TITLE XII

COLLABORATIVE RESEARCH SUPPORT PROGRAM

POND DYNAMICS/ AQUACULTURE

INTERIM WORK PLAN

1 SEPTEMBER 1995 to 31JULY1996

PRINTED MAY 1996

Pond Dynamics/Aquaculture CRSP

Office of International Research and Development

Oregon State University

Snell Hall 400

Corvallis, Oregon 97331-1641 USA

1. This work plan describes a standardized set of experiments to be undertaken by the

Collaborative Research Support Program in Pond Dynamics/Aquaculture during the

" period 1 September 1995 to 31JULY1996. Program activities are fended in part by

Grant No. DAN- 4023-G-00-0031-00 from the United States Agency for International

Development. Egypt activities are fended under Grant No. 263-0152-G-00-2231-00.

TITLE XII

COLLABORATIVE RESEARCH SUPPORT PROGRAM

POND DYNAMICS/AQUACULTURE

INTERIM WORK PLAN

1 SEPTEMBER 1995 to 31JULY1996

PRINTED MAY 1996

Pond Dynamics/ Aquaculture CRSP

Office of International Research and Development

Oregon State University

Snell Hall 400

Corvallis, Oregon 97331-1641 USA

This work plan describes a standardized set of experiments to be undertaken by the

Collaborative Research Support Program in Pond Dynamics/Aquaculture during the

period 1 September 1995 to 31JULY1996. Program activities are funded in part by

Grant No. DAN- 4023-G-00-0031-00 from the United States Agency for International

Development. Egypt activities are funded under Grant No. 263-0152-G-00-2231-00.

TABLE OF CONTENTS

Global Experiment: Effect of Pond Management Strategy on Nutrient Budgets ............................... 1

Honduras Project lntroduction .................................................................................................................... 3 Study 1: Effects of Diet Protein on Food Conversion and Nitrogen Discharge

During Semi-Intensive Production of Penaeus vannamei... .......................................... .4 Study 2: Estuarine Water Quality .................................................................................................... 6 Study3: Tidal Effects on Nutrient, Oxygen, Temperature, and Salinity Profiles in

Estuaries of Two Major Shrimp Producing Areas in Southern Honduras ................ 7 Study 4: Sex Reversal of Tilapia 17-Alpha Methyltestosterone Dose Rate by

Environment, and Efficacy of Bull Testes ....................................................................... 8

African Project Study 1: African Site Evaluation and Development Planning ................................................. 10 Study 2: Masculinization of Tilapia ............................................................................................. 12 Study3: Experimental Evaluation of Lime Requirement Estimators for

Global Sites Extension of WP7 Study B ....................................................................... 13 Study4: Characterization of Soils from Potential New Sites ................................................... 15 Study5: Relationships between Concentrations of Chemically-Extracted Soil

Nutrients and Equilibrium Concentrations of Nutrients in Soil-Water Microcosms ...................................................................................................................... 16

Study6: Reproductive Efficiency of Nile Tilapia and "Red" Tilapia and Fry Growth and Response to Sex Reversal ........................................................................ 17

Study 7: Gonadal Differentiation in Nile Tilapia Fry as a Function of Temperature ............ 9 Study8: The Effect of Feed Storage Time and Storage Temperature on

Growth Rate of Tilapia Fry and Efficacy of Sex Reversal Treatment ..................... 21 Study9: Feed Consumption, Growth, Conversion Efficiency of Tilapia Fry and

Efficacy of Sex Reversal Treatment as a Function of Water Temperature .............. 23 Study 10: Economics ......................................................................................................................... 25

Thailand Project Introduction .................................................................................................................. 26 Activity 1: Net Photosynthetic Oxygen Production and Fish Production During

PD/ A CRSP Pond Experiments .................................................................................... 27 Activity 2: Relationships among Measures of Suspended Particulate Material

in the CRSP Data Base .................................................................................................... 28 Activity 3: Polyculture in Deep Ponds ............................................................................................ 29 Activity 4: Finishing System for Large Tilapia: A Continuation, Part 3, of WP7 Study 3 ....... 30

