College of Agriculture

SOKOINE UNIVERSITY OF AGRICULTURE

COLLEGE OF AGRICULTURE

DEPARTMENT OF ANIMAL, AQUACULTURE AND RANGE SCIENCES

PROFESSORIAL INAUGURAL LECTURE:

Development of sustainable Nile tilapia (Oreochromis niloticus) culture for improved food security and poverty reduction

Sebastian Wilson Chenyambuga

Department of Animal, Aquaculture and Range Sciences, Sokoine University of Agriculture, P.O. Box 3004, Morogoro, Tanzania.

Executive summary Fish is used in many countries as a primary source of protein. Fish is estimated to account for 17% of the global intake of animal protein and 6.5% of all protein consumed. Fish is also a major source of livelihoods and income, particularly in developing countries. More than 158 million people in the world are estimated to depend directly on fish-related activities. Since the number of fish stocks in natural water bodies has greatly declined, more emphasis on fish production has been directed to aquaculture. Aquaculture has the potential of enhancing food security directly through producing fish for household consumption, improving the supply, and reducing the price of fish in the market. Aquaculture also contributes to farm diversification and creation of new employment opportunities and income streams. Thus, aquaculture is currently promoted as a mechanism for rural development with a focus on poverty alleviation in developing countries.

In Tanzania, aquaculture is an emerging industry, which dominated by pond culture of Nile tilapia (Oreochromis niloticus). The Nile tilapia is a good fish for resource poor farmers because it is easy to raise, it grows fast, and it is tasty, it can eat many types of foods, it is highly tolerant to diseases, it is able to reproduce easily under captivity and can tolerate poor water quality conditions. The demand for tilapia for both domestic consumption and export is high and increasing, but the production ii

from natural water bodies has shown a declining pattern due to overfishing. Thus, there is a need of improving fish production from aquaculture to complement the declining capture of fisheries. Early sexual maturity and prolificacy are the problems, which are associated with pond culture of tilapia, leading to uncontrolled reproduction and overcrowding which, in turn, results in large populations of small-sized fish that are of low value. Poor nutrition is another problem facing pond culture of Nile tilapia. Protein concentrate feeds contribute a major cost component of fish feeds. Because of high cost of conventional protein concentrates, fish farmers use locally available feeds (rice and maize brans, kitchen leftovers, and garden remains) to feed the cultured fish. These are of low quality and fish reared on these feeds are unable to meet their maintenance and production requirements, especially for protein. This practice prolongs the time of reaching the market weight and consequently leading to the production of poor quality fish and low profitability of fish farming. Therefore, there is a need of identifying appropriate alternative cheap sources of protein. To address the above problems a number of studies were done between 2009 and 2018 as described below. The first study was carried out to compare the growth performance and survival of Nile tilapia (O. niloticus), Wami tilapia (O. urolepis) and Ruvuma tilapia (O. ruvumae) under earthen pond and hapa-in-pond culture conditions in order to identify the species that can complement O. niloticus in aquaculture. Two growth experiments were conducted for 90 days. In the first iii

experiment, the fish were raised in earthen ponds while in the second experiment they were grown in hapas installed in earthen ponds. On average, O. niloticus had higher (p ≤ 0.0001) overall mean final body weight (47.17 ± 1.95 g), weight gain (42.75 ± 2.04 g), growth rate (0.48 ± 0.02 g/d) and survival rate (92.82 ± 0.01%) than both O. urolepis and O. ruvumae. On the other hand, O. urolepis had higher overall mean final body weight (33.79 ± 2.24 g), weight gain (28.54 ± 2.34 g) and growth rate than O. ruvumae. To address the problem of over-reproduction of pond cultured Nile tilapia and enable the production of large market size tilapia, two methods for controlling unwanted reproduction were evaluated; polyculture of tilapia with predatory fish (African catfish) and the culture of all-male population. A growth trial was conducted for 120 days in earthen ponds and tilapia were subjected to three treatments, that is, stocking only male tilapia, stocking tilapia of mixed sex together with African catfish at a ratio of 1:10 for catfish: tilapia and stocking tilapia of mixed sex only. The tilapia raised under all-male culture had higher (P ≤ 0.0001) weight gain (65.95 ± 1.19 g) and growth rate (0.55 ± 0.01 g/d) than those under polyculture of tilapia and catfish. On the other hand, tilapia cultured under polyculture system had higher (P ≤ 0.0001) weight gain (55.43 ± 1.81 g) and growth rate (0.46 ± 0.02 g/d) compared to those raised under mix sex culture without catfish (weight gain of 37.71 ± 3.03 g and growth rate of (0.31 ± 0.02 g/d). Since the culture of all-male tilapias resulted in higher growth performance than the polyculture of tilapia with iv

catfish and manual sexing is tedious, a study was conducted to determine the best ways of producing all- male population. Two methods were compared; hormonal sex reversal using 17α-methyltestosterone (17α-MT) and hybridization of Wami tilapia (O. urolepis) and Nile tilapia (O. niloticus). Sex reversed Nile tilapia were produced by feeding newly hatched fry a diet containing 60 mg of 17α-MT per kg of the diet for 28 days. Hybrid tilapias were produced by making a reciprocal cross of O. niloticus ♂ x O. urolepis ♀ and O. niloticus ♀ x O. urolepis ♂. The resulting F1 hybrids and hormonal sex reversed males were compared in a growth experiment conducted for 98 days. The results showed that hybridization of O. niloticus x O. urolepis resulted in slightly higher percentage of males (94%) compared to 17α-MT hormone treatment (90%). However, the percentages of males produced from the two methods were not significantly different (P > 0.05). The hybrids showed higher growth performance (weight gain = 31.54 ± 1.0 g, growth rate = 0.32 ± 0.01 g/d) than the hormonal sex reversed tilapia (weight gain = 25.91 ± 0.95 g, growth rate = 0.26 ± 0.01 g/d), which, in turn, had higher growth performance compared to the mixed sex tilapia (weight gain = 19.30 ± 0.98 g, growth rate = 0.20 ± 0.01 g/d).

To address the problem of unavailability and high price of protein concentrates in fish diets, a study was carried out to evaluate the possibility of replacing soybean meal, which is expensive, with Moringa oleifera leaf meal and sunflower seed cake in tilapia diets. Nine diets were v

formulated with different combination of soybean meal, Moringa leaf meal and sunflower seed cake. Nile tilapia fingerlings were grown in concrete tanks and fed the formulated diets for 90 days. Results indicated that fish fed the diet containing 50% sunflower seed cake and 50% Moringa leaf meal as sources of protein had higher (P ≤ 0.05) average body weight gain (23.03 ± 1.3 g) and growth rate (0.27 ± 0.01 g/d) than the fish fed other diets. Likewise, higher gross margin was obtained from the fish fed the diet containing 50% sunflower seed cake and 50% Moringa leaf meal as protein sources. Therefore, the diet containing a mixture of Moringa leaf meal and sunflower leaf cake in equal proportions was found to be better than that with soybean meal.

In order to avoid unnecessary cost of feed, a study was carried out to determine the appropriate amount of feed and feeding frequency in order to minimize feed wastage and increase profitability. Three levels of feeding (i.e. 1%, 2.5% and 5% of fish body weight) and two feeding frequencies (i.e. daily feeding and alternate days feeding) were evaluated. Nile tilapia fingerlings were grown in concrete tanks for 90 days under a 3 x 2 factorial experiment. Results showed that the fish fed at the feeding level of 5% of body weight had higher (P ≤ 0.05) weight gain, growth rate, survival rate and feed conversion ratio while those fed at 1% feeding level had the lowest weight gain, growth rate, survival and feed conversion ratio. The fish under daily feeding regime had higher (P ≤ 0.05) weight gain, average daily gain and average feed conversion ratio than those under alternate vi

days feeding regime. Feeding level of 5% of body weight and skipping a day resulted into better growth performance than daily feeding at 2.5% level.

To solve the problem of feed wastage in ponds and, hence, minimize production cost, an experiment was conducted to assess the effects of pond fertilization and supplementary feeding on the growth performance and profitability of Nile tilapia. Three treatments were compared namely, weekly fertilization alone with urea and Di-Ammonium Phosphate (DAP) at a rate of 3 g/m2 and 2 g/m2, respectively, concentrate feeding alone at 5% of fish body weight and weekly fertilization with urea and DAP plus concentrate feeding at 2.5% of fish body weight. Nile tilapia were grown in nine concrete tanks and subjected to the three treatments for a period of 166 days. Results indicated that fish reared in concrete tanks under fertilization plus concentrate feeding at 2.5% of fish body weight had higher (P ≤ 0.0001) weight gain (257.37 ± 5.71 g), growth rate (1.50 ± 0.03 g/d) and gross margin (TZS 51,692,352 per ha per year) and lower feed conversion ratio (1.49 ± 0.08) than the fish under concentrate feeding alone at 5% of fish body weight and fertilization alone.

The following conclusions were made: - i. Nile tilapia is superior in growth performance and survival compared to other tilapia species. ii. The culture of all-male population of Nile tilapia results in higher growth performance and bigger fish than polyculture of mixed sex Nile tilapia and African catfish. vii

iii. Growing Nile tilapia of mixed sex together with African catfish in the same pond produces tilapia of relatively larger size than is the case with only growing mixed male and female tilapias in the same pond. iv. The hybrids of Nile tilapia and Wami tilapia have higher growth performance compared to hormonal sex-reversed Nile tilapia. v. The diet containing a mixture of 50% Moringa leaf meal and 50% sunflower seed cake as sources of protein promotes higher growth rate and results in higher profit than the diet containing soybean meal. vi. Daily feeding at the level of 5% of fish body weight promotes higher growth rate than feeding at 2.5% daily or at 5% on alternative days. vii. Feeding at the level of 5% of fish body weight on alternative days (skipping a day) results into better growth performance than daily feeding at 2.5% level. viii. Combination of weekly fertilization of ponds and concentrate feeding at 2.5% of fish body weight promotes higher growth rate and results into higher profit than either weekly fertilization alone or feeding alone at 5% of fish body weight.

viii

Table of Contents

Executive summary ii

List of Tables xi

List of figues xii

Backgraund Infirmation 1

Status of aquaculture in Tanzania 6

Study 1: Assessment of growth performance of three tilapia species raised in earthen ponds and hapas-in- ponds 11

Study 2: Effects of recruitment control by culturing all- male population and using predatory fish on growth performance of Nile tilapia (Oreochromis niloticus) 19

Study 3: Comparison of growth performance of hormonal sex reversed Oreochromis niloticus and hybrids (Oreochromis niloticus x Oreochromis urolepis hornorum) 27

Study 4: Development of a low cost feed based on Moringa oleifera leaf meal and sunflower seed cake as sources of protein in Nile tilapia (Oreochromis niloticus) diets 41

Study 5: Effects of feeding strategies on growth performance and feed utilization of Nile tilapia (Oreochromis niloticus) 55

Study 6: Effects of fertilization and supplementary feeding on water quality, growth performance and profitability of Nile tilapia (Oreochromis niloticus) grown in concrete tanks 65

Conclusions 79

Acknowledgement 81

References 83

LIST OF TABLES

Table 1: Comparison of growth performance, survival, yield and condition factor of mixed-sex, hormonal sex reversed Oreochromis niloticus and hybrids ...... 39

Table 2 Experimental diets containing different replacement levels of soybean meal with moringa leaf meal and sunflower seed cake ...... 46

Table 3: Least squares means for growth performance and nutrient utilization efficiency ...... 53

Table 4:Least squares means for proximate body composition of Oreochromis ...... 54

Table 5: Comparison of growth performance, feed utilization efficiency and ...... 63

Table 6:Comparison of variable costs, revenue and profit obtained from O.niloticus ...... cultured in concrete tanks under three different treatments ...... 78

LIST OF FIGURES

Figure 1:Comparison of growth performance of three tilapia species under pond culture conditions ...... 17

Figure 2:Comparison of growth performance of three tilapia species under hapa-in-pond culture conditions ...... 18

Figure 3: Comparison of growth performance of Nile tilapia grown under all-male monosex culture, mixed sex tilapia-African catfish polyculture and mixed sex tilapia culture without catfish ...... 26

Figure 4: Growth performance of Nile tilapia reared in concrete tanks under different feeding levels and frequencies ...... 62

Figure 5: Comparison of growth performance of O. niloticus cultured in concrete tanks under three different treatments ………………………………………… 76

xii

BACKGROUND INFORMATION

World human population is estimated at 7.6 billion and is expected to reach 9.6 billion by 2050. Feeding the world population of 9.6 billion by 2050 is a challenge all over the world. Hunger and malnutrition are the world’s most devastating problems and are highly linked to poverty. On the other hand, poverty is intimately associated with under-nutrition. Between 2015 and 2016, the prevalence of hunger increased from 10.6% of the global population (777 million people) to 11% (815 million people) (Bennett et al., 2018). According to Bennett et al. (2018), one out of 10 people on the planet suffers from hunger and more than one in five children are stunted (have low height relative to weight), indicating under-nutrition or micronutrient deficiency. Moreover, more than 250 million children worldwide are at risk of vitamin A deficiency (leading to blindness), more than 30% of the world’s population has iron deficient, 200 million people have goitre with 20 million suffering from learning difficulties as a result of iodine deficiency; and 800,000 child deaths per year are attributable to zinc deficiency (Quaas et al., 2016). The challenge facing the world governments and international development agents is to ensure adequate food and nutrition security for all. It is widely recognized that fish production has the potential of contributing positively towards eradication of hunger, food insecurity and malnutrition, if supported and developed in a regulated and sensitive manner that is both environmentally and socially responsible. According to FAO (2014), fish production can play a 1

major role in satisfying the needs of the world’s growing middle-income group while also meeting the food security needs of the poorest.

Fish, both caught from wild water bodies (both marine and freshwater) and produced from aquaculture, are used in many developing countries as a primary source of protein. The estimates by FAO (2014) indicate that fish account for 17% of the human global population’s intake of animal protein and 6.5% of all protein consumed. According to FAO (2014), a portion of 150 g of fish can provide about 50 – 60% of an adult’s daily protein requirements. Moreover, fish is rich in numerous micronutrients that are often missing in diets, particularly those of the poor. Fish contains essential nutrients (such as iodine, vitamin B12 and vitamin D) and long-chain fatty acids (LC-PUFA) (eicosapentaenoic (EPA) and docosahexaenoic (DHA)) omega-3 fatty acids. Moreover, fish is very rich in calcium, iron, zinc and vitamin. Fish consumption provides health benefits to the adult population. There is strong evidence that fish, in particular oily fish, lowers the risk of coronary heart disease mortality by up to 36 percent due to a combination of EPA and DHA (Quaas et al., 2016). Because of health benefits, WHO recommends on average an annual intake of 11.7 kg of fish per person (about 32 g per day or 225 g per week).

In addition to direct consumption, fisheries contribute to food and nutrition security through income generation, increase in the household’s ability to purchase food and 2

provide a source of employment for people who participate in fishing and postharvest activities. Although it is difficult to quantify the extent of their total contribution to income and impact on food security, fisheries activities are a crucial source of income for many people. Recent estimates suggest that fisheries and aquaculture support the livelihoods of 10 to 12% of the world's population and provide income to more than half a billion people worldwide (WorldFish, 2011). It is estimated that more than 158 million people in the world depend directly on fish-related activities (fishing, fish farming, processing, and trading) (HLPE, 2014). Among these people, more than 90% are small-scale operators living in developing countries.

Improvement of fish production and consumption are vital in achieving Sustainable Development Goals number one, which intends to end poverty in all its forms everywhere and number two, which intends to end hunger, achieve food security, improve nutrition, and promote sustainable agriculture. The targets include the reduction at least by half of the proportion of men, women, and children of all ages living in poverty, in all its dimensions, as per the national definitions by 2030. In addition, by 2030, end hunger and ensure access by all people, in particular the poor and people in vulnerable situations, including infants, to safe, nutritious and sufficient food all year round. Both capture fisheries and aquaculture are important food production systems that can make complementary contributions to food and nutrition security. Capture (or wild-caught) fisheries and 3

aquaculture (farmed fish production) together produced about 171 million tonnes in 2016, with aquaculture representing 47% of the total production (FAO, 2018). Capture fisheries, although still the major source of supply of fish, have become static or are in a decline due to over-fishing and environmental degradation. Aquaculture is now believed to have the greatest potential of meeting the growing demand for fish resulting from the increasing population. Half of the fish produced for human consumption came from aquaculture in 2012, compared to just 5% in 1962 and 37% in 2002 (FAO, 2015).

Aquaculture has been growing more rapidly than any other animal food-producing sector in the world. Between 1970 and 2000, the global aquaculture production grew at an average annual rate of 9.2%, compared with only 1.4% for capture fisheries and 2.8% for terrestrial farmed meat production (FAO, 2002). In the last three decades, aquaculture has expanded, intensified, and made major technological advances. Because of its rapid expansion, aquaculture is now considered to have the potential of enhancing food security among adopters and the population at large (Murshed-e-Jahan et al., 2010). Aquaculture can enhance food security directly by the production of fish for household consumption and by improving the supply and reducing the price of fish in the market, and indirectly by contributing to farm diversification and the creation of new employment opportunities and income streams. Because of this logic, aquaculture has been promoted as 4

a mechanism for rural development focusing on poverty alleviation for several decades (World Bank, 2006).

With increasing popularity among consumers, tilapia has become the world’s second most popular farmed fish after carps. Global production of farmed tilapia has increased tremendously from less than 398,000 tonnes in 1991 to a predicted global production of 6.4 million metric tonnes in 2017 (FAO, 2018). Nowadays, tilapias are farmed in at least 85 countries, with most production coming from the developing countries of Asia and Latin America. The global supply of farmed tilapia greatly increased in the 1990s and in early 2000s, largely due to genetic improvements through conventional breeding methods, widespread use of improved tilapia breeds, feed supply availability, effective management of reproduction through sex reversal and hybridization, and expansion of consumer markets. Asia and Latin America dominate the world’s top producers of farmed tilapia. China, Philippines, Mexico, Thailand, Brazil, Egypt, Indonesia, Colombia, Cuba, and Ecuador, together account for 93% of farmed tilapia, and from this Asia, Latin America, and Egypt contribute 70, 19 and 4%, respectively.

The world’s fish consumers are likely to increase due to an increase in the number of people with rising incomes. People with high income seek for higher value protein food of animal origin. Therefore, aquaculture is expected to contribute significantly to meet the increasing demand of fish (World Bank, 2013). It is well accepted that large- 5

scale investment in aquaculture can result in improved nutrition, increased income for the household and enhanced market availability of fish (Fiedler et al., 2016). To produce fish sustainably, aquaculture needs to be carefully managed to protect the resources from over- exploitation and pollution, but at the same time increase production in order to meet the growing demand for nutritious foods.

Status of aquaculture in Tanzania

In Tanzania, fish culture for food production started in 1950s with tilapia species that are native to the region. These include Mozambique tilapia (Oreochromis mossambicus), Nile tilapia (Oreochromis niloticus) and Wami/Zanzibar tilapia (Oreochromis urolepis) (Rice et al., 2006). Other species, which have been used commercially in aquaculture, are O. karongae and O shiranus native to Lake Nyasa (Lake Malawi). Nowdays, aquaculture production includes Nile tilapia (Oreochromis niloticus), African catfish (Clarius gariepinus), rainbow trout (Onchorynchus mykiss) which are cultured in freshwater, and seaweeds (Eucheuma spinosum and Kappaphycus cottonii)), milkfish (Chanos chanos) and prawns produced in marine aquaculture (mariculture) (URT, 2016). In 2014, the total production of aquaculture was estimated to be 3,942 tonnes, and contributed 1% of the total fish production in the country (URT, 2014). Of the 3,942 tonnes, tilapia accounted for 75% of the total production (URT, 2016).

In Tanzania, aquaculture is being promoted as an option for rural development because it provides an important opportunity for reducing poverty and protein malnutrition of the rural poor people. Furthermore, aquaculture is emphasized as an alternative to capture fisheries due to the decline of wild fish from natural water bodies. Aquaculture is important for employment, especially of the rural poor people. It is estimated that 20,000 and 14,100 people are engaged in seaweed and freshwater fish farming, respectively (URT, 2016). Freshwater fish farming, whereby small-scale farmers practice both extensive and semi-intensive fish farming, dominates the industry. In recent years, many non- governmental organizations, church based organizations and research institutes/universities have been promoting aquaculture by distributing Nile tilapia fingerlings to small-scale farmers in rural areas as an effort to contribute effectively to poverty alleviation and improve food and nutrition security among the rural poor people. Because of these efforts, the number of fish ponds has increased from 4,000 in 2000 to 22,702 in 2016 and correspondingly the production of Nile tilapia and African catfish has increased from just 200 tonnes in 2000 to 2,989.5 tonnes in 2012.

In rural areas, small fish ponds of an average size of 10 m x 15 m (150 m2) are integrated with other agricultural activities such as gardening and livestock and poultry production on small pieces of land. Most fish farmers prefer to culture Nile tilapia (Oreochromis niloticus) and few cultures African catfish (Clarias gariepinus). Kaliba 7

et al. (2006) estimated that more than 95% of the farmers culture Nile tilapia (Oreochromis niloticus) in earthen ponds under mixed-sex culture system. On average, small-scale farmers own one to three ponds and the main reasons for engaging in fish farming are the production of fish for home consumption and generation of household income (Chenyambuga and Madalla, 2017). Under small-scale production system, management practices for pond cultured Nile tilapia involve only pond fertilization and feeding. Most small-scale fish farmers in the rural areas fertilize their ponds before stocking using cattle, sheep, goat and chicken manures (Chenyambuga et al., 2014). During the grow-out period, most of the fish farmers apply manure to their ponds irregularly, either once per month or once for every three months and they put the manures in cribs. Majority of the farmers use cattle manure because it is readily available and can be obtained free of charge from the agro-pastoralists living in the nearby areas. Chicken manure is rarely used, despite the fact that almost every household keep chickens rather than cattle or goats. Chicken manure is not preferred because of the small quantity produced from the chickens. Pond fertilization stimulates the growth of phytoplankton. Phytoplankton is food for other organisms (zoo-plankton and larger animals) which are eaten by fish. Fish in the pond can live on natural foods such as phytoplankton, insect larvae, worms, and other small animals. However, natural foods in the ponds are not always abundantly available year-round, hence, fish reared in the ponds need to be fed with concentrate feeds. In Tanzania, feeding of pond cultured Nile tilapia under 8

small-scale production system depends on natural food and concentrate. Most fish farmers in the rural areas provide maize bran, kitchen leftovers and vegetables/weeds as supplementary feeds to their fish (Chenyambuga et al., 2014). This is because these materials are readily available in the rural areas at low price.

Under small-scale production system, it takes six to eight months from stocking to harvesting of Nile tilapia and the yield at harvest is relatively low, averaging about 4,900 kg/ha/year (Chenyambuga et al., 2016). According to Chenyambuga et al. (2016), on average, the family consumes 32% of the fish harvested while selling the remaining 68%. The fish are sold as fresh fish directly to consumers (75%), fish vendors (35%) and retailers (20%). The study carried out in Morogoro Rural, Kilosa, Mpwapwa and Mufindi districts, Tanzania showed that fish farming contributes 19.3% of the total household income (Chenyambuga and Madalla, 2017). The income from aquaculture enterprise is used for house construction, for paying school fees, buying consumer goods, buying livestock to increase herd size, paying medical bills, paying the costs for crop farming and for buying food during the period of food shortage. Therefore, small-scale aquaculture enterprise contributes significantly to household income and wellbeing of rural farmers.

The aquaculture sub-sector in Tanzania has the great potential for expansion since the demand for fish is 9

increasing because of population growth and stagnant production from capture fisheries. Moreover, the export drive for fish and fish products could most likely lead to aquaculture development in the country. Pond culture of Nile tilapia is now viewed as a possible source of livelihood for farmers residing in the proximity of the city and town markets. The emphasis of the national fisheries policy (URT, 2015) is on a semi-intensive integrated mode of fish culture, focusing on Nile tilapia. The Nile tilapia is given the first priority due to its more desirable aquaculture characteristics, including fast growth, short food chain, efficient conversion of food, high fecundity (which provides opportunity for distribution of fingerlings from farmer to farmer), tolerance to a wide range of environmental parameters, and good product quality.

Currently, the Nile tilapia raised by small-scale farmers are of low genetic potential in terms of traits of favourable economic importance such as growth rate and size at maturation. Thus, low productivity and profitability are typical characteristics of small-scale production of tilapia in Tanzania. The demand for tilapia for both domestic consumption and export is high and increasing, but the production from natural water bodies has shown a declining pattern due to overfishing. Thus, there is a need of improving fish production from aquaculture to complement the declining capture fisheries. As an effort of addressing the low productivity and profitability of pond cultured Nile tilapia, many studies were done between 2009 and 2018. 10

Study 1: Assessment of growth performance of three tilapia species raised in earthen ponds and hapas-in- ponds

Introduction

In Tanzania, aquaculture production is dominated by the culture of Nile tilapia (Oreochromis niloticus) and the species have spread in all agro-ecological zones. Apart from the Nile tilapia, there are many other tilapia species, which can be used for aquaculture in the country. Most tilapia species can grow in brackish water, and some are adapted to seawater and most often are grown in ponds, cages and rice fields (Rice et al., 2006). The tilapia species native to Tanzanian inland waters that have the potential for aquaculture production include Oreochromis korogwe, Oreochromis urolepis, Oreochromis jipe, Oreochromis ruvumae, Oreochromis variabilis and Oreochromis esculentus (Trewavas, 1983). However, little research have been conducted on their performance under pond culture conditions. Moreover, there is no research, which has been conducted to compare the growth performance of O. niloticus with the species native to Tanzania under the same culture conditions. Therefore, a study was carried out to evaluate the growth performance, survival and body chemical composition of O. niloticus in comparison to other two tilapia species native to Tanzania, namely O. urolepis and O. ruvumae under earthen pond and hapa-in-pond culture conditions so as to identify the species that can complement O. niloticus in aquaculture. 11

Experimental procedure and fish management

Two studies were conducted to evaluate the growth performance of Nile tilapia (Oreochromis niloticus), Wami tilapia (Oreochromis urolepis) and Ruvuma tilapia (Oreochromis ruvumae). Fingerlings of O. niloticus were obtained from Kingolwira Fish Farming Centre, which is located near Morogoro town. O. urolepis fingerlings were collected from River Wami, around Dakawa village, Morogoro region and O. ruvumae from Ruvuma River at Litapwasi village in Ruvuma region. The fingerlings from each source were kept in plastic containers filled with water and oxygen added and transported to Sokoine University of Agriculture (SUA), Morogoro, Tanzania. The fingerlings were then acclimatized separately in concrete tanks for one month and then stocked separately in earthen ponds where they were cultured to maturity (four months) when they produced the first fingerlings crop. The first fingerlings crop was used in the experiment.

The first study was conducted at Magadu fish farm located at SUA main campus using two ponds, each with a size of 300 m2. Prior to the commencement of the experiment, the ponds were drained, cleaned and allowed to dry for one week. Three hapas were installed in each pond and each hapa had a size of 3 x 2 x 1 m3. The ponds were filled with water and fertilized with poultry manure at a rate of 25 g per m2. During the experimental period, pond fertilization was repeated at monthly intervals. The three tilapia species were randomly allocated to the hapas 12

in a completely randomized design, making two replicates for each species. The fingerlings of each species (O. niloticus, O. urolepis and O. ruvumae) were stocked in separate hapas installed in each pond at 2 fish/m2. During the experimental period, the fish were supplemented with a diet comprised of soybean meal, maize bran, vitamin, and mineral premixes and having crude protein content of 30%. The fish were fed the diet at a feeding rate of 10% of body weight for the first 30 days and then the amount was reduced to 7 and 5% of body weight during the second and third month of the experiment, respectively. The fish were fed once per day at 11:00 hours throughout the experimental period. The experiment was conducted for 90 days. Body weight, length and width were measured from each fish (12 fish from each hapa) at the beginning of the experiment and then at monthly intervals up to the end of the experiment. Fish body weight was measured using an electrical weighing balance while body length and width were measured using a measuring board fixed with a ruler. The following growth parameters were computed: the average weight gain (g), average daily gain (ADG) (g/d), specific growth rate (SGR) (%), the survival rate (%) and condition factor (K).

The second study was conducted using six earthen ponds with an average size of 125 m2 at Changa and Kibwaya villages, Mkuyuni ward, Morogoro Rural District. Before the start of the experiment, the ponds were drained, cleaned and left to dry for one week. The ponds were filled with water and fertilized with chicken manure at a 13

rate of 25 g/m2. The fingerlings were stocked at 2 fish/m2 in each pond. Each species (O. niloticus, O. urolepis and O. ruvumae) was randomly allocated to two ponds. During the experimental period, the same pond fertilization and feeding regimes, which was used in the first study, was adopted. In each pond, a sample of 20 fish was selected for measurement of body weight, length and width. All measurements were carried out at the start of the experiment and then at monthly intervals for a period of 90 days. Water quality parameters (temperature, dissolved oxygen (DO), pH, nitrate (NO3), nitrite (NO2) and turbidity.) were monitored to ensure that they are within the ideal range for tilapia survival. At the end of the experiment, four fish per species in each study were randomly sampled for body chemical composition analysis. All fish were analysed for ash, crude protein (CP) and ether extract (EE) contents according to the proximate analysis scheme of AOAC (2000).

Results and Discussion

Water quality parameters For the ponds at Changa and Kibwaya villages, water temperature ranged from 23.0 to 27.80oC, DO ranged from 2.77 to 7.80 mg/L, pH ranged from 6.0 to 8.40 and the secchi disk reading ranged from 20.0 to 50.0 cm. For the hapas-in-ponds at SUA the temperature, DO, pH and secchi disk readings were 22.60 – 28.60oC, 3.27 – 6.85 mg/L, 6.40 – 7.20 and 30.0 – 56.0 cm, respectively. The concentration of nitrate and nitrite remained in the range 14

of 0 - 10 and 0 - 0.5 mg/L, respectively, in both locations throughout the experimental period. The values for all water quality parameters (temperature, DO, pH, nitrate and nitrite) remained within the safe limits acceptable for tilapia growth and survival under culture conditions throughout the experimental period.

Growth performance The growth performance of the three tilapia species cultured in earthen ponds and hapas-in-ponds are shown in Figures 1 and 2, respectively. The Nile tilapia showed higher growth performance compared to the other species under both culture conditions. The Nile tilapia (O. niloticus) had higher (p ≤ 0.0001) overall mean final body weight (47.17 ± 1.95 g), weight gain (42.75 ± 2.04 g), ADG (0.48 ± 0.02 g/day), SGR (2.30 ± 0.11 % per day), final body length (13.27 ± 0.20 cm) and width (3.89 ± 0.07 cm) than both O. urolepis and O. ruvumae. On the other hand, O. urolepis had higher overall mean final body weight (33.79 ± 2.24 g), weight gain (28.54 ± 2.34 g), ADG (0.32 ± 0.03 g/day), SGR (1.98 ± 0.13 % per day), body length (12.38 ± 0.23 cm) and width (3.61± 0.08 cm) than O. ruvumae. The superiority of O. niloticus to other tilapia species has also been reported by El-Zaeem (2011) who compared O. niloticus and O. aureus. Similarly, Verster (2017) who studied the growth rates of O. niloticus and O. mossambicus in biofloc and standard recirculating aquaculture system found the O. niloticus to be significantly superior to O. mossambicus in terms of average weight gain and specific growth rate.

The condition factor differed significantly among the species. Among the species, O. niloticus showed higher overall mean condition factor (1.83 ± 0.03) than both O. urolepis (1.68 ± 0.03) and O. ruvumae (1.67 ± 0.03). This shows that the health and well wellbeing of Nile tilapia were better than those of the other species. This is supported by the fact that the O. niloticus had the highest survival rate while the O. urolepis had the lowest survival rate in both pond and hapa culture conditions. Therefore, the Nile tilapia is well suited to the culture conditions compared to the other species.

Body chemical composition

Body chemical composition of the three tilapia species did not differ (p ˃ 0.05) in terms of crude protein (CP) and ash contents, but differed (p ≤ 0.001) in ether extract (EE) content. The highest CP content was observed in O. niloticus (60.5 ± 0.3%), followed by O. urolepis (57.4 ± 0.3%) and O. ruvumae (54.3 ± 0.3%) which had the lowest CP content in both culture conditions. The O. urolepis had significantly higher EE content (34.9 ± 3.5%) than both O. niloticus (16.8 ± 3.5%) and O. ruvumae (18.3 ± 3.5%). Ash content was slightly higher in O. urolepis (15.2 ± 0.8%) compared to O. niloticus (13.7 ± 0.8%) and O. ruvumae (14.0 ± 0.8%). Generally, O. urolepis had higher fat content compared to other Oreochomis species. On the other hand, the O. niloticus had higher protein content and low fat content and thus, it is well suited for human consumption.

Figure 1: Comparison of growth performance of three tilapia species under pond culture conditions

Figure 2: Comparison of growth performance of three tilapia species under hapa-in-pond culture conditions

Study 2: Effects of recruitment control by culturing all- male population and using predatory fish on growth performance of Nile tilapia (Oreochromis niloticus)

Introduction

In Tanzania, Nile tilapia (Oreochromis niloticus) is given first priority in aquaculture due to its better characteristics that include fast growth, short food chain, efficient food conversion, high fecundity, tolerance to a wide range of environmental parameters and good product quality. The major drawback of pond culture of Nile tilapia is uncontrolled reproduction that occurs in grow-out ponds. This uncontrolled multiplication caused by the prolific reproduction of Nile tilapia (Oreochromis niloticus) results in overcrowding and competition for food between the adult fish and the newly produced fingerlings (Banerjee, 2010). The ultimate result is that the original stock becomes stunted, resulting into low yield at harvest composed mainly of small fish of low market value.

Techniques for controlling overcrowding in the pond culture setting and thus, enable the production of large market size Nile tilapia include high density stocking, polyculture of tilapia with predatory fish, monosex culture, cage culture, sterilization (through the use of irradiation, chemosterilants and other reproduction inhibitors) and intermittent/selective harvesting (Mair and Little, 1991). Among these methods, the use of local 19

predatory fish species to control unwanted/undesirable tilapia recruitment in ponds seems to be more convenient in rural areas under small-scale farmers’ management level. One of the predatory fish used to control pond overpopulation is the African catfish. Stocking African catfish with mixed sex tilapia populations controls recruitment and allows the original stock to attain a larger market size (de Graaf et al., 1996). Another method for resolving the problem of uncontrolled reproduction in fish ponds is the culture of all-male population. Since males grow faster than females, the production of larger fish and higher yields are possible when all-male stocks are grown. All-male tilapia populations can be produced by hand sexing of the fingerlings, crossing of some tilapia species and hormonal sex reversal. Hand sexing of the fingerlings is the only feasible method that can be applied under farmers’ condition in Tanzania.

Despite the potential benefits of tilapia-catfish polyculture and all-male culture, not much has been done in Tanzania to explore the use of these technologies as a means of controlling tilapia overpopulation in ponds and hence, the need for this study. This study was carried out to compare the effects of all-male culture and tilapia- catfish polyculture on growth performance of Nile tilapia under farmers’ conditions.

Experimental procedure and fish management

The study was carried out at Changa and Kibwaya villages, Mkuyuni Division, Morogoro Rural District, Tanzania. Six ponds, that is three ponds per village, were used in the experiment. The sizes of the ponds ranged from 84 to 285 m2 at the surface area and had the depth of 1.2 m. Prior to the commencement of the experiment, all the ponds were drained, cleaned, allowed to dry for one week and then filled with water from the peripheral canal up to a depth of 1 m. Each pond was fertilized with farmyard manure one week before stocking. In each village, one pond was stocked with male Nile tilapia only (all-male culture) (treatment 1) while the other pond was stocked with mixed sex Nile tilapia and African catfish (Tilapia-catfish polyculture) (treatment 2). The third pond was stocked with Nile tilapia fingerlings of mixed sex (male and female tilapia culture (treatment 3). The tilapias were stocked at a density of three fingerlings per m2 in all the three treatments. In treatment 1, the tilapia fingerlings were manually sexed and the females were removed before being introduced into the ponds designated for all-male culture treatment. For treatment 2, the ratio of catfish: tilapia was 1:10. The catfish were introduced into the ponds 45 days after stocking the ponds with tilapia fingerlings. The experiment was carried out for 120 days. During the experimental period, the fish were supplemented with a concentrate diet comprised of maize bran (79%), soybean meal (20%) and mineral premix (1%). The quantity of feed offered 21

per day was 5% of fish body weight and the amount was adjusted after every measurement of body weight.

Body weight and length of a random sample of 20 Nile tilapia from each pond were measured by using an electronic balance and a measuring board fixed with a ruler, respectively, at the start of the experiment and then at monthly intervals throughout the 120 days of the experiment. Body weight gain, growth rate, specific growth rate and change in body length were computed. In addition, water temperature was determined using a digital thermometer, dissolved oxygen (DO) concentration was determined using a digital DO meter (Jennway 2001) and pH was determined using a digital pH meter (Portmass 911) at monthly intervals during the experimental period.

Results and Discussion

Water quality parameters

The analysis for water quality parameters indicated that the values of pH, DO and temperature did not differ significantly (p > 0.05) among the treatments. The average (± s.e.) pH values ranged from 7.1 ± 0.18 to 7.2 ± 0.13. The highest DO value (7.7 ± 0.21 mg/L) was observed in July while the lowest value (7.4 ± 0.18) mg/L was observed in October 2009. The temperature ranged from 27.2 ± 0.11oC in July to 28.0 ± 0.01oC in November 2009. Growth Performance 22

Figure 3 shows the growth performance of Nile tilapia cultured under all-male culture (treatment 1), tilapia- catfish polyculture (treatment 2) and only mixed sex culture (treatment 3). Results indicated that treatment had significant (p ≤ 0.0001) effects on weight gain and growth rate of the fish. The fish on treatment 1 had the highest weight gain (65.95 ± 1.19 g), followed by those on treatment 2 (55.43 ± 1.81 g) and lastly the fish on treatment 3 (37.71 ± 3.03 g). Likewise, the fish on treatment 1 had significantly (p ≤ 0.0001) higher growth rate (0.55 ± 0.01 g/d) than those on the other treatments. Moreover, fish on treatment 2 had significantly higher growth rate (0.47 ± 0.02 g/d) than those on treatment 3 (0.31 ± 0.02 g/d). Similarly, the final body length values differed significantly (P ≤ 0.001) among treatments. The highest body length value was observed on fish under treatment 1 (12.04 ± 0.09 cm), this was followed by those on treatment 2 (11.23 ± 0.10 cm) and fish on treatment 3 had the lowest body length (9.61 ± 0.2 cm). With regard to feed utilization efficiency, the tilapia on treatment 1 had the lowest (1.23 ± 0.04) feed conversion ratio while those on treatment 3 had the highest feed conversion ratio (2.38 ± 0.63).

Generally, the Nile tilapia under all-male culture showed higher growth rate and gained more weight than the tilapia, which were grown under tilapia-catfish polyculture and mixed sex. This observation compares well with observations in other studies. Lin (1996) who studied the effects of African catfish in tilapia farming 23

obtained significantly higher growth rate and yield of Nile tilapia under all-male culture system than in mixed sex and mixed sex tilapia-catfish polyculture. This is because male tilapia grows faster than females. Therefore, all male culture is desirable in pond farming of tilapia not only to prevent overpopulation and competition for food but also because males grow faster. Studies have shown that all male tilapia populations have higher growth potential because no energy is directed towards the reproduction and there is no competition with younger fish (Green et al., 1997). On the other hand, when females become sexually mature at three to six months, they devote more energy and resources into egg production than into growth and do not eat when they are incubating eggs. This makes them stunted, resulting in small sized fish at harvest. Therefore, the culture of all-male population of Nile tilapia should be promoted as it results in fish with higher growth rate and higher yield at harvest.

The results of this study revealed that the growth performance of tilapia under tilapia and catfish polyculture was significantly better than that under mixed male and female culture system without catfish. This observation concurs with the observation by de Graaf (2004) who reported that the polyculture of Oreochromis niloticus and Clarias gariepinus results in faster growth and higher final weight of both male and female tilapia compared to rearing tilapia alone. Normally, tilapias are cultured with other species to take advantage of many natural foods, which are available in 24

the ponds, to produce a secondary crop, and to control tilapia recruitment. According to de Graaf (2004), stocking predators with mixed sex tilapia populations controls recruitment and allows the original stock to attain a larger market size. In this study, the growth performance of Nile tilapia under polyculture with African catfish was significantly better than that of Nile tilapia cultured under mixed males and females without African catfish.

Figure 3: Comparison of growth performance of Nile tilapia grown under all-male monosex culture, mixed sex tilapia-African catfish polyculture and mixed sex tilapia culture without catfish

Study 3: Comparison of growth performance of hormonal sex reversed Oreochromis niloticus and hybrids (Oreochromis niloticus x Oreochromis urolepis hornorum)

Introduction

Nile tilapia (Oreochromis niloticus) is one of the most important and widely cultured fish species in the world. In Tanzania, more than 95% of fish farmers culture O. niloticus in earthen ponds under mixed-sex culture system (Kaliba et al., 2006). One of the main problems of mixed-sex culture of O. niloticus in the ponds is precocious maturity and uncontrolled reproduction which result into overcrowding and subsequent stunted growth due to competition for feed and space (Omitoyin et al., 2013), hence, leading to low yield at harvest and poor returns. One way of overcoming these undesirable effects of mixed culture of O. niloticus is to culture monosex populations. The culture of all-male populations is preferred to all-female populations because males grow faster and at harvest are larger than females. The culture of all-male populations prevents fingerling recruitment, hence, avoiding overpopulation in the pond and competition for feed and space between recruits and stocked adults.

All-male population of O. niloticus can be produced through several ways, including manual sexing, hormonal sex reversal, hybridization and use of super males as breeding males (Mair and Little, 1991; Abucay 27

and Mair, 1997; Mair et al., 1997). Manual sexing of tilapia is tedious, tricky and requires specially trained personnel. During manual sexing, a significant percentage of females can be mistakenly included in a population of males, and this affects the maximum attainable size of the original stock in grow-out ponds. On the other hand, the production of super males (YY- males) is expensive and not affordable by small-scale farmers. Direct application of male hormones, most commonly 17α-methyltestosterone (17α-MT) is one of the easiest methods of commercial production of monosex male populations in many countries (Abucay and Mair, 1997; Mair et al., 1997). Another method for production of monosex male population which is ideal for small-scale farmers is hybridization of different tilapia species (Mair et al., 1997). According to Wohlfarth (1994) hybrid males can be produced from the following crosses; O. mossambicus x O. urolepis hornorum, O. niloticus x O. urolepis hornorum and O. niloticus x O. aureus. In countries where hormonal sex- reversal of O. niloticus using 17α-MT is prohibited, crossing of O. niloticus with O. aureus is a popular practice and forms one of the most effective methods for producing all-male progenies of tilapia (Marengoni et al., 1998). Moreover, the hybridization of Nile tilapia (O. niloticus) and Wami tilapia (O. urolepis hornorum) results into predominantly male offsprings (Wohlfarth, 1994).

In Tanzania, unfortunately, there is limited information on the most effective method for production of all-male 28

tilapia population. Furthermore, it is not well known whether the hybrid males produced by crossing O. niloticus and O. urolepis hornorum perform better than the sex-reversed males. Therefore, this study was carried out to compare the effectiveness of hormonal sex reversal using 17α-MT and hybridization (O. niloticus x O. urolepis hornorum) for producing all-male population. In addition, the study compared the growth performance and survival rate of hormonal sex reversed O. niloticus males and hybrids of O. niloticus and O. urolepis hornorum.

Experimental procedure and fish management The experiment was conducted at Magadu Fish Farm, Department of Animal, Aquaculture and Range Sciences, of Sokoine University of Agriculture, Morogoro, Tanzania. Fingerlings of O. niloticus with an average weight of 5.70 ± 0.45 g were collected from Lake Victoria at Nyegezi-Bay in Mwanza, Tanzania while O. urolepis hornorum fingerlings with an average weight of 6.10 ± 0.20 g were obtained from Wami River at Wami- Dakawa, Morogoro, Tanzania. The fingerlings were raised for four months before being used as broodstock.

Production of hybrids Before using O. urolepis hornorum and O. niloticus as broodstock, they were morphologically identified according to Eccles (1992) and Skelton (1993). Hybrids were produced by pairing ripe females and males of O. niloticus and O. urolepis hornorum each with an average live weight of 28.00 ± 1.40 g. The chosen broodstock 29

were stocked at a sex ratio of 1 male to 3 females in four 4.5 m2 separate concrete tanks and allowed to spawn naturally. A reciprocal cross of O. niloticus male x O. urolepis hornorum female and O. niloticus female x O. urolepis hornorum male was done. The resulting F1 hybrids were used for growth performance studies.

Production of hormonal sex reversed O. niloticus Two concrete tanks with the capacity of 4.5 m2 were stocked with one O. niloticus male and three O. niloticus females for production of fry for sex reversal. After three weeks, 270 healthy eggs were collected from the mouth of different females from each tank. All eggs were disinfected using 10 mg L-1 potassium permanganate before been put in jar incubators. They were then placed in basket, counted and siphoned up and down into the jar incubator. A constant flow of water was maintained to ensure constant suspension and aeration of the eggs. After four days, all the eggs hatched and were left until the yolk sac has been absorbed as indicated by inflation of the swim bladders. A diet composed of fishmeal (35%), whole maize meal (59%), sunflower oil (2%), wheat meal (3%), and mineral premix (1%) was prepared. A standard solution of 17α-methyltestosterone (17α-MT) hormone was prepared by dissolving 60 mg of 17α-MT in one litre of 95% ethanol. After dissolving the hormone, the solution was evenly sprayed over one kilogram of the diet and mixed by hand several times to ensure homogenous mixing. To allow evaporation of ethanol, the treated feed was spread thinly onto a plastic sheet overnight in large 30

and well-ventilated building at room temperature. The dry hormonal treated feed was packed in bags and crushed into powder form. The feed was then sealed in airtight black container and stored in a refrigerator until use.

A total of 152 fry with a mean weight of 0.0066 g were subjected to hormonal treatment in a 60 litre aquarium at a stocking density of 3 fry/L. All fry were fed the hormonal treated diet twice per day for 28 days at a rate of 20, 15, 10, and 5% of the average body weight during the first, second, third and fourth week, respectively. The aquarium water was continually aerated and siphoning of any debris was done daily to ensure optimum water quality parameters. The water was completely replaced on daily basis.

Growth and survival experiment Nine circular concrete tanks, each having an area of 7 m2 and depth of 1 m were used in the experiment. Water in each tank was completely renewed fortnightly. The experiment consisted of three treatments i.e. mixed-sex O. niloticus (as a control), hormonal sex reversed male O. niloticus and hybrid males of O. niloticus x O. urolepis hornorum. Each tank was stocked with 14 fingerlings having an average weight of 0.01 g. Fish in all tanks were fed a diet containing 30% crude protein. The diet was composed of fishmeal (35%), whole maize meal (59%), sunflower oil (2%), wheat meal (3%) and mineral premix (1%) for the entire period of the study. For the hormonal sex reversed treatment, the fish were 31

fed the diet mixed with 17α-MT for the first 28 days; afterwards they were fed normal diet. During the first week, the fish were provided with the diet at a feeding rate of 20% and the amount was progressively decreased to 15, 10 and 5% of the average body weight during the second, third, and fourth week of the experiment. From the fourth week up to the end of the experiment, the diet was provided at 5% of fish body weight. The fish were fed twice daily, in the morning (between 09:00 and 10:00 h) and in the evening (between 15:00 and 16:00 h). The amount of the diet provided was adjusted every two weeks using the average body weights obtained after each sampling.

Before the experiment, all fish in each tank were individually measured to determine the initial body weight and length. Body weight was measured in g by using a sensitive weighing balance (Boeco BBL, model 43, Hamburg, Germany). Body length was measured in cm by using a measuring ruler. During the experimental period, the fish in each tank were measured for body weight and length every 14 days. At the end of the experiment, the water was completely drained out in each tank and all the fish were collected and counted to determine the survival rate. The hybrids and hormonal treated fish were sexed manually to determine the proportions of males. Then all fish were individually weighed to determine final body weight and length. The following growth performance parameters were computed: weight gain (g), daily weight gain (g/d),

specific growth rate (%), survival rate (%), estimated annual yield (kg/ha/year), and condition factor (K).

Water quality parameters were measured twice daily in the morning (08:30 - 09:00 h) and in the evening (16:30 - 17:00 h). Temperature and dissolved oxygen were measured by using YSI 55 dissolved oxygen meter (model 55, Yellow Spring Instrument Co. Ohio, USA). Water pH was measured using a pH meter (Mardel 5, R029B-MARDEL, USA).

Results and Discussion

Water quality parameters Water quality parameters during the experimental period did not differ among the treatments (p ˃ 0.05). Temperature ranged from 23.03 ± 0.16°C in tanks for hybrids to 23.48 ± 0.13°C in tanks with mixed sex O. niloticus. Dissolved oxygen (DO) was the highest (8.07 ± 0.19 mg/L) in the tanks in which hybrids were raised, followed by the tanks for mixed sex O. niloticus (7.73 ± 0.26 mg/L) and sex reversed O. niloticus (7.72 ± 0.24 mg/L). Water pH values in the tanks were close to neutral with values of 7.19 ± 0.05, 7.23 ± 0.05 and 7.27 ± 0.02 for tanks with sex reversed O. niloticus, hybrids and mixed sex O. niloticus, respectively. Generally, the observed temperature, DO and pH values were within the suitable range of water quality parameters that can ensure better growth and survival of Nile tilapia.

Effectiveness of hormonal sex reversal and hybridization (O. niloticus x O. urolepis hornorum) for producing all-male population

The efficacies of the two methods (i.e. hormonal sex reversal and hybridization of O. niloticus x O. urolepis hornorum) on the production of all male tilapia fingerlings were compared at the end of the experiment. The results show that interspecific hybridization (O. niloticus x O. urolepis hornorum) resulted into slightly higher percentage of male population (94% males) compared to 17α-MT hormone treatment (90% males). However, the chi-square analysis indicated that the percentages of males produced from the two methods were not significantly different (p > 0.05). The observation in this study is consistent with the findings of Mair et al. (1997) who reported that hybridization produces consistently high percentages of males compared to other methods, especially if Oreochromis urolepis hornorum is used as the paternal parent. According to Mair et al. (1997) hormonal sex reversal can produce up to 98% males if properly applied. However, inconsistencies in application often result in lower sex ratios.

Comparison of growth performance, condition factor and survival rate of mixed-sex and hormonal sex reversed Oreochromis niloticus and hybrids

The results on growth performance indicate that mean final body weight, weight gain, growth rate, specific 34

growth rate, estimated annual yield and condition factor differed significantly (p ≤ 0.001) among the treatments. The hybrids had higher final body weight, weight gain, growth rate, specific growth rate, estimated annual yield and condition factor than both sex reversed and mixed- sex O. niloticus (Table 1). The mean final body weight of the hybrids exceeded that of sex-reversed and mixed sex O. niloticus by 5.63 and 12.24 g, respectively. Consequently, the annual yield of the hybrids (2,013.53 ± 187.33 kg/ha/year) was higher than that of sex reversed (1,837.95 ± 199.73 kg/ha/year) and mixed-sex O. niloticus (1,266.15 ± 100.42 kg/ha/year). The faster growth and higher yield of hybrids in the present study is attributed to heterosis effect resulting from hybridization of Oreochromis niloticus and O. urolepis hornorum. Usually, tilapia hybrids show accelerated growth and development, higher viability and greater tolerance to unfavourable environmental conditions and diseases (Cai et al., 2004; Pongthana et al., 2010). The results in the present study are in agreement with the findings of Rothbard et al. (1988) and Al-Hakim et al. (2012) who observed higher growth rate for O. niloticus x O. aureus F1 hybrids compared to pure O. niloticus. Similarly, Garduño-Lugo et al. (2003) obtained higher weight gain and specific growth rate in Red Hybrid Tilapia (RHT) than in pure Stirling Nile Tilapia (SNT). The improvement in growth performance of RHT was attributed to heterosis effect resulting from the cross of Florida Red Tilapia x Stirling red O. niloticus. Moreover, El-Zaeem and Salam (2013) observed the highest heterosis for daily weight gain and specific growth rate 35

in hybrids from the cross between female Oreochromis aureus and male Oreochromis niloticus.

The final body weight, weight gain, growth rate, specific growth rate, estimated annual yield and condition factor for sex reversed O. niloticus were higher (p ≤ 0.0001) than those of mixed-sex O. niloticus (Table 1). The final body weight and annual yield of the fish under mixed sex was lower by 6.61 g and 571.80 kg, respectively, compared to that of sex reversed O. niloticus. Under mixed-sex culture, fish growth rate was lower due to the presence of males and females, which mated after attaining maturity, thus the fish diverted the energy required for growth towards reproduction. Furthermore, in the mixed-sex culture, the growth of the fish was constrained by competition for food and insufficient space due to the existence of both adult fish and fingerlings. The observation that sex reversed O. niloticus had significantly higher mean body weight gain and daily weight gain than mixed-sex O. niloticus corroborates well with the results obtained by Vera Cruz and Mair (1994), Mateen and Ahmed (2007) and Chakraborty and Banerjee (2010). In their studies, they observed that the fish treated with 17α-MT showed higher growth performance than the untreated mixed-sex fish reared under similar conditions. The higher growth performance of hormonal sex reversed O. niloticus is because oral administration of 17α- MT has both androgenic and anabolic effects. According to Yilmaz et al. (2013), the application of 17α- MT hormone in O. niloticus and other species develops male sexual 36

characters and at the same time promotes muscle growth. Moreover, the 17α-MT hormone acts as an appetite stimulant in fish and functions synergistically with insulin to increase protein anabolism (Mukhopadhyay et al., 1986).

Condition factor (K) indicates the health status, fitness or well-being of fish in their habitat and it is assumed that the fish with higher K values are in a better condition. In the present study, the condition factor of O. niloticus differed significantly among the treatments (p ≤ 0.05). Hybrids (1.71 ± 0.05) had higher condition factor than both sex reversed (1.46 ± 0.01) and mixed-sex O. niloticus (1.43 ± 0.01). However, the condition factors of hormonal treated and mixed-sex O. niloticus were not significantly different (p > 0.05). Similar results have been obtained by El-Hawarry (2012) who found higher K values for the hybrids of O. aureus x O. niloticus than the purebred O. niloticus. Ahmad et al. (2002) also obtained higher condition factor values in hybrids compared to hormonally reversed males. The higher condition factor of hybrid tilapia obtained in this study indicates that the hybrid fish had a good health condition and were growing better than hormonal sex reversed and mixed-sex O. niloticus.

Survival rate of sex reversed O. niloticus was slightly higher (95.08 ± 2.92%) than that of mixed sex O. niloticus and hybrids, though the difference was not statistically significant (p > 0.05). The survival rates observed in the present study were somehow higher in all 37

treatments. The high survival rates of the fish in all treatments indicate that the treatments and culture environment in the tanks had no adverse effect on fish survival. The hormone treatment and hybridization did not affect the water quality in the tanks as the water quality did not reach critical levels, which can affect survival rate of the O. niloticus. Indeed, all water quality parameters were optimum for O. niloticus growth and did not differ significantly among the treatments.

Overall, the estimated annual yield was significantly higher in hybrids and hormonal sex reversed O. niloticus than mixed-sex O. niloticus. Similar results have been obtained by El-Hawarry (2012) who observed higher yields for O. aureus x O. niloticus hybrids. Also, Mateen and Ahmed (2007) obtained higher yield, which was 1.6 times greater for 17α-MT treated O. niloticus than for the mixed sex O. niloticus. The higher yields of hybrids and hormonal sex reversed O. niloticus compared to mixed- sex O. niloticus are due to their significant higher growth rates, which resulted into larger body size at the end of the experiment than mixed-sex O. niloticus, hence, higher yields. Therefore, the culture of hybrids of O. niloticus and O. urolepis hornorum is more advantageous compared to the culture of hormonal sex reversed O. niloticus.

Table 1: Comparison of growth performance, survival, yield and condition factor of mixed-sex, hormonal sex reversed Oreochromis niloticus and hybrids

Study 4: Development of a low cost feed based on Moringa oleifera leaf meal and sunflower seed cake as sources of protein in Nile tilapia (Oreochromis niloticus) diets

Introduction Good nutrition in fish production is essential to ensure fast growth and produce good quality products. Furthermore, nutrition is critical in fish farming because feed represents more than 50% of the production costs (De Silva, 1992; Rana et al., 2009). Among the ingredients, protein is the most expensive part of fish feed. For several decades, fishmeal has been used as the main source of protein in fish feeds (El-Sayed, 1999). However, the limited supplies due to competition with humans and livestock and the continuously rising prices of fishmeal make it to be too expensive for small-scale fish farmers in developing countries. Consequently, efforts have been shifted to evaluating potential alternative protein sources for use in fish diets. These alternative protein sources include soybean meal, meat meal and blood meal. Soybean meal has been the focus of attention for substituting fishmeal because of its advantage in terms of protein quality, competitive price and adequate supply (Sharma and Saini, 2017). However, in practice the use of soybean meal has not been readily adopted by small-scale fish farmers in sub-Saharan Africa because of its high price and limited supplies. In Tanzania fish meal and soybean are likely to be reserved for human rather than animal diets. Therefore, there is a need to identify less expensive alternative sources of 41

protein from locally available feed resources and to select protein sources that do not conflict with human food security interests (El-Sayed, 1999; El-Saidy and Gaber, 2002).

Leguminous tree leaves and their pods seem to be appropriate alternative protein sources to fishmeal and soybean (Fernandes et al., 1999; El-Saidy and Gaber, 2003; Richter et al., 2003; Kaushik et al., 2004) as they are readily available in many places and have high protein content. Among the leguminous trees, Moringa oleifera leaves have a considerable potential for becoming fish feed ingredients because of their high nutritional quality. Moringa leaves have crude protein (CP) content of about 25% of dry matter (DM) and contain at least 19 amino acids, several minerals and vitamins (Brilhante et al., 2017). The content of essential amino acids in the raw and extracted moringa leaves is almost similar to that of soybean (Foidl et al., 2001). Moreover, Moringa leaves are free from anti-nutritional factors except for saponins and phenols (Egwui et al., 2013). Studies have shown that in tilapia 10% replacement of fishmeal-based dietary protein using moringa leaf meal in the diets does not cause any adverse effect on growth performance, but higher inclusion levels (20 - 30%) in the diet severely depress growth (Richter et al., 2003). On the other hand, replacement of up to 50% of the more expensive fishmeal and soybean meal protein with sunflower seed cake in Nile tilapia diets does not cause any negative effects on productivity and have some positive impact on profitability (Hossain et al., 2018). 42

Therefore, sunflower seed cake (SFC) is a possible substitute for increasingly expensive and environmentally unsustainable fishmeal in aquaculture feeds (Maina et al., 2007). Sunflower seed cake contains high level of crude protein (28 - 37%), which varies with seed quality and processing (Munguti et al., 2006). According to Maina et al. (2007), the amino acid profiles of SFC and fishmeal are comparable, though SFC has relatively lower levels of lysine and threonine than fishmeal.

Therefore, the present study was designed to evaluate whether addition of sunflower seed cake into diets containing moringa leaf meal would overcome the negative effects of moringa leaf meal and improve the growth performance of Nile tilapia (Oreochromis niloticus L.) cultured under semi-intensive production system. Specifically, the study evaluated the effects of different dietary inclusion levels of moringa leaf meal and sunflower seed cake as replacement of soybean meal on growth performance, feed utilization and profitability of Oreochromis niloticus raised under semi-intensive system.

Experimental procedure and fish management

This study was carried out at Magadu aquaculture research facility of the Department of Animal, Aquaculture and Range Sciences, Sokoine University of Agriculture (SUA). Moringa leaves were harvested from plots located within SUA main campus. The leaves were 43

dried under shade to constant weight, crushed by hand into small pieces and soaked into water for 24 hours to remove water soluble anti-nutritional factors such as saponins and tannins. Then the leaves were dried under the sun. The dried leaves were ground using a hammer mill into powder form to pass through 1 mm sieve and then stored in plastic containers at room temperature until use. Other ingredients used to formulate the diets were soybean meal, sunflower seed cake, hominy meal, sunflower oil, wheat meal and mineral premix and were purchased from livestock input supply shops in Morogoro.

The chemical compositions of moringa leaf meal, soybean meal, sunflower seed cake, hominy meal and wheat meal were analyzed using proximate analysis scheme procedure (AOAC, 2000). Following chemical composition analysis, nine isonitrogenous diets, each containing about 20% crude protein, were formulated. The diets were designated as D1, D2, D3, D4, D5, D6, D7, D8 and D9. Diet D1 was a control diet and contained 100% soybean meal as a major source of protein. For the rest of the diets soybean meal (SBM) was replaced with either moringa leaf meal (MLM) or sunflower seed cake (SFC) or a combination of MLM and SFC at different levels. In diets D2 and D3 SBM was completely replaced with SFC (100%) and MLM (100%), respectively, as the main sources of protein. In diet D4, SBM was replaced with a combination of 75% SFC and 25% MLM. Diet D5 contained 50% SFC and 50% MLM as the main sources of protein while D6 had combination of 25% SFC and 44

75% MLM as the main source of protein. In diets D7, D8 and D9 MLM replaced SBM at the levels of 25, 50 and 75%, respectively. The proportions of all ingredients in the nine diets are shown in Table 2 below.

The mixture of the feed ingredients of each diet was made into pellets using an electric meat grinder (NMG- 745, Nikai Japan Ltd, Japan). The pellets were dried under the sun and stored in plastic containers at room temperature until use. The pellets from each diet were analyzed to determine the chemical composition using the proximate analysis scheme procedure (AOAC, 2000).

Table 2 Experimental diets containing different replacement levels of soybean meal with moringa leaf meal and sunflower seed cake

Ingredients D1 D2 D3 D4 D5 D6 D7 D8 D9 SFC 0 84 0 66 41 21 0 0 0 SBM 25 0 0 0 0 0 24 20 13 MLM 0 0 80 22 41 63 7 20 39 HM 69 10 14 6 12 10 63 54 42 SFO 3 3 3 3 3 3 3 3 3 MIN 1 1 1 1 1 1 1 1 1 WM 2 2 2 2 2 2 2 2 2 Total 100 100 100 100 100 100 100 100 100 Note: SFC = Sunflower seed cake, SBM = Soybean meal, MLM = Moringa leaf meal, HM = Hominy meal, SFO = Sunflower oil, MIN = Mineral premix, WM = Wheat meal, D1 = Diet 1, D2 = Diet 2, D3 = Diet 3, D4 = Diet 4, D5 = Diet 5, D6 = Diet 6, D7 = Diet 7, D8 = Diet 8, D9 = Diet 9.

The nine diets were randomly allocated to 18 concrete tanks in a completely randomized design. Each diet was replicated twice. Each tank had a surface area of 7.06 m2 and depth of 1 m. The tanks were filled with water. A total 378 Nile tilapia fingerlings with a mean weight of 5.50 g were stocked in 18 experimental concrete tanks at a density of three fingerlings/m2. The fingerlings were randomly assigned to the tanks. The fingerlings were fed their

respective experimental diets at 5% of body weight and three times per day at 09:00 h, 13:00 h and 17:00 h. The amount of feed provided to the fish in each tank was adjusted fortnightly in response to the changes in body weight of the fish. Before stocking, the fingerlings in each experimental tank were measured for body weight to determine initial body weight. Body weight of each fingerling was measured using a digital weighing balance (Mettler PM 11, Mettler Instrument LTD, Switzerland). Then body weight of individual fish in each tank was weighed every two weeks up to the end of the experiment. The experiment lasted for 90 days. At the end of the experiment, all fish in each tank were harvested, counted and weighed. Weight gain, growth rate, specific growth rate and survival rate were computed. Also, feed intake and feed conversion ratio were computed.

Water quality parameters were monitored during the experimental period. Dissolved oxygen and temperature were measured using dissolved oxygen metre (YSI 55- 12FT). Nitrogenous compounds (nitrates and nitrites) and pH were measured using test strips (JBL EasyTest) and test vials (Tetratest vials). Conductivity and salinity were measured using an automatic salinity refractometer (ACT- S/Mill-E, Cat.NO.2444) while water turbidity was measured using a locally made Secchi disk.

At the end of the experiment six fish were sampled from each treatment, three fish per tank, and analyzed for protein, fat and ash contents of the carcasses using the proximate analysis scheme (AOAC, 2000). Economic analysis was 47

done by computing gross margin. The gross margin (GM) for each diet was obtained by subtracting total variable costs (TVC) from total revenue (TR). The total variable costs were obtained by adding the feed cost and cost of fingerlings. The total revenues were obtained by multiplying the total weight of fish harvested in each tank by market fish price.

Results and Discussion The values of mean water quality parameter were within the acceptable limits for Oreochromis niloticus throughout the experimental period. The mean (±s.e.) temperature was 29.21 ± 0.30C while mean pH and dissolved oxygen were 7.65 ± 0.1 and 7.03±0.3 mg/L, respectively. The mean values for the other parameters were; salinity = 4.27 ± 0.3 ppt, nitrate = 7.82±0.3 mg/L; nitrite = 0.25±0.02 mg/L and secchi disk readings = 76.11±2 cm. Generally, all water quality parameters were within the favorable range for Nile tilapia growth. Therefore, the growth variations of the fish fed different diets could not be due to differences in water quality.

Both moringa leaf meal and sunflower seed cake have been reported to be suitable for use as protein sources in place of soybean meal in tilapia feeds (Afuang et al., 2003; Hossain et al., 2018). Moringa leaf meal and sunflower seed cake have a great potential for being used as ingredients in fish diet formulations in Tanzania because they are locally available, affordable and have relatively good nutritional value. In the present study the effect of replacing soybean meal with sunflower seed cake and moringa leaf meal either singly or in various combinations on growth performance 48

of Nile tilapia was evaluated. Results on growth performance of Nile tilapia fed different diets containing different levels of sunflower seed cake and moringa leaf meal are presented in Table 3. The results indicate that mean final body weight, weight gain, growth rate and specific growth rate differed significantly among the fish fed different diets. The highest mean final body weight, weight gain, growth rate and specific growth rate were observed in the fish fed diet D5 in which soybean meal was completely replaced with a combination of 50% sunflower seed cake and 50% moringa leaf meal as protein sources. These were followed by the fish, which were fed diet D6 that contained 25% sunflower seed cake and 75% moringa leaf meal as protein source. The fish fed the control diet D1, which contained soybean meal as the only main source of protein, ranked third in terms of final body weight, weight gain, growth rate and specific growth rate while those fed diet D2, which contained sunflower seed cake as the only source of protein, showed the least growth performance. Similarly feed utilization efficiency differed significantly among the fish fed different diets. Like for growth performance, the fish fed diet D5 had the highest feed intake and lowest FCR, indicating that the diet was efficiently utilized for growth. Generally, this study has shown that the inclusion of a mixture of 50% sunflower seed cake and 50% Moringa leaf meal in Nile tilapia diet promotes faster growth. And the fish fed this diet show better feed utilization efficiency compared to the fish fed diet based on either soybean meal or sunflower seed cake or moringa leaf meal as the sole source of protein. It seems that the combination of sunflower seed cake with moringa leaf meal complements better than the combination of soybean meal

with moringa leaf meal and can completely replace soybean meal in fish diet without negatively affecting growth and feed utilization efficiency of Nile tilapia.

Comparison of the fish fed diets D1, D2 and D3 which were based on soybean meal, sunflower seed cake and moringa leaf meal as sources of protein, respectively, indicated that fish fed diet D2 had lower (p > 0.05) body weight gain, growth rate and specific growth rate compared to those fed diet D1 and D3 (Table 3). The body weight gain, growth rate and specific growth rate of the fish, which were fed the diet based on moringa leaf meal as sole protein source (D3), were not significantly different from that of the fish fed diet D1 based on soybean meal. Therefore, moringa leaf meal can be used in Nile tilapia diet to replace soybean meal as a plant protein source. This is supported by the findings of Puycha et al. (2017) who said that moringa leaf meal could be used as a supplement in fish diet at more than 100 g/kg to support growth with no adverse effect on the digestibility and serum biochemistry parameters. According to Makkar and Becker (1996), moringa leaf meal contains high crude protein of about 260 g/kg of leaf, of which about 87% is true protein. Moreover, the amino acid profile of moringa leaf is comparable to that of soybean (Foidl et al., 2001). Moringa leaves contain abundant essential amino acids methionine, cysteine, typtophan and lysine, hence, their inclusion in the diets improves the nutritional value of the animal feeds (Afuanget et al., 2003). The use of the mixture of sunflower seed cake and moringa leaf meal in Nile tilapia diets seems to be promising for improving productivity while at the same time reducing diet cost and increasing profitability. This is supported by the fact that

the use of diet D5 resulted in higher gross margin compared to the other diets.

In the present study the use of sunflower seed cake as the sole source of protein reduced significantly the growth performance of Nile tilapia. This is contrary to Hossain et al. (2018) who reported that sunflower seed cake can be used up to 40% inclusion in fish feed as an alternative plant protein source to fish meal and soybean without depressing growth performance. The lower growth performance of fish fed the diet containing sunflower seed cake as the only source of protein may be the result of low lysine and threonine content. Sunflower seed cake is rich in methionine and arginine, but has low levels of lysine and threonine (Merida et al., 2010). Both lysine and threonine are essential amino acids in O. niloticus and are, therefore, required in the diet as they limit growth of juvenile and finishing Nile tilapia (Michelato et al., 2016). Studies have shown that the deficiencies of threonine negatively affect weight gain in juvenile tilapia (Yue et al., 2014). Therefore, the addition of moringa leaf meal, which is rich in lysine and threonine (Abbas et al., 2018), complemented the amino acids lacking in sunflower seed cake and contributed to increased growth performance for the fish fed the diets containing the mixture of sunflower seed cake and moringa leaf meal.

The effects of diets on proximate composition of fish whole body is shown in Table 4. Results show that the type of diet influenced (p ≤ 0.05) the proximate composition of fish carcasses. Fish fed diet D1 had significantly higher carcass crude protein and fat contents than the fish fed the rest of

the diets while those fed diet D3 showed the lowest. Soybean meal, which was the main protein source in diet D1, had significantly higher protein (41.64% CP) and fat (11.75% EE) contents compared to sunflower seed cake (22.03% CP and 10.91% EE) and Moringa leaf meal (22.50% CP and 4.18% EE). Consequently, the fish fed the diet containing soybean meal as the main source of protein contained higher protein and fat than those fed either sunflower seed cake or moringa leaf meal based diets. The ash content, which is an indication of mineral content in fish bodies, differed (P ≤ 0.05) among the fish fed different diets. Carcasses of fish fed diet D3 that contained moringa leaf as the main protein source had significantly higher ash content compared to that of those fed other diets. Generally, the ash content of carcasses of fish fed diets containing moringa leaf meal was higher than that of the fish fed soybean meal and sunflower seed cake based diets. This study has demonstrated that Nile tilapia (O. niloticus) diets formulated with a combination of 50% sunflower seed cake and 50% moringa leaf meal or 25% sunflower seed cake and 75% moringa leaf meal as replacement for soybean meal protein promote higher growth performance and result into higher gross margin.

= Diet 7, D SR = Survival r weight gain, ADG = Average daily gain, SGR = Specific growth rate, FCR = Feed conversion rate, 0.05p ≤ abcd

Table

Least squares means with different superscript

IniWt = Initial body weight, FiWt = Final body weight, FI = Feed intake, BWtG = Body

parameters and

8 Least squares means for growth performance and nutrient utilization efficiency

= Diet 8, D

ate,

= Diet 1, D

= Diet

survival rate

= Diet 2, D

of of

Oreochromis niloticus

= Diet 3, D

letter within the same row differ significantly at

= Diet 4, D

fed fed different diets (mean ±sd)

= Diet 5, D

= Diet = Diet 6, D

8, D D Nitrogen con Moisture = Moi Note: 0.05p ≤ abcd

Table

= Diet 1, D 1, Diet =

Least squares means with different superscript letter within the same row differ significantly at significantly differ row same the within letter superscriptdifferent with means squares Least

= Diet 9

free extract.

Least square niloticus

D 2, Diet =

fed different diets

s means for proximate body composition of composition body fors means proximate

tent, DM = Dry matter, CP = Crude protein, EE = Ether extract, NFE = NFE extract, Ether = EE protein, Crude = CP matter, Dry = DM tent,

= Diet 3, D 3, Diet =

= Diet 4, D 4, Diet =

= Diet 5, D 5, Diet =

= Diet 6, D 6, Diet =

Oreochromis Oreochromis

= Diet 7, D 7, =Diet

= Diet =

Study 5: Effects of feeding strategies on growth performance and feed utilization of Nile tilapia (Oreochromis niloticus)

Introduction

In fish farming most of production cost is contributed by feed. It is estimated that about 60 to 80% of cost of inputs under intensive fish culture are contributed by feed (Hossain et al., 2018). Thus, the amount of feeds offered to fish determine the economics of production because feed contributes more to the cost of production than other factors. Low feeding level can cause stunted growth and hence, low yield at harvest and less income while feeding excessively leads to feed wastage and hence, water pollution and increased production cost (Abou-Zied and Ali, 2015). Both feed wastage and water quality deterioration increase the production cost through the amount of feed used and the cost of replacing water in the pond. Therefore, it is important to develop a good feeding strategy that can minimize feed wastage and deterioration of water quality. By using the correct amount of feed and feeding frequency, unnecessary cost can be avoided, and thus increase profit of the aquaculture enterprise.

A feeding strategy is the feeding plan designed to achieve more profit and sustainability of fish production with more advantage and less disadvantage to the people and environment. In fish management, use of appropriate feeding strategies is necessary in order to gain an 55

economic advantage and to maximize growth and feed conversion efficiency. Improper feeding or feeding strategies may result in slow growth and poor feed conversion efficiency, and thus cause unnecessary cost. Reduction of production costs and maintaining the quality of water are the most important priorities in aquaculture. According to Cho et al. (2003) the determination of optimum ration amount and feeding frequency are important prerequisites in aquaculture operation since they ensure better feed conversion ratio of cultured fish. It is, therefore, important to determine the optimal feeding level and frequency, which will maximize the utilization of feed by fish and reduce feed wastage, which can negatively affect water quality and profitability during the course of production. Therefore, this study was carried out to determine the most appropriate level of feeding and feeding frequency, which can ensure that the amount of feed provided, is reasonable and promotes fast growth.

Experimental procedure and fish management

A study was conducted at Magadu Aquaculture Research unit, Department of Animal, Aquaculture and Range Sciences, Sokoine University of Agriculture. The study aimed at determining the appropriate feeding level and feeding regime, which can support higher growth of Nile tilapia (Oreochromis niloticus) under small-scale production system. Three feeding levels were evaluated i.e. 1, 2.5 and 5% of the fish live weight. In addition, two feeding frequencies were assessed i.e. daily feeding and 56

alternate days feeding. A 3 x 2 factorial experiment under completely randomised design was adopted and treatments were made of the combinations between feeding level and feeding frequency Thus, the total number of treatments were six and they were randomly allotted to 18 concrete tanks. Each treatment was replicated three times. A total of 378 fingerlings with mean weight of 10 g were stocked in 18 concrete tanks, each with a size of 7.06 m2. The fingerlings were stocked at a stocking density of three fingerlings per m2, making the total number of fingerlings to be 21 per tank. The fish were fed twice per day at 10:00 h and 17:00 h with a diet containing 28.3% crude protein and energy content of 15.6 MJ/kg feed. The main sources of protein in the diet were fishmeal, Moringa oleifera leaf meal and sunflower seed cake. The ingredients for the diet were Moringa oleifera leaf meal (34.5%), sunflower seed cake (34.5%), hominy meal (12%), fishmeal (13%), sunflower oil (3%), wheat flour (2%) and mineral premix (1%). The study was carried out for 90 days. Data on feed offered were recorded daily and body weights of individual fish were measured every two weeks. The amount of feed provided to the fish in each tank was adjusted every two weeks based on the body weight. In addition, water quality parameters were measured to ensure that the conditions were favourable for fish growth. At the end of the experiment all fish in each tank were harvested, counted and individually weighed to determine final body weight. Survival rate was computed. From the body weight data, the following variables were computed: mean final weight, weight gain, average daily gain (ADG), specific 57

growth rate (SGR), feed intake (FI) and feed conversion ratio (FCR). Also after harvesting, 10 fish per treatment (five per tank and 10 for each combination of feeding level and feeding frequency) were randomly sampled and their crude protein content in the carcass determined using proximate analysis scheme (AOAC, 2000).

Results and Discussion

Figure 4 shows the growth performance of Nile tilapia subjected to different feeding levels and frequencies. The fish fed daily at a feeding level of 5% of body weight had the highest growth performance, followed by those fed at 5% of body weight on alternative days. The lowest growth performance was observed on the fish fed at 1% of body weight on alternative days. The analysis of variance revealed that the interaction between feeding level and feeding frequency was not significant. Results in Table 5 indicate that feeding level significantly (p ≤ 0.05) affected all growth performance parameters i.e. final body weight, weight gain, average daily gain, specific growth rate (SGR) and feed conversion ratio (FCR). Fish provided feed at the feeding level of 5% of body weight showed the highest final body weight, weight gain, average daily gain, specific growth rate and FCR, followed by those fed at 2.5% while those fed at 1% of body weight had the lowest values. Similarly, the fish fed at the level of 5% of body weight had higher (p ≤ 0.05) carcass crude protein content compared to those fed at 2.5 and 1% of body. The fish reared under the feeding regime of daily feeding at 5% of body weight 58

had higher feed intake, and hence, more nutrients were assimilated into their bodies and this resulted into faster growth. This concurs with the findings of Deyab and Hussein (2015) who said that feeding rate of 5% of body weight daily is the optimal feeding rate for red tilapia fingerlings and significantly enhances fish growth and feed utilization. According to El‐Saidy and Gaber (2005) growth rate, specific growth rate and body weight increase significantly with increasing feeding levels .

Feed contributes substantially to the cost of fish production, and studies have shown that the expenses on feed determine the difference between profitable and unprofitable aquaculture enterprises (Bolivar and Jimenez, 2006). Reducing the amount of feed offered to fish in the ponds is a means of lowering costs if production is not reduced. Optimal feeding increases feed intake and promotes higher growth rate. In the present study, the effects of daily feeding and alternate day feeding were compared. The analysis of variance revealed that the fish fed daily had higher (p ≤ 0.05) final body weight, weight gain, average daily gain, specific growth rate (SGR) and feed conversion ratio compared to those fed on alternate days (Table 6). Nile tilapia fed two times every day showed the best growth performance, indicating that daily feeding is better than alternate days feeding and that the feeding interval of 5 - 12 hours between meals corresponds well with gastric evacuation times and this improves feed intake and digestibility. The results of the present study are supported by the findings reported by Byamungu et al. (2001) that feeding tilapia 59

only five days a week (instead of seven) has a negative effect on growth while increased feeding frequency improves growth. However, the findings in the current study differs with the results of Bolivar and Jimenez (2006) who found that feeding on alternate days does not reduce significantly the growth performance and yield of Nile tilapia. This is also supported by Ali et al. (2016) who said that it is not necessary to feed daily in order to obtain maximum growth rates. The difference between the results of the current study and those obtained by Bolivar and Jimenez (2006) may be because the fish in the current study were raised in concrete tanks, thus, depended entirely on the feeds provided for their nutrient intake. Bolivar and Jimenez (2006) used earthen ponds; hence, the fish got some nutrients from natural food available in the ponds. Pond water usually contains phytoplankton and zooplanton, which are the natural food that are eaten by fish.

Feed conversion ratio was significantly affected by both feeding level and feeding frequency. The FCR of the fish fed at a feeding level of 5% of body weight was higher than of those fed at 2.5%, which in turn, had higher FCR than those fed at 1% of body (Table 5). In addition, the fish, which were fed daily, had higher FCR than those fed on alternate days (Table 6). This is in agreement with the observation made by El‐Saidy and Gaber (2005) that FCR increases with increase in feeding rate. The provision of large quantity of feed results into excess intake, which, in turn, causes a worse FCR. On the other hand, reducing the amount of feed through lowering 60

feeding level or reducing feeding frequency decreases feed intake and maximizes the efficiency of converting food into body tissues. This is supported by the findings by Bolivar and Jimenez (2006) who reported that a significantly better FCR is obtained when fish are fed on alternate days than for fish fed daily.

The survival rate of Nile tilapia did not differ significantly (p ˃ 0.05) between the two feeding frequencies (daily feeding and alternate days feeding), but differed (p ≤ 0.05) among the different feeding levels. The highest survival rate was observed on fish subjected to daily feeding and 5% of body weight (98.41%) while the lowest was found on the fish, which were fed daily at 2.5% of body weight (82.54%). On average, the fish which were subjected to daily feeding had higher survival rate than those under alternate days feeding regime (Table 6). With regard to feeding levels, overall, survival of the fish decreased with decrease in feeding level. Generally, daily feeding and feeding level of 5% of fish body weight is the best feeding strategy for Nile tilapia.

Figure 4: Growth performance of Nile tilapia reared in concrete tanks under different feeding levels and frequencies

INWTNote: FNWT = Initial, weight, = Final WG = Weight gain, ADG daily gain, = Average SGR ab

Table

Means with different superscript letter within the same row differ significantly p at 0.05≤

content and SR = survival rate. (mineral) ash body Fish = Ash content, extract ether body Fish = EE content, protein crude Conver Feed = FCR rate, growth Specific =

chemical of Nile tilapia under body composition

Comparison growth of efficiency utilization performance, feed and

sion Ratio, FI = Feed intake, CP = Fish body Fish = CP intake, Feed = FI Ratio, sion

differentfeeding levels

abc

Table Specific = SGR gain, daily Average = ADG gain, weight = WG weight, final = FNWT Note: ab

Means with different superscript letter within the same row differ significantly p at

rti cnet E = ih oy te etat otn, s = ih oy s (minera ash body Fish = Ash content and SR = survival rate. content, crude extract body ether Fish body = Fish CP = intake, EE content, Feed protein = FI Ratio, Conversion Feed = FCR rate, growth 6

of Nile tilapia under daily feeding and alternate feeding days Comparison growth of performancefeed and efficiency parameters utilization

≤ 0.05≤

Study 6: Effects of fertilization and supplementary feeding on water quality, growth performance and profitability of Nile tilapia (Oreochromis niloticus) grown in concrete tanks

Introduction

Feed costs in aquaculture production account for approximately 50% of total operational costs (Rana et al., 2009) and is considered to be the major constraint for both small-scale fish farmers and commercial fish farmers. Reducing the amount of feed provided to fish in a pond is a means of lowering costs and increasing profit margin, if production is not reduced. One way of reducing cost is fertilizer application. When ponds are fertilized, nutrients in the fertilizer stimulate the growth of phytoplankton. Phytoplankton is food for other organisms (zoo-plankton and larger animals) that are eaten by fish. Weekly fertilization of fish pond has been shown to increase fish yields by increasing primary productivity through released inorganic nutrients, or by providing organic carbon through heterotrophic pathways (Knud-Hansen et al. 1991). Availability of natural food in pond water reduces fish requirement for artificial feeds, leading to low production costs and increased farm income. However, the application of fertilizers of any type and form alone cannot suffice the nutritional requirements of fish for reasonable growth (Abbas et al., 2014) and thus, supplementation of artificial feed is inevitable for satisfactory fish growth and obtaining higher yields. 65

Water quality in fish ponds is a major factor determining the growth performance of fish (Egna and Boyd, 1997). Maintaining proper water quality parameters is very important for survival, growth, and reproduction of aquatic organisms, hence, they need to be monitored. Pond fertilization with excessive amount of fertilizer can cause severe environmental issues due to high concentration of algae that lead to algal bloom. Algal bloom hinders light penetration in the water leading to decrease in photosynthesis rate and consequently decrease in the amount of dissolved oxygen in fish ponds. Furthermore, excessive pond fertilization results into excess nitrogen input that causes high unionized ammonia concentrations, which may reduce fish growth or cause mortality. Moreover, provision of feeds in excess of what can be taken by the fish leads to wastage of diet and diet waste means deteriorated water quality and economical losses (Ali et al., 2010). This situation often causes mass mortality of tilapia and consequently the farmers abandon fish farming.

Pond culture is the most common method of raising Nile tilapia in Tanzania. Among the primary factors that limit the production capacity of a pond is the quantity of available nutrients, which form basic materials for structure and growth of living organisms. Proper pond fertilization and supplementary feeding techniques are used to supply these nutrients in optimal quantities, thereby overcoming natural deficiencies, in order to obtain maximum possible fish yield from a pond. A study done in Bangladesh (Wahab et al., 2014) indicated 66

that proper pond fertilization combined with supplementary feeding at 50% satiation level results in higher fish yield and benefit-cost ratio for polyculture of tilapia and Silver carp than culturing on fertilizer application or 100% feeding alone. Another study done in Cambodia (Phanna et al., 2014) showed that pond fertilization plus supplementary feeding (50% satiation) is the most effective feeding for optimization of production, feed conversion ratio, and growth performance of tilapia cultured under semi-intensive system or small-scale aquaculture. In Tanzania, information on the combined effects of fertilization and supplementary feeding on water quality parameters in ponds and growth performance of fish is lacking. It is not known which pond management system can be used by small-scale farmers in order to reduce cost and increase production and profit. Therefore, the overall purpose of this study was to identify the best management practice that can be used by small-scale fish farmers for improving fish production in Tanzania. Specifically, the study aimed at establishing a low cost sustainable tilapia culture method by evaluating the effects of pond fertilization alone, feeding alone and combination of pond fertilization with supplementary feeding on water quality parameters, growth performance, yield and profit margin of Nile tilapia.

Experimental procedure and fish management

The study was conducted at Magadu fish farm, Sokoine University of Agriculture (SUA), Morogoro. The experiment was conducted using nine concrete tanks, each having a surface area of 3.36 m2. The treatments were weekly fertilizer application alone with urea and Di-Ammonium Phosphate (DAP) at a rate of 3 g/m2 and 2 2 g/m , respectively (T1), concentrate feeding alone at 5% of fish body weight (T2) and weekly fertilization with urea and DAP plus concentrate feeding at 2.5% of fish body weight (T3). The three treatments were assigned randomly to the concrete tanks and each treatment was replicated three times. The concentrate feed comprised of fish meal (25%), cotton seed cake (10%), sunflower seed cake (10%), maize meal (4%), wheat bran (50%) and mineral premix (1%) and had a crude protein content of 30% CP. Prior to the start of the experiment, all tanks were drained, cleaned, dried for five days and then refilled with water and fertilized with urea and DAP at a rate of 3 g/m2 and 2 g/m2, respectively (except those under T2), and left for 14 days before being stocked with fingerlings. The tanks were stocked with sex-reversed Nile tilapia with average weight of 1.1 g at a stocking density of 3 fingerlings per m2. Fingerlings cultured in the tanks under T2 and T3 were fed 10 and 5% of their body weights, respectively, for the first two months and thereafter the amount of feed was reduced to 5 and 2.5% for T2 and T3, respectively. The fish were provided with feed twice per day at 10:00 h and 16:00 h and the experiment took 166 days. 68

Fish body weights and lengths were measured at the beginning of the experiment and then after every two weeks. All fish in each tank were taken for measurement of body weight and length. Weight of individual fish was measured by using digital weighing scale while body length was measured by using a ruler. Pond water pH, dissolved oxygen (DO), conductivity, total dissolved solids, salinity, temperature, alkalinity, total phosphorus, ammonia and nitrate were measured in each tank every week. Water samples were collected at three depths (i.e.at the top, middle and just off the bottom of the tank) of the water column and then mixed and put into 500 ml vials for determination of the water quality parameters. Water quality parameters were measured weekly between 09:00 h and 10:00 h.

Results and Discussion

Water quality parameters Water quality parameters important for fish growth are dissolved oxygen, temperature, pH, salinity and ammonia. Dissolved oxygen (DO) is the most important water quality parameter in fish ponds, as oxygen is needed by fish and other aquatic organisms for breathing and survival. In the present study, results on water quality parameters show that concrete tanks subjected to the treatment of fertilization alone (T1) (9.97 ± 0.21 mg/L) and fertilization plus feeding (T3) (7.84 ± 0.20 mg/L) had higher (p ≤ 0.0001) DO values than those under concentrate feeding alone (T2) (6.51 ± 0.20 mg/L). This is because fertilizer application in T1 and T3 69

increased the production of phytoplankton. The phytoplankton produce most oxygen in water through photosynthesis, the process in which green plants use solar energy to convert water and carbon dioxide to oxygen and carbohydrates (Knud-Hansen, 1998). Furthermore, the lower DO levels in T2 were caused by oxygen depletion as a result of decomposition of uneaten feeds. Aerobic bacteria use oxygen to decompose feed materials, and this contributes to low oxygen levels in pond water. The average pH values were not different (p ˃ 0.05) among the concrete tanks under T1 (8.53 ± 0.10), T2 (8.06 ± 0.44) and T3 (7.90 ± 0.13). The results of this study are in agreement with the findings by Ibrahim and Nagdi (2006) who reported that pond fertilization with chemical fertilizers helps to maintain high DO and moderate pH values. Water salinity, temperature, alkalinity and ammonia were not affected (p ˃ 0.05) by treatment. Average salinity was 0.04, 0.04 and 0.03 PSU in concrete tanks under T1, T2 and T3, respectively. Mean temperature ranged from 27.53 ± 0.11 to 27.82 ± 0.10 oC in concrete tanks under T1 and T3, respectively. Average alkalinity values were 74.20 ± 5.18 mg/L for T1, 78.21 ± 5.10 mg/L for T2 and 71.76 ± 3.52 mg/L for T3. The average levels of ammonia were5.32 ± 0.27, 4.72 ± 0.19 and 4.77 ± 0.23 mg/L for T1, T2 and T3, respectively. The concentrations of nitrate and phosphorus in the tanks under the three treatments were significantly different. The highest concentration of nitrate was observed in concrete tanks under T1 (6.54 ± 0.35 mg/L) while the lowest was found in tanks under T3 (5.68 ± 0.28 mg/L). The levels of phosphorus in water ranged from 0.21 ± 70

0.04 mg/L in T2 to 0.69 ± 0.08 mg/L in T3. Similar results on water quality parameters were obtained by El Naggar et al. (2008) in Egypt. Generally, the values for water pH, DO, salinity and temperature in the current study were within the normal range for Nile tilapia growth. According to Popma and Lovshin (1995) the ideal water temperature, pH, DO and salinity for Nile tilapia growth range from 25°C to 30°C, 6 to 9, 3 to 6 mg/L and 5 to 10 ppt, respectively. These values make the fish thrive well in pond water and grow fast.

Growth performance, feed utilization efficiency and condition factor of Nile tilapia (O. niloticus) cultured in concrete tanks under three treatments

Figure 5 shows the growth performance of Nile tilapia cultured under three treatments (fertilization alone (T1), feeding alone (T2) and a combination of fertilization and feeding (T3)). Fish cultured under T3 showed the highest growth performance, followed by those under T2 while those under T1 had the lowest growth performance. Table 7 shows the growth performance, feed conversion ratio, condition factor (K), survival and estimated yield of Nile tilapia (O. niloticus) cultured in concrete tanks under three treatments (fertilization alone (T1), feeding alone (T2) and a combination of fertilization and feeding (T3)). The results indicate that fish cultured in concrete tanks under T1 had significantly lower final body weight, weight gain, growth rate, specific growth rate and yield than those cultured in tanks under T2 and T3. Fish cultured in tanks under T2 on average gained 173 g while 71

those on T3 gained 216.26 g more weight than those on T1. The results of the present study concur with the findings of Abbas et al. (2014) who concluded that application of fertilizers of any type and form cannot suffice the nutritional requirements of fish for reasonable growth and thus, supplementation of artificial feed is inevitable for satisfactory fish growth and higher yields. This implies that natural food produced as the result of pond fertilization should not be used as sole source of food for fish, but should be combined with concentrate supplementation in order to promote fast growth and produce high yield of farmed Nile tilapia.

When comparison is made between T2 and T3, it can be seen that fish reared in concrete tanks subjected to weekly fertilization and daily feeding (T3) had significantly higher final body weight, weight gain, growth rate, specific growth rate and yield than those under daily feeding alone (T2). The mean final body weight, growth rate and estimated yield per ha per year of fish reared in tanks under T3 exceeded those of fish reared in tanks under T2 by 43.26 g, 0.25 g/d and 2,440.29 kg, respectively. Moreover, the fish under T3 had significantly lower FCR than those on T2, mainly because they were given half the amount of feed provided to the fish under T2, but they grew faster compared to those under T2. This is because weekly fertilization of concrete tanks with urea and DAP fertilizers in T3 increased the production of natural food organisms (phytoplankton and zooplankton) which were then eaten by the fish cultured in those tanks. According 72

to Knud-Hansen (1998), pond fertilization increases natural food production by stimulating algal productivity. This, in turn, promotes higher fish growth due to increase in the available natural food (phytoplankton and zooplankton). Elnady et al. (2010) showed that the growth of fish is strongly correlated with increase in phytoplankton and zooplankton productivity as a result of fertilization.

Condition factor (K) reflects the physiological state of a fish in relation to its welfare. Higher value of condition factor indicates better condition experienced by the fish (Anani and Nunoo, 2016). In the present study, the condition factors of the fish under T2 and T3 were significantly higher than that of the fish under T1, implying that the environmental conditions in concrete tanks under T2 and T3 were more favourable for the growth and survival of Nile tilapia compared to that of the concrete tanks under T1. On the other hand, the condition factor for the fish under T2 was slightly higher than of those on T3, though not significantly different. Similar results have been obtained by Abdel-Warith (2013) who obtained higher condition factor for the fish, which received only artificial feed without any fertilization compared to those under weekly fertilization with either chicken manure or chemical fertilizer plus artificial feeding. In the present study survival rate of fish did not differ (P ˃ 0.05) among the treatments and ranged from 87.7 to 90.0% in concrete tanks under T1 and T2, respectively.

The results for economic analysis are shown in Table 8 for fish reared in concrete tanks subjected to the three treatments. Among the variable costs, fertilizer and fingerling costs did not differ (p ˃ 0.05) among the treatments while labour and feed costs differed significantly among the treatments. Tanks under T3 had the highest labour costs while those under T1 had the lowest. The highest feed cost was observed under T2. Feed cost accounted for 66.06 and 44.95% of total variable costs in treatment T2 and T3, respectively. Fertilizer cost did not differ (p ˃ 0.05) between the treatments T1 and T3, mainly because the same amount of fertilizer was used in the two treatments. Treatment influenced significantly the revenue and gross margin. Fish reared in tanks under T3 resulted into significantly higher revenue and gross margin than those cultured under T2 and T1. Fish reared under T2 had higher revenue and gross margin than those under T1. The fish reared under T1 resulted into a loss mainly because of low yield because of significantly slow growth rate. The results in the present study concur with Waidbacher et al. (2006) who observed higher fish growth, yield, better feed utilization and profits for fish reared under pond fertilization and feeding concentrate diet at 6% of body weight compared to those which were subjected to either feeding alone or fertilization alone. According to Elnady et al. (2010) application of medium amount of chemical fertilizers (urea and TSP) as pond inputs can produce up to a three- fold increase in fish yield. In this study, it was observed that fish reared under T3 produced the highest profit, followed by those reared under T2 while those 74

under T1 resulted into a loss. These results indicate that pond fertilization combined with concentrate feeding at 2.5% of fish body weight reduced feed costs, and hence, increased the profit of Nile tilapia reared in concrete tanks under T3 compared to those reared by only feeding at 5% of fish body weight without any fertilization. This agrees with Wahab et al. (2014) who reported that availability of natural food in pond water reduces fish requirement for artificial feeds, leading to reduced production costs and improved farm income. Artificial feed costs in aquaculture operations account for approximately 50 - 70% of total operational costs. Therefore, the application of inorganic fertilizers can be used as a means for reducing the need for supplementary feeds and, therefore, can lower the production cost and increase the profit of fish farming enterprises.

Figure 5: Comparison of growth performance of O. niloticus cultured in concrete tanks under three different treatments

those with different ≤ 0.001. at P significantly differ superscript abc of Nile tilapia ( Table

lette superscript same the with Means

Comparison growth of condition factor, performance, and yield FCR, survival

O. niloticus

) cultured under in concrete three tanks

( differ not do row same the in r

different treatments

P ˃ 0.05) while while 0.05) ˃ P

Table

abc

Means different with at letters superscript differ p ≤ 0.05. significantly

cultured under in concrete three tanks different treatments

Comparison variable of profit obtained costs, and from revenue

O.niloticus

CONCLUSIONS

Six studies were conducted to develop low cost methods for improving the production of Nile tilapia (Oreochromis niloticus) under semi-intensive production systems. From these studies, the following conclusions were made: - i. The Nile tilapia (Oreochromis niloticus) is superior to Wami tilapia (Oreochromis urolepis hornorum) and Ruvuma tilapia (Oreochromis ruvumae) in terms of growth performance, survival and body chemical composition. ii. The Wami tilapia (Oreochromis urolepis hornorum) ranks second to Nile tilapia (O. niloticus) in terms of growth performance and can be used in aquaculture in place of Nile tilapia in areas where it is readily available. iii. The culture of all-male Nile tilapia significantly improves growth performance and results into larger fish and higher yields at harvest than the polyculture of mixed sex Nile tilapia and African catfish. iv. The culture of Nile tilapia of mixed sex together with African catfish in the same pond controls excessive reproduction and produces tilapia of larger size compared to the culture of mixed male and female tilapias without the African catfish. v. The hybrids of Nile tilapia (O. niloticus) and Wami tilapia (O. urolepis hornorum) grow faster and produce higher yield at harvest compared to

hormonal sex reversed and mixed-sex O. niloticus. vi. Hormonal treatment of O. niloticus results into fish, which grow faster and produce higher yields at harvest than mixed sex O. niloticus. vii. Nile tilapia (O. niloticus) diets formulated with a combination of 50% sunflower seed cake and 50% moringa leaf meal or 25% sunflower seed cake and 75% moringa leaf meal as protein sources promote higher growth performance and result into higher profit that the diets containing soybean meal as the main protein source. viii. The use of diet based on sunflower seed cake as the main source of protein in place of soybean meal reduces the growth performance of Nile tilapia and result into lower profit. ix. Moringa leaf meal is a better plant protein source than sunflower seed cake and its use in Nile tilapia diet result in growth performance that is not significantly different from that of the fish fed soybean meal based diet. However, for better results moringa leaf meal should be mixed with sunflower seed cake in equal proportions to replace the more expensive soybean meal. x. Daily feeding of Nile tilapia at 5% of body weight promotes higher growth rate and survival. Moreover, feeding at the level of 5% of body weight on alternate days results into better growth performance compared to daily feeding at 2.5% of body weight.

xi. The combination of weekly fertilization of concrete tanks with inorganic fertilizers and concentre diet feeding at 2.5% of fish body weight promotes higher growth rate and results into higher profit than either weekly fertilization alone or feeding alone at 5% of fish body weight. Moreover, the combination of weekly fertilization and concentre diet feeding at 2.5% of fish body weight does not affect water quality parameters beyond the range recommend for tilapia growth.

Acknowledgement

Funding for five studies, out of the six studies, was provided by the AquaFish Innovation Lab. The AquaFish Innovation Lab is supported in part by United States Agency for International Development (USAID) Cooperative Agreement No. EPP-A-00-06-00012-00 and by contributions from participating institutions. One study was funded by NORAD through the programme for Enhancing Pro-Poor Innovation in Natural Resources and Agricultural Value Chain (EPINAV) at Sokoine University of Agricultre, Tanzania. The author acknowledges the financial support from both USAID and NORAD. Moreover, I am grateful for the assistance provided by farmers and village extension officers during the undertaking of the various studies in Mvomero, Kilosa, Mpwapwa and Mbarali districts. I would like to thank the Vice Chancellor of Sokoine University of Agriculture for funding the preparation of this document. 81

References

Abbas R.K., Elsharbasy F.S. and Fadlelmula A.A. (2018). Nutritional values of Moringa oleifera, total protein, amino acid, vitamins, minerals, carbohydrates, total fat and crude fiber, under the semi-arid conditions of Sudan. Journal of Microbial & Biochemical Technology 10(2): 56 - 58. Abbas S., Ashraf M., and Ahmed I. (2014). Effect of fertilization and supplementary feeding on growth performance of Labeo rohita, Catla catla and Cyprinus carpio. The Journal of Animal and Plant Sciences 24(1): 142 - 148. Abdel-Warith A-W.A. (2013). Effect of fertilization sources with artificial feeding on growth performance, water quality and returns of monosex Nile tilapia (Oreochromisniloticus) reared in earthen ponds. Egyptian journal of aquatic biology and fisheries Vol 17. DO - 10.12816/0011036 Abou-Zied R.M. and Ali A.A.A. (2015). Effect of feeding rate and frequency on growth performance, sex conversion ratio and profitability of Nile tilapia (Oreochromis niloticus) fry in hapa at commercial hathceries. Egyptian Journal of Nutrition and Feeds 18(2): 451 - 459. Abucay J.S. and G.C. Mair (1997). Hormonal sex reversal of tilapias: implications of hormone treatment application in closed water systems. Aquaculture Research 28: 841 - 845. Afuang W., P. Siddhuranju, and K. Becker. (2003). Comparative nutritional evaluation of raw, methanol 83

extract residues and methanol extract of Moringa (Moringa oleifera Lam.) leaves on growth performance and feed utilization in Nile Tilapia (Oreochromis niloticus L.). Aquaculture Research 34:1147 - 1159. Ahmad M.H., A.M.E. Shalaby, Y.A.E. Khattab and M. Abdel-Tawwab (2002). Effects of 17 α– methyltestosterone on growth performance and some physiological changes of Nile tilapia fingerlings (Oreochromis niloticus L.). Egyptian Journal of Aquatic Biology and Fisheries 4: 295 - 311. Al-Hakim N.F.A., M. Saleh, A.Z. Hegazi, A.I.K. Aly and A.M. Tahoun (2012). Induction of mono-sex (male tilapia) population by inter-specific hybridization and hormonal sex reversal of Nile tilapia. Egyptian Journal of Aquatic Biology and Fisheries 17: 23 - 33. Ali M., Hayward R.S., Bajer P.G. and Whitledge G.W. (2010). Maintenance/submaximum feeding schedules for reducing solid wastes and improving feed conversion in aquaculture. Journal of the World Aquaculture Society 41(3): 319 - 331. Ali T. E., Silvia Martínez-Llorens, A.V. Moñino, Cerda´ M.J., Tomás-Vidal A. (2016). Effects of weekly feeding frequency and previous ration restriction on the compensatory growth and body composition of Nile tilapia fingerlings. Egyptian Journal of Aquatic Research (2016) 42, 357–363. Anani F.A. and Nunoo F.K.E. (2016). Length-weight relationship and condition factor of Nile tilapia, 84

Oreochromis niloticus fed farm-made and commercial tilapia diet. International Journal of Fisheries and Aquatic Studies 4(5): 647 – 650. AOAC (2000). Official methods of analysis. Fifteen Edition. Association of Official Analytical Chemists (AOAC), Arlington, Virginia, USA. 1298p. Banerjee S. (2010). Comparative growth performance of mixed-sex and monosex Nile tilapia population in freshwater cage culture system under Indian perspective. International Journal of Biology 2: 44 – 50. Bennett A., Patil P., Kleisner K., Rader D., Virdin J., and Basurto X. (2018). Contribution of fisheries to food and nutrition security: Current knowledge, policy, and research. NI Report 18-02. Durham, NC: Duke University, http://nicholasinstitute.duke.edu/publication. Bolivar R.B. and Jimenez E.B. (2006). Alternate-day feeding strategy for Nile tilapia grow out in the Philippines: Marginal cost–revenue analyses. North American Journal of Aquaculture 68:192–197. Brilhante R.S.N., Sales J.A., Pereira V.S., Castelo- Branco D.S.C.M., Cordeiro R.A., de Souza Sampaio C.M., de Araújo Neto Paiva M., Santos J.B.F.D., Sidrim J.J.C., Rocha M.F.G. (2017). Research advances on the multiple uses of Moringa oleifera: A sustainable alternative for socially neglected population. Asian Pacific Journal of Tropical Medicine 10(7): 621 - 630. doi: 10.1016/j.apjtm.2017.07.002.

Byamungu N., Darras V.M. and Kuhn E.R. (2001). Growth of heat-shock induced triploids of blue tilapia, Oreochromis aureus, reared in tanks and in ponds in Eastern Congo: Feeding regimes and compensatory growth response of triploid females. Aquaculture 198:109–122. Cai W.Q., S.F. Li and J.Y. Ma (2004). Diseases resistance of Nile tilapia (Oreochromis niloticus), blue tilapia (Oreochromis aureus) and their hybrid (female Nile tilapia× male blue tilapia) to Aeromonas sobria. Aquaculture 229: 79 - 87. Chakraborty S.B. and S. Banerjee (2010). Comparative growth performance of mixed-sex and monosex Nile tilapia population in freshwater cage culture system under Indian perspective. International Journal of Biology 2, 44-50. Chenyambuga S W, Mwandya A, Lamtane H A and Madalla N A (2014). Productivity and marketing of Nile tilapia (Oreochromis niloticus) cultured in ponds of small-scale farmers in Mvomero and Mbarali districts, Tanzania. Livestock Research for Rural Development. Volume 26, Article #43. http://www.lrrd.org/lrrd26/3/chen26043.htm. Retrieved on 13th March 2014. Chenyambuga S.W., Elibariki E. Msuya and Nazael A. Madalla. (2016). Assessment of Value Chain of Farmed Nile tilapia (Oreochromis niloticus) in Coastal and Lake zones of Tanzania. Presented at the ASIAN-PACIFIC AQUACULTURE 2016 held at Grand City Convention Center, Surabaya, Indonesia, 27 - 29th April 2016. 86

Chenyambuga S.W. and Nazael A. Madalla (2017). Contribution of small-scale aquaculture to rural livelihoods in Tanzania. Presented at the WORLD AQUACULTURE 2017 Conference, held at the International Convention Centre, Cape Town. South Africa, June 26 - 30th, 2017. Cho S.H., Lim Y.S., Lee J.H., Lee J.K. and Park S. (2003). Effects of feeding rate and feeding frequency on survival, growth, and body Composition of Ayu post-larvae, Plecoglossusaltivelis. Journal of the World Aquaculture Society 34: 85 - 91. de Graaf G. (2004). Optimisation of the pond rearing of Nile tilapia (Oreochromis niloticus L.): The impact of stunting processes and recruitment control. PhD thesis, Wageningen University, The Netherlands. de Graaf G.D., Galemoni F. and Banzoussi 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: 85-100. De Silva S.S. (1992). Fish nutrition research for semi- intensive culture systems in Asia. Asian Fisheries Science 5: 129-144. Deyab E.D.M.S. and Hussein E.E.M. (2015). Effects of Different Feeding Rates on Growth Performance and Body Composition of Red Tilapia, Oreochromis mossambiquse x O. niloticus, Fingerlings. International Journal of Aquaculture 5(12). doi: 10.5376/ija.2015.05.0012. 87

Eccles D.H. (1992). FAO species identification sheets for fishery purposes. Field guide to the freshwater fishes of Tanzania: United Nations Development Programme, Project URT/87/016, FAO, Rome, Italy. Egna H. and Boyd C. (1997). Dynamics of Pond Aquaculture. CRC Press, Boca Raton, FL. 472p. Egwui P.C., Mgbenka B.O. and Ezeonyejiaku C.D. (2013). Moringa plant and its use as feed in aquaculture development: A review. Animal Research International 10. (1):1673 - 1680. El-Hawarry W.N. (2012). Growth performance, proximate muscle composition and dress-out percentage of Nile tilapia (Oreochromis niloticus), Blue tilapia (Oreochromis aureus) and their interspecific hybrid (♂ O. aureus X ♀ O. niloticus) cultured in semi-intensive culture system World's Veterinary Journal 2: 17-22. Elnady M.A., Hassanien H.A., Salem M.A. and Samir H. M. (2010). Algal abundances and growth performances of Nile tilapia (Oreochromis niloticus) as affected by different fertilizer Sources. Journal of American Science 6(11): 584 - 593. El Naggar G. O., Ibrahim N.A. and Abou Zead M.Y. (2008). Influence of fertilizers' types and stocking density on water quality and growth performance of Nile tilapia - African catfish in polyculture system. In: Proceedings of the 8th International Symposium on Tilapia in Aquaculture, page 157 - 171. El-Saidy D.M.S.D. and Gaber M.M.A. (2002). Complete Replacement of Fish Meal by Soybean Meal with 88

Dietary L-Lysine Supplementation for Nile Tilapia Oreochromis niloticus (L.) Fingerlings. Journal of the World Aquaculture Society 33 (3):297 - 306. El-Saidy D.M.S.D. and Gaber M.M.A. (2003). Replacement of fish meal with a mixture of different plant protein sources in juvenile Nile tilapia, Oreochromis niloticus (L.) diets. Aquaculture Research 34 (13):1119-1127. El-Sayed A.F.M. (1999). Alternative dietary protein sources for farmed tilapia, Oreochromis spp. Aquaculture 179 (1-4):149-168. El‐Saidy D.M.S.D. and Gaber M.M.A. (2005). Effect of dietary protein levels and feeding rates on growth performance, production traits and body composition of Nile tilapia, Oreochromis niloticus (L.) cultured in concrete tanks. Aquaculture Research 36(2): 163 - 171. https://doi.org/10.1111/j.1365-2109.2004.01201.x El-Zaeem S.Y. (2011). Growth comparison of Nile tilapia (Oreochromis niloticus) and Blue tilapia, (Oreochromis aureus) as affected by classical and modern breeding methods. African Journal of Biotechnology 10(56): 12071 - 12078. El-Zaeem S. and G. Salam (2013). Production of genetically male tilapia through interspecific hybridization between Oreochromis niloticus and O. aureus. Iranian Journal of Fisheries Sciences 12: 813 - 832. FAO. (2002). The State of World Fisheries and Aquaculture. Food and Agriculture Organization of the United Nations, Rome. 124 p. 89

FAO (2014). The State of World Fisheries and Aquaculture: Opportunities and challenges. Food and Agriculture Organization of the United Nations, Rome. FAO (2015). The State of Food and Agriculture - Social protection and agriculture: breaking the cycle of rural poverty. Food and Agriculture Organization of the United Nations, Rome. FAO (2018). The State of World Fisheries and Aquaculture 2018 - Meeting the sustainable development goals. Food and Agriculture Organization of the United Nations, Rome. Fernandes A.E.,Gomes M.A., Henriques R.and Combra J. (1999). Replacement of fishmeal by plant proteins in the diet of Nile tilapia: digestibility and growth performance. Aquaculure International 7: 57-67. Fiedler J.L., Lividini K., Drummond E. and Thilsted S.H. (2016). Strengthening the contribution of aquaculture to food and nutrition security: The potential of a vitamin A-rich, small fish in Bangladesh. Aquaculture 452: 291–303. Foidl N., Makkar H. P. S.and Becker K. (2001). The potential of Moringa oleifera for agricultural and industrial uses. In: What development potential for Moringa products? 20th October - 2nd November, 2001. Dar Es Salaam, Tanzania. Garduño-Lugo M., I. Granados-Alvarez, M.A. Olvera- Novoa and G. Muñoz-Córdova (2003). Comparison of growth, fillet yield and proximate composition between Stirling Nile tilapia (wild type) (Oreochromis niloticus, Linnaeus) and Red hybrid 90

tilapia (Florida red tilapia × Stirling red O. niloticus) males. Aquaculture Research 34: 1023-1028. Green B.N.W., K.L. Veverica and M.S. Fitzpatrick. (1997). Fry and fingerlings production. In: Dynamics of Pond Aquaculture (ed. Egna, H. and C. Boyd), pp: 215 – 243. CRC Press, Boca Raton, Florida, USA. HLPE (2014). Sustainable fisheries and aquaculture for food security and nutrition: A report by the High Level Panel of Experts on Food Security and Nutrition of the Committee on World Food Security, Food and Agriculture Organization of the United Nations, Rome. Hossain M.M., Chakraborty S.C. and Brown C.L. (2018). Evaluation of diets with sunflower cake as a replacement for fish meal in feeds for Nile tilapia (Oreochromis niloticus) production. Australian Journal of Science and Technology 2(2): 90 - 98. Ibrahim N. A. and Z. A. Nagdi. (2006). Effect of nitrogen fertilizer source on water quality and performance of two tilapia species. Egyptian Journal of Applied Science 21 (6): 12 - 22. Kaliba A.R., K.O. Osewe, E.M. Senkondo, B.V. Mnembuka and K.K. Quagrainie (2006). Economic analysis of Nile tilapia (Oreochromis niloticus) production in Tanzania. Journal of the World Aquaculture Society 37: 464 - 473. Kaushik S.J., Covès D., Dutto G. and Blanc D. (2004). Almost total replacement of fish meal by plant protein sources in the diet of a marine teleost, the

European seabass, Dicentrarchus labrax. Aquaculture 230, 391–404. Knud-Hansen C.F. (1998). Pond fertilization: Ecological approach and practical applications. Aquaculture Collaborative Research Support Program, Oregon State University, Corvallis, Oregon, USA. pp125. Knud-Hansen C.F., Batterson T.R., McNabb C.D., Harahat I.S., Sumantadinata K. and Eidman H.M. (1991). Nitrogen input, primary productivity and fish yield in fertilized freshwater ponds in Indonesia. Aquaculture 94: 49 - 63. Lin, C.K. (1996). Clarias and Tilapia interaction in polyculture. http://pdacrsp.oregonstate.edu/pubs/nops/ful_rprts/9 6-94.pdf. accessed on 26 September 2018. Maina J.G., Beames R.M., Higgs D., Mbugua P.M., Iwama G. and Kisia S.M. (2007). The feeding value and protein quality in high-fibre and fibre-reduced sunflower cakes and Kenya’s “omena” fishmeal for tilapia (Oreochromis niloticus). Livestock Research Rural Development. Available from: http:www.lrrd.org/lrrd19/11/main1964.htm. Retrieved on 3rd November 2018. Mair G.C., L.R. Dahilig, E.J. Morales, J.A. Beardmore and D.O.F. Skibinski (1997). Application of genetic techniques for the production of monosex male tilapia in aquaculture: Early experiences from the Philippines. In: Proceedings of the Fourth Central America Symposium on Aquaculture, pp. 225-227, Tegucigalpa, Honduras.

Mair G.C. and D.C. Little (1991). Population control in farmed tilapias. NAGA, International Center for Living Aquatic Resources Management (ICLARM.) Quarterly 4: 8 - 13. Makkar H.P.S. and Becker K. (1996) Nutritional value and anti-nutritional components of whole and ethanol extracted Moringa oleifera leaves. Animal Feed Science and Technology 63:211 - 228. Marengoni N.G., Y. Onoue and T. Oyama (1998). All- male tilapia hybrids of two strains of Oreochromis niloticus. Journal of the World Aquaculture Society 29: 108 - 113. Mateen A. and I. Ahmed (2007). Effect of androgen on sex reversal and growth of Nile tilapia (Oreochromis niloticus). Pakistan Journal of Agricultural Sciences 44: 272 - 276. Merida S.N., Tomas-Vidal A., Martinez-Llorens S., Cerda M.J. (2010). Sunflower meal as a partial substitute in juvenile sharpsnout sea bream (Diplodus puntazzo) diets: Amino acid retention, gut and liver histology. Aquaculture 298:275 - 281. Michelato M., Vidal L.V., Xavier T.O., de Moura L.B., de Almeida F.L., Pedrosa VB, Furuya V.R.B. and Furuya W.M. (2016). Dietary lysine requirement to enhance muscle development and fillet yield of finishing Nile tilapia. Aquaculture 457:124 - 130. Mukhopadhyay P., B. Venkatesh and P. Das (1986). Growth and some biochemical changes in Clarias batrachus due to methyltestosterone. Indian Journal of Fisheries 33: 262 - 269.

Munguti J.M., Liti D.M., Waidbacher H., Straif M., Zollitsch W. (2006). Proximate composition of selected potential feedstuffs for Nile tilapia (Oreochromis niloticus Linnaeus) production in Kenya. Die Bodenkultur 57(3):131 - 141. Murshed-e-Jahan K., Ahmed M. and Belton B. (2010). The impacts of aquaculture development on food security: Lessons from Bangladesh. Aquaculture Research 41: 481 - 495. Omitoyin B.O., E.K. Ajani and H.O. Sadiq (2013). Preliminary investigation of Tribulus terrestris (Linn., 1753) extracts as natural sex reversal agent in Oreochromis niloticus (Linn., 1758) larvae. International Journal of Aquaculture 3: 133 - 137. Phanna N., Nam S., Ramara O. and Pomeroy R. (2014). Experimental pond unit assessment in Cambodia. In: AquaFish Collaborative Research Support Program. Goetting, K., Price, C., and Egna, H., (eds). AquaFish CRSP Technical Reports: Investigations 2011 – 2013. AquaFish Innovation Lab, Oregon State University. Corvallis, Oregon. p 70 - 93. Pongthana N., N.H. Nguyen & R.W. Ponzoni (2010) Comparative performance of four red tilapia strains and their crosses in fresh-and saline water environments. Aquaculture 308, S109-S114. Popma T.J. and Lovshin L.L. (1995). Worldwide Prospects for Commercial Production of Tilapia. Auburn University, Alabama, USA. http://www.aces.edu/dept/fisheries/aquaculture/docs /worldtilapia.pdf.

Puycha K., Yuangsoi B., Charoenwattanasak S., Wongmaneeprateep S., Niamphithak P. and Wiriyapattanasub P. (2017). Effect of moringa (Moringa oleifera) leaf supplementation on growth performance and feed utilization of Bocourti's catfish (Pangasius bocourti). Agriculture and Natural Resources 51: 286 - 291. Quaas M., Hoffmann J., Kamin K., Kleemann L. and Schacht K. (2016). Fishing for proteins: How marine fisheries impact on global food security up to 2050 - A global prognosis. WWF Report. International WWF Centre for Marine Conservation, Hamburg, Germany. Rana K.J., Siriwardena S., Hasan, M.R. (2009). Impact of rising feed ingredient prices on aquafeeds and aquaculture production. FAO Fisheries and Aquaculture Technical Paper. No. 541. Rome, FAO. 63p. Rice M.A., Mmochi A.J., Zuberi L. and Savoie R.M. (2006). Aquaculture in Tanzania. World Aquaculture 37(4):50 - 57. Richter N., Siddhuraju P. and Becker K. (2003). Evaluation of nutritional quality of moringa (Moringa oleifera Lam.) leaves as an alternative protein source for Nile tilapia (Oreochromis niloticus L.). Aquaculture 217 (1-4): 599-611 Rothbard S., Z. Yaron and R. Moav (1988). Field experiments on growth enhancement of tilapia (Oreochmmis niloticus x O. aureus F1 hybrids) using pellets containing an androgen (17α- ethynyltestosterone) In: The Second International 95

Symposium on Tilapia in Aquaculture. ICLARM Conference Proceedings (eds R.S.V. Pullin, T. Bhukaswan, K. Tonguthai & J.L. Maclean), pp 367- 375. Department of Fisheries, Bangkok, Thailand and International Center for Living Aquatic Resources Management, Manila, Philippines. Sharma S.O.P. and Saini V.P. (2017). Replacement of fishmeal with soybean meal in Nile tilapia (Oreochromis niloticus) diet. Journal of Entomology and Zoology Studies 5(4): 845 - 849. Trewavas E. (1983). Tilapiine fishes of the genera Sarotherodon, Oreochromis and Danakilia. British Museum of Natural History, London, 583pp. Skelton P. (1993) A complete guide to the freshwater fishes of Southern Africa. Southern Book Publishers, Halfway House, 388 p. URT (2014). Fisheries annual statistics report - 2013. Ministry of Livestock and Fisheries Development, Fisheries Development Division, Dar es Salaam, Tanzania.60p. URT (2015). National fisheries policy of 2015. United Republic of Tanzania, Ministry of Livestock and Fisheries Development, Dar es Salaam, Tanzania.44p. URT (2016). The Tanzanian fisheries sector: Challenges and opportunities. United Republic of Tanzania, Ministry of Agriculture, Livestock and Fisheries, Dar es Salaam, Tanzania. Vera Cruz E.M. and G.C. Mair (1994). Conditions for effective androgen sex reversal in Oreochromis niloticus (L.). Aquaculture 122: 237 - 248. 96

Verster N. (2017). Comparison of growth rates of tilapia species (Oreochomis mossambicus and Oreochromis niloticus) raised in a biofloc and a standard recirculating aquaculture system (RAS). MSc. Dissertation, Ghent University, Belgium. Wahab Md.A., Rahaman S.M.B., Milstein A., Daniels H.V. and Borski R.J. (2014). Experimental pond unit assessment in Cambodia. In: AquaFish Collaborative Research Support Program. Goetting, K., Price, C., and Egna, H., (eds). AquaFish CRSP Technical Reports: Investigations 2011 – 2013. AquaFish Innovation Lab, Oregon State University. Corvallis, Oregon. p 1 - 37. Waidbacher H., Liti D.M., Fungomeli, M., Mbaluka R.K. Munguti J., and Straif M. (2006). Influence of pond fertilization and feeding rate on growth performance, economic returns and water quality in a small‐scale cage‐cum‐pond integrated system for production of Nile tilapia (Oreochromis niloticus L.). Aquaculture Research 37(6):594 - 600. DOI: 10.1111/j.1365-2109.2006.01467. x. Wohlfarth G.W. (1994). The unexploited potential of tilapia hybrids in aquaculture. Aquaculture Research 25: 781 - 788. World Bank (2006). Aquaculture: Changing the face of the waters. World Bank, Washington, DC, USA,136pp. World Bank (2013). FISH TO 2030: Prospects for Fisheries and Aquaculture. WORLD BANK REPORT NUMBER 83177-GLB. World Bank, Washington, D.C., USA. 97

WorldFish. (2011). Aquaculture, Fisheries, Poverty and Food Security. Working Paper 2011-65. The WorldFish Center, Penang, Malaysia. Yılmaz E., Ş. Çek and Y. Mazlum (2013). Effects of synthetic and natural steroids on the growth, sex reversal and gonadal development of rainbow trout, Oncorhynchus mykiss (Walbaum). Ege Journal of Fisheries and Aquatic Sciences 30: 123 - 131. Yue Y., Zou Z., Zhu J., Li D., Xiao W. and Han J. (2014). Dietary threonine requirement of juvenile Nile tilapia, Oreochromis niloticus. Aquaculture International 22:1457 - 1467.