.

FCR Results
Commercial BSFL/Duckweed
Metric Feed Feed
Total Initial Population Weight (grams) 2849.2 2818.4
Total Final Population Weight (grams) 4412 3527.2
Average Initial Population Weight
(grams) 48.3 47.8
Average Final Population Weight
(grams) 81.7 64.1
Amount Fed (grams) 2678.6 3092.2
Initial Number of Individuals 59 59
Final Number of Individuals 54 55
Mortality 5 4
FCR 1.7 4.4
Average Daily Gain (g/day) 0.76 0.37

Table 2 FCR Summary- Summary of the data acquired from the feed conversion experiment.

24  
 3.1.1. Comparison of Means

The mean weight of the fish that were fed commercial pellets (Ncom= 54,

M=81.7g, SD=40.5) was significantly different than the fish fed the BSFL and duckweed

feed (NBSFL=55, M=64.1, SD=29.4), t(107)=2.596, tcritical=1.982, p=0.01076, using a

student’s two sample t-test (Figure 1). However, the mean final weight of the fish fed

commercial pellets was substantially influenced by tail of the distribution (see following

section). Removing the six individuals in the highest bins resulted in means that were not

significantly different (Ncom= 48, M=71.7g, SD=30.0, NBSFL=55, M=64.1, SD=29.4),

t(101)=1.287, tcritical=1.984, p=0.201 using a student’s two sample t-test. The fish fed the

commercial pellets had a higher final weight than the fish fed the BSFL and duckweed

feed, but this was mostly due to six large individuals in treatment receiving commercial

feed.

25  
 .

Figure 2 Final Means- The final mean weights of the two treatment
groups is shown with error bars representing the standard error of
the mean.

3.1.2. Binned Comparison of Variance

The distribution of initial and final weights of each treatment group were compared

using histograms to bin the weights in 20-gram intervals (Figure 2). The initial groups

started out with similar variances (Table 4). The variance of the final weights of both

treatments was substantially higher than the initial weights. The variance of the

commercial feed increased much more than the variance of the BSFL and duckweed

treatment group. The distributions of the final weights for each group were significantly

different χ2(9, N=109)= 18.395 p=0.0309, however the tails of the distribution contributed

most of the observed difference (Table 3). The histograms show that the commercial

26  
feed led to larger maximum weights of individual fish and heavier fish overall than the

BSFL and duckweed feed.

Commercial Feed BSFL Duckweed Feed

Bin count X2 count X2
0-20 0 0.4954 1 0.4864
20-40 6 0.4683 10 0.4598
40-60 13 0.6859 20 0.6734
60-80 11 0.4864 7 0.4775
80-100 5 0.7954 10 0.7809
100-120 11 1.7139 4 1.6827
120-140 2 0.0919 3 0.0902
140-160 2 1.0279 0 1.0092
160-180 3 1.5418 0 1.5138
180-200 1 0.5139 0 0.5046

Table 3 Histogram Bin Counts-Histogram bin counts with corresponding chi-squared

values.

Measure Commercial BSFL/Duckweed

Initial Standard Deviation 20.3 19.4
Final Standard Deviation 40.5 29.4
Initial Variance 407 370.5
Final Variance 1609.6 848.9

Table 4 Treatment Group Standard Deviation- Standard deviation and variance of the
two treatment groups at the beginning and end of the study.

27  
 Figure 3 Population Histograms- Histograms of the population weights at the beginning and
end of the feed conversion efficiency study.

3.1. Feed Assimilation Study

Several assumptions made in the design of the feed assimilation study were

proven to be false and a result, no data was obtained. The first assumption was that

tilapia would have completely emptied their guts after 26 hours of fasting. This was

proven to be false when all 14 fish defecated into the buckets at least 33 hours after their

last feeding. This also proved the assumption that the fish defecated once per feeding to

be false, which was further emphasized when the fish labeled as control defecated again

at 52 hours after the last feeding. An additional complication was that only three fish in

28  
each treatment group ate the feed given, probably due to handling stress. Future

attempts at this study should have a continuous fecal collection method and an

adequate way of assuring the fish have emptied their guts at the beginning of the study.

3.2. BSFL Conversion Study

The four BSFL bins were fed a total of 2339 pounds of fruit and vegetable scraps

and together produced 57 pounds of BSFL. The average conversion ratio of fruit and

vegetable waste to BSFL was 2.44% with a standard deviation of 0.33% between bins.

29  
 4. Discussion

The commercial feed outperformed the BSFL and duckweed feed in growth rate and

weight gain but that was mostly due to a minority of the commercial population growing

much larger than the rest. Gaigher et al. 1984 found that tilapia fed only fresh duckweed

at a rate matching their consumption rate over the course of the day experienced an

average daily gain of 1g/day (Gaigher, Porath, and Granoth 1984). Since both the

commercial feed and BSFL and duckweed feed treatment groups grew well under

1g/day, (0.76g/day and 0.37g/day respectively), that would suggest the feeding rate in

the feed conversion study was below the optimum rate necessary for maximum weight

gain. Bondari and Sheppard 1981 found that blue tilapia fed chopped BSFL at 3% of

their body weight gained an average of 0.56g/day over a 10-week period which was not

significantly different from the tilapia receiving commercial pellets in that study (Bondari

and Sheppard 1981). This result is closer to the weight gains observed in this study but

still higher than the BSFL and duckweed feed. The difference could be due to the

nutritional differences in the feed or differences in environmental conditions the fish were

reared in.

The BSFL and duckweed feed probably had a higher moisture content than the

commercial pellets, which could have led to the fish eating less total macronutrients

despite eating a larger total mass. The chitin content of the BSFL could also lower the

amount of digestible protein in the BSFL and duckweed feed which was not adjusted for

in the feed formulation. To track environmental conditions, water quality was monitored

twice daily at each feeding session by checking dissolved oxygen, unionized and ionized

ammonia, temperature, and pH using a YSItm pro plus multimeter. Another possible

30  
reason for the depressed weight gain was high concentrations of ionized ammonia

(NH4+) that quickly rose at the beginning of the study for unknown reasons and

averaged 7.1 mg/L over the course of the feed study in both treatment tanks. Unionized

ammonia was undetectable or low for the entire study but prolonged exposure ionized

ammonia could have negatively affected fish health.

The fish fed commercial pellets had more variable weights than the fish fed the

BSFL and duckweed feed (σ2 commercial= 1609.6, σ2 BSFL/Duckweed = 848.9).

However if the six largest individuals were excluded from the commercial population, the

variances would be similar (σ2 commercial= 897.5, σ2 BSFL/Duckweed = 848.9). This

difference could be due to differences in the physical characteristics of the feed. The

commercial pellets were much harder and fish that started off small may not have been

able to fit the pellets in their mouths. This could have led to the large fish getting larger

and excluding the smaller fish from feeding. In contrast, the BSFL and duckweed feed

was relatively soft and easy for fish to break pieces off of when they were unable to

swallow them whole. Nutritional differences such as the increased fat content of the

BSFL and duckweed compared to commercial pellets could also have had an effect on

the weight distributions of the fish.

Despite not performing as well as the commercial pellets in overall weight gain,

the BSFL and duckweed feed did produce steady growth in tilapia and resulted in a

similar mortality rate as the commercial pellets. Feeding the BSFL and duckweed feed

also resulted in more uniform growth than the commercial pellets, which may be

desirable for aquaculture. Having all of the fish in a cohort grow at an even rate could

make harvesting and planning easier and also reduce the need for grading sizes. Since

the BSFL and duckweed feed did not have vitamin or mineral additives and was not as

heavily researched as the commercial pellets, the performance was a satisfactory

starting point for developing sustainable feed from food waste. The feed was produced

31  
with close to no costs aside from the capital cost of the bins, collecting the food waste,

and oven drying the feed. Since the tilapia were in a recirculating aquaponic system,

with high value lettuces being produced as well as tilapia, the profit margins were further

increased.

32  
 5. Conclusion

More research should be done on streamlining the production of fish food from

the urban waste stream using BSFL and duckweed. The duckweed for this study was

grown using inorganic fertilizer but BSFL leachate may contain enough nutrients to be a

suitable alternative, which would help to completely recover wasted nutrients. The food

waste processed by the BSFL could be used to feed vermiculture which would produce

earthworms and valuable castings (Newton et al. 2005). Additional research on

streamlining the production of the feed and using solar dehydration rather than a

conventional oven to dry it would also help reduce costs and make the process more

scalable. It would also be interesting to see more in depth life cycle and economic

analyses of using food waste for decentralized fish feed production using BSFL and

duckweed. This information would be relevant not only to those seeking to reduce costs

in the aquaculture industry, but also to the subsistence farmers of the world who, more

than anyone, know that wasting massive amounts of nutrients and energy is an

unacceptable practice. As a society we must rethink what we label “wastes” and instead

see them as resources that are to be cycled and conserved.

33  
 References

Amirkolaie, Abdolsamad K., Saber A. El-Shafai, Ep H. Eding, Johan W.

Schrama, and Johan A. J. Verreth. 2005. “Comparison of Faecal
Collection Method with High- and Low-Quality Diets Regarding
Digestibility and Faeces Characteristics Measurements in Nile Tilapia.”
Aquaculture Research 36 (6): 578–85. doi:10.1111/j.1365-
2109.2005.01257.x.

Appenroth, Klaus-J., Nikolai Borisjuk, and Eric Lam. 2013. “Telling Duckweed
Apart: Genotyping Technologies for the Lemnaceae.” Chinese Journal of
Appplied Environmental Biology 19 (1): 1–10.
doi:10.3724/SP.J.1145.2013.00001.

Belal, Ibrahim E.H. 2005. “A Review of Some Fish Nutrition Methodologies.”

Bioresource Technology 96 (January): 395–402.
doi:10.1016/j.biortech.2003.11.030.

Bondari, K., and D. C. Sheppard. 1981. “Soldier Fly Larvae as Feed in

Commercial Fish Production.” Aquaculture 24: 103–9. doi:10.1016/0044-
8486(81)90047-8.

Bradley, Susan W., and D. C. Sheppard. 1984. “House Fly Oviposition Inhibition
by Larvae ofHermetia Illucens, the Black Soldier Fly.” Journal of Chemical
Ecology 10 (6): 853–59. doi:10.1007/BF00987968.

Diener, Stefan, Christian Zurbrügg, and Klement Tockner. 2009. “Conversion of

Organic Material by Black Soldier Fly Larvae: Establishing Optimal
Feeding Rates.” Waste Management & Research 27 (6): 603–10.
doi:10.1177/0734242X09103838.

El-Sayed, Abdel-Fattah M. 1999. “Alternative Dietary Protein Sources for Farmed

Tilapia, Oreochromis Spp.” Aquaculture 179 (1–4): 149–68.
doi:10.1016/S0044-8486(99)00159-3.

El-Sayed, A. F. M., and A. G. J. Tacon. 1997. “Fishmeal Replacers for Tilapia: A

Review.” In Cahiers Options Mediterraneennes (CIHEAM). CIHEAM-
IAMZ. http://agris.fao.org/agris-
search/search.do?recordID=QC1997065922.

Erickson, M. C., M. Islam, C. Sheppard, J. Liao, and M. P. Doyle. 2004.

“Reduction of Escherichia Coli O157  : H7 and Salmonella Enterica

34  
 Serovar Enteritidis in Chicken Manure by Larvae of the Black Soldier Fly.”
Journal of Food Protection 67 (4): 685–90.

FAO. 2003. World Agriculture: Towards 2015/2030 a FAO Perspective. London:

Earthscan Publications Ltd.
http://www.fao.org/docrep/005/y4252e/y4252e00.htm#TopOfPage.

———. 2011a. Global Food Losses and Food Waste. Rome.

http://www.fao.org/docrep/014/mb060e/mb060e00.htm.

———. 2011b. “The State of the World’s Land and Water Resources for Food
and Agriculture  : Managing Systems at Risk.” In . Vol. 1st ed. Rome: FAO.

———. 2014a. The State of Food and Agriculture 2014. Rome: Food and
Agriculture Organization of the United Nations,.
http://www.fao.org/publications/sofa/2014/en/.

———. 2014b. The State of World Fisheries and Aquaculture 2014.

http://www.fao.org/3/a-i3720e/index.html.

Fasakin, E.a., A.m. Balogun, and B.e. Fasuru. 1999. “Use of Duckweed,
Spirodela Polyrrhiza L. Schleiden, as a Protein Feedstuff in Practical Diets
for Tilapia, Oreochromis Niloticus L.” Aquaculture Research 30 (5): 313–
18.

Foley, Jonathan A., Navin Ramankutty, Kate A. Brauman, Emily S. Cassidy,

James S. Gerber, Matt Johnston, Nathaniel D. Mueller, et al. 2011.
“Solutions for a Cultivated Planet.” Nature 478 (7369): 337–42.
doi:10.1038/nature10452.

Folke, C, and N Kautsky. 1992. “Aquaculture with Its Environment - Prospects for
Sustainability.” OCEAN & COASTAL MANAGEMENT 17 (1): 5–24.

Gaigher, I.G., D. Porath, and G. Granoth. 1984. “Evaluation of Duckweed

(Lemna Gibba) as Food for Tilapia (Oreochromis Niloticus X Oreochromis
Aureus) in a Recirculating Unit.” Aquaculture 41 (3): 235–44.
doi:10.1016/0044-8486(84)90286-2.

Gatlin, Dm, Ft Barrows, P Brown, K Dabrowski, Tg Gaylord, Rw Hardy, E

Herman, et al. 2007. “Expanding the Utilization of Sustainable Plant
Products in Aquafeeds: A Review.” AQUACULTURE RESEARCH 38 (6):
551–79.

Hale, O.M. 1973. “Dried Hermetia Illucens Larvae (Diptera: Stratiomyidae) as a

Feed Additive for Poultry.” Journal of the Georgia Entomological Society,
no. 1 (January): 16–20.

35  
Hardy, Ronald W. 2010. “Utilization of Plant Proteins in Fish Diets: Effects of
Global Demand and Supplies of Fishmeal.” Aquaculture Research 41 (5):
770–76. doi:10.1111/j.1365-2109.2009.02349.x.

Hardy, R. W., and A. G. J. Tacon. 2002. “Fish Meal: Historical Uses, Production
Trends and Future Outlook for Sustainable Supplies.” In , edited by R. R.
Stickney and J. P. McVey, 311–26. Responsible Marine Aquaculture.
Wallingford Oxon, New York, CABI.

Hasan, M. R., and R. Chakrabarti. 2009. “Use of Algae and Aquatic Macrophytes
as Feed in Small-Scale Aquaculture A Review.” FAO Fisheries and
Aquaculture Technical Paper, no. 531: ALL.

Hassan, Mohammad S., and Peter Edwards. 1992. “Evaluation of Duckweed

(Lemna Perpusilla and Spirodela Polyrrhiza) as Feed for Nile Tilapia
(Oreochromis Niloticus).” Aquaculture 104 (3): 315–26. doi:10.1016/0044-
8486(92)90213-5.

Heng, Hock Gan, T W Ong, and M D Hassan. 2007. “Radiographic Assessment

of Gastric Emptying and Gastrointestinal Transit Time in Hybrid Tilapia
(Oreochromis Niloticus X O. Mossambicus).” Veterinary Radiology &
Ultrasound: The Official Journal Of The American College Of Veterinary
Radiology And The International Veterinary Radiology Association 48 (2):
132–34.

Hillman, William S. 1961. “The Lemnaceae, or Duckweeds: A Review of the

Descriptive and Experimental Literature.” Botanical Review 27 (2): 221–
87.

Hillman, William S., and Dudley D. Culley. 1978. “The Uses of Duckweed: The
Rapid Growth, Nutritional Value, and High Biomass Productivity of These
Floating Plants Suggest Their Use in Water Treatment, as Feed Crops,
and in Energy-Efficient Farming.” American Scientist 66 (4): 442–51.

Kroeckel, S., A. -G. E. Harjes, I. Roth, H. Katz, S. Wuertz, A. Susenbeth, and C.

Schulz. 2012. “When a Turbot Catches a Fly: Evaluation of a Pre-Pupae
Meal of the Black Soldier Fly (Hermetia Illucens) as Fish Meal Substitute
— Growth Performance and Chitin Degradation in Juvenile Turbot (Psetta
Maxima).” Aquaculture 364–365 (October): 345–52.
doi:10.1016/j.aquaculture.2012.08.041.

Leng, R.A., J.H. Stamboli, and R. Bell. 1995. “Duckweed - a Potential High
Protein Feed Resource for Domestic Animals and Fish.” Livestock
Research for Rural Development 7 (1).

36  
Li, Qing, Longyu Zheng, Hao Cai, E. Garza, Ziniu Yu, and Shengde Zhou. 2011.
“From Organic Waste to Biodiesel: Black Soldier Fly, Hermetiaillucens,
Makes It Feasible.” Fuel 90 (April): 1545–48.
doi:10.1016/j.fuel.2010.11.016.

Liu, Qiaolin, Jeffery K. Tomberlin, Jeff A. Brady, Michelle R. Sanford, and Ziniu
Yu. 2008. “Black Soldier Fly (Diptera: Stratiomyidae) Larvae Reduce
Escherichia Coli in Dairy Manure.” Environmental Entomology 37 (6):
1525–30. doi:10.1603/0046-225X-37.6.1525.

Muztar, A. Jabbar, S.J. Slinger, and J.H. Burton. 1978. “Chemical Composition of
Aquatic Macrophytes. II. Amino Acid Composition of the Protein and Non-
Protein Fractions.” Canadian Journal of Plant Science - CAN J PLANT
SCI 58 (3): 843–49. doi:10.4141/cjps78-123.

Naylor, Rosamond L., Ronald W. Hardy, Dominique P. Bureaus, Alice Chiu,

Matthew Elliott, Anthony P. Farrell, Ian Forster, et al. 2009. “Feeding
Aquaculture in an Era of Finite Resources.” Proceedings of the National
Academy of Sciences of the United States of America.

Neori, Amir, Max Troell, Thierry Chopin, Charles Yarish, and et al. 2007. “The
Need for a Balanced Ecosystem Approach to Blue Revolution
Aquaculture.” Environment 49 (3): 37–43,2.

Newton, G. Larry, D. C. Sheppard, D.W. Watson, G. Burtle, and R. Dove. 2005.

“Using the Black Soldier Fly, Hermetia Illucens, as a Value- Added Tool
for the Management of Swine Manure.” Animal and Poultry Waste
Management Center, North Carolina State University, Raleigh, NC, USA.

Newton, G. L., O. M. Hale, C. V. Booram, and R. W. Barker. 1977. “Dried

Hermetia Illucens Larvae Meal as a Supplement for Swine.” Journal of
Animal Science 44 (3): 395.

Nguyen, Tri N., D. Allen Davis, and I. Patrick Saoud. 2009. “Evaluation of
Alternative Protein Sources to Replace Fish Meal in Practical Diets for
Juvenile Tilapia, Oreochromis Spp.” Journal of the World Aquaculture
Society 40 (1): 113–21. doi:10.1111/j.1749-7345.2008.00230.x.

Nguyen, Trinh T. X., Jeffery K. Tomberlin, and Sherah Vanlaerhoven. 2015.

“Ability of Black Soldier Fly (Diptera: Stratiomyidae) Larvae to Recycle
Food Waste.” Environmental Entomology 44 (2): 406–10.
doi:10.1093/ee/nvv002.

Olsen, Rl, and Mr Hasan. 2012. “A Limited Supply of Fishmeal: Impact on Future
Increases in Global Aquaculture Production.” Trends in Food and Science
Technology 27 (2): 120–28.

37  
Peters, Christian J., Jamie A. Picardy, Amelia Darrouzet-Nardi, and Timothy S.
Griffin. 2014. “Feed Conversions, Ration Compositions, and Land Use
Efficiencies of Major Livestock Products in U.S. Agricultural Systems.”
Agricultural Systems 130 (September): 35–43.
doi:10.1016/j.agsy.2014.06.005.

Rakocy, James E., Michael P. Masser, and Thomas M. Losordo. 2006.

“Recirculating Aquaculture Tank Production Systems: Aquaponics—
integrating Fish and Plant Culture.” SRAC Publication 454: 1–16.

Rockström, J. (1, 2), W. ( 1 Steffen 3 ), K. ( 1 Noone 4 ), A. ( 1 Persson 2 ), C. ( 1

Folke 9 ), B. ( 1 Nykvist 2 ), S. ( 1, 15 ) Sörlin, et al. 2009. “Planetary
Boundaries: Exploring the Safe Operating Space for Humanity.” Ecology
and Society 14 (2).

Rusoff, Louis L., Ernest W. Blakeney, and Dudley D. Culley. 1980. “Duckweeds
(Lemnaceae Family): A Potential Source of Protein and Amino Acids.”
Journal of Agricultural and Food Chemistry 28 (4): 848–50.
doi:10.1021/jf60230a040.

Sheppard, Craig, G. Larry Newton, S.A. Thompson, and S. Savage. 1994. “A

Value Added Manure Management System Using the Black Soldier Fly.”
Bioresource Technology 50 (3): 275–79. doi:10.1016/0960-
8524(94)90102-3.

Skillicorn, Paul*Spira. 1993. “Duckweed Aquaculture  : A New Aquatic Farming

System for Developing Countries.” 11733. The World Bank.
http://documents.worldbank.org/curated/en/1993/03/699091/duckweed-
aquaculture-new-aquatic-farming-system-developing-countries.

St-Hilaire, Sophie, Craig Sheppard, Jeffery K. Tomberlin, Stephen Irving, Larry

Newton, Mark A. McGuire, Erin E. Mosley, Ronald W. Hardy, and Wendy
Sealey. 2007. “Fly Prepupae as a Feedstuff for Rainbow Trout,
Oncorhynchus Mykiss.” Journal of the World Aquaculture Society 38 (1):
59–67. doi:10.1111/j.1749-7345.2006.00073.x.

Tacon, Albert G. J. 2006. “Use of Fishery Resources as Feed Inputs to

Aquaculture Development  : Trends and Policy Implications / by Albert G.J.
Tacon, Mohammad R. Hasan, Rohana P. Subasinghe.” In FAO Fisheries
Circular, No. 1018. Rome  : Food and Agriculture Organization of the
United Nations, 2006.

“The Commercial Bioconversion Digester and Food Scrap Recycler: The

ProtaPodTM.” 2015. Accessed December 1.
http://www.thebiopod.com/pages/pages/protapod.html.

38  
Tomberlin, Jeffery K., D. Craig Sheppard, and John A. Joyce. 2002. “Selected
Life-History Traits of Black Soldier Flies (Diptera: Stratiomyidae) Reared
on Three Artificial Diets.” Annals of the Entomological Society of America
95 (3): 379–86. doi:10.1603/0013-
8746(2002)095[0379:SLHTOB]2.0.CO;2.

Ulloa, J.b., J.h. van Weerd, E.a. Huisman, and J.a.j. Verreth. 2004. “Tropical
Agricultural Residues and Their Potential Uses in Fish Feeds: The Costa
Rican Situation.” Waste Management 24 (January): 87–97.
doi:10.1016/j.wasman.2003.09.003.
US EPA. 2015. “Food Recovery Hierarchy.” Overviews and Factsheets.
September 14. http://www2.epa.gov/sustainable-management-food/food-
recovery-hierarchy.

Wohlfarth, G.W., and G. Hulata. 1987. “Use of Manures in Aquaculture.” In

Detritus and Microbiological Ecology in Aquaculture, 353–67. 14. ICLARM
Conference Proceedings.

Wu, Y.v. ( 1 ), D.j. ( 1 ) Sessa, R. ( 2 ) Rosati, and P. ( 3 ) Brown. 1994.

“Utilization of Protein-Rich Ethanol Co-Products from Corn in Tilapia
Feed.” Journal of the American Oil Chemists’ Society 71 (9): 1041–43.
doi:10.1007/BF02542277.

39