Philippines Project Introduction ............................................................................................................. 31 Activity 1: Stocking Density in Relation to Elevation in the Philippines ................................... 32 Activity 2: Yield Trials with Genetically Selected Tilapia ............................................................ 33 Activity 3: Philippines Outreach ...................................................................................................... 34

Database Management. ............................................................................................................................. . 35

Data Analysis and Synthesis Team (DAST) Introduction .................................................................. 36 Study 1: Development and Use of POND as a Tool for Analysis and Planning ................... 37 Study 2: Decision Support for Pond Production Systems in the United States ..................... 39 Study 3: Application of POND for Decision Support at Other International Sites .............. .40 Study 4: A Water Quality /Fish Yield Model with Stochastic Inputs .................................... .41 Study 5: Aquaculture Pond Modeling for the Analysis of Integrated Aquaculture/

Agriculture Systems ........................................................................................................ 43

Social Science Project: Socioeconomic Dimensions of Aquaculture Development: A Comparative Assessment of Financial Returns, Adoption Barriers, and Impacts of Tilapia Production Regimes ..................................................................................................................................... 44

GLOBAL EXPERIMENT: EFFECT OF POND MANAGEMENT STRATEGY ON NUTRIENT BUDGETS

Objectives: 1. To develop nutrient budgets for nitrogen and phosphorus for semi-intensively managed freshwater and brackishwater production ponds. 2. To quantify effect of pond management strategy on water quality and sediment quality.

Significance: PD I A CRSP pond management strategies rely on high nutrient loading rates to increase pond productivity. Discharge of nutrient-rich pond water may result in deteriorated quality of receiving waters. However, there is little information on the effect of semi-intensive pond management strategies on quality of pond effluents.

Nutrient budgets are used to assess the fate of nutrients added to an ecosystem and indicate .the relative importance of each nutrient source and sink. Development of nutrient budgets would permit quantification of potential pollutional impact of a specific pond management strategy.

Experimental Design: Nutrient budgets will be developed for freshwater and brackishwater ponds. Two treatments, each replicated three times, will be tested in each environment. Freshwater treatments: (1) chemical fertilization plus commercial formulated fish ration (min. 20% crude protein) added beginning day 80, and (2) chemical fertilization followed by commercial formulated fish ration (min 20% crude protein) added beginning day 80. Brackishwater treatments: (1) commercial formulated shrimp ration (20% crude protein) and (2) commercial formulated shrimp ration (30% crude protein).

Stocking Rate: Freshwater environment: male (sex reversed) tilapia (Oreochromis niloticus) stocked at 3 fish/m2; a predator fish may be stocked in accordance with normal host country management practices. Brackishwater environment: post-larval penaeid shrimp (Penaeus vannamei) will be stocked to achieve a final stocking rate of 7.5 PLs/m2.

Nutrient Inputs: Freshwater environment: chemical fertilizer (urea and DAP) applied at 28 kg N/ha/wk at 4N:1P; and, commercial pelleted (extruded) fish ration of 20% crude protein minimum. Feed will be offered beginning on day 80 of the culture cycle. Feed rate will be determined for each pond as the amount of feed consumed during a 1-h period each in the morning (1000-1100 h) and the afternoon (1400-1500 h). The average for each treatment will be fed twice daily to all ponds within a treatment. Feeding rate will be adjusted weekly according to the above procedure. Brackishwater environment: commercial pelleted shrimp ration of 20% and 30% crude protein. Each feed will be offered at 50% of the theoretical feeding curve for P. vannamei: log 10 (% biomass fed) = -0.899 - 0.561 log 10 (mean wt of shrimp).

Data Collection: Water Budget: Pond volume will be determined. Pond water levels will be maintained 5 cm below drain tops to provide storage capacity for rainfall. Replacement water will be added to freshwater ponds weekly, if necessary, to replace losses to seepage and evaporation. In brackishwater ponds, 10% of the pond volume will be replaced weekly; pond water will be discharged first, the drain closed and replacement water added. Emergency water exchange (5% of pond volume) will be used to mitigate decrease in dissolved oxygen concentrations to critical minimums. Stage gages installed in each pond will be read daily to determine change

1

in storage. Regulated inflow water will be quantified from change in stage. Evaporation and rainfall will be measured.

Nutrients: Samples of feed and fertilizer added to ponds will be collected at monthly intervals and analyzed for total nitrogen and phosphorus. Five animals per pond (per species in the event a polyculture is used) will be sampled at stocking and at harvest for total nitrogen and total phosphorus analyses.

Water Quality: Total nitrogen, nitrate, nitrite, total ammonia and pH will be determined weekly. Total phosphorus and soluble reactive phosphorus will be determined every two weeks. Chlorophyll a and total alkalinity will be determined every two weeks. Samples of water entering and leaving the pond will be analyzed during any water exchange. During harvest, water samples will be collected and 100%, 10% and 0% of pond water volume for analysis. BODS will be determined on pond effluent.

Sediment Quality: Sediments will be sampled using a core sampler. Five or more cores will be collected from one site near the center of each pond. The top 5 cm of each core will be collected, pooled, dried at 100°C to constant with, and analyzed for total nitrogen and total phosphorus. Samples will be collected immediately following pond inundation and immediately prior to pond harvest.

Stock/Harvest: Standard PD I A CRSP protocol for data collection on animals stocked/harvested will be followed. Fish will be harvested after 240 days and shrimp after about 120 days.

Null Hypothesis: Nutrient budgets are not affected by pond management strategy.

Report Due: April 1996

2

HONDURAS PROJECT

Cooperating Institutions:

Auburn University Dr. Claude E. Boyd Dr. David Teichert-Coddington Dr. Bartholomew W. Green Dr. Bryan L. Duncan

General Directorate of Fisheries and Aquaculture, Ministry of Natural Resources Dr. Marco Polo Micheletti.

Introduction: The Honduras research effort will continue work on the brackish water environment in southern Honduras that was initiated as part of the Seventh Work Plan. Investigations will target selected shrimp farms and associated estuaries which serve as water sources and sinks. Data collection to quantify impact of shrimp farming activities on estauarine water quality will be initiated as part of this work plan; quantification of impact will be relative to the baseline estaurine water quality profile that was established as part of Seventh Work Plan activities. Specific farm resource management issues also will be addressed through replicated pond studies.

Interaction between shrimp farming and estuarine ecology is critical to the sustainability of the nascent shrimp industry Gulf of Fonseca, Central America. Experience in southeast Asia and China demonstrates that shrimp farming can become untenable if the carrying capacity of shrimp-farming estuaries is exceeded. Incidence of disease increases and production decreases as estuarine water quality deteriorates. The Gulf of Fonseca is the common water source for shrimp farms in Honduras, Nicaragua and, to some extent, El Salvador. The gulf is shallow and is reported to exchange its total volume of water with the Pacific Ocean about once every seven years. Therefore, the potential for deterioration of gulf water quality also exists. Every shrimp farming area is different because of estuarine size, estuarine exchange rates with the ocean or bays, freshwater inflow, and the amount and type of land suitable for shrimp farming in a particular area. Development of shrimp farming therefore must be accompanied by objective studies that make information available on environmental impact of farming to permit industry or government regulation of industry development. The industry also can be encouraged to implement management techniques that will minimize deterioration of its water supplies.

Honduran shrimp farms are located on several different estuaries of the Gulf of Fonseca where water quality varies widely in a relatively small geographic area. Estuaries comprising the confluence of the Choluteca and Negro rivers and the Gulf are muddy and carry urban and agricultural effluents. Salinity fluctuates widely during the year, ranging from 5 ppt or less during the rainy season to in excess of 45 ppt during the dry season. Appearance of water changes with onset of rains. Estuaries formed from inlets of the Gulf appear relatively clear and unpolluted, and apparently have more stable water quality than riverine estuaries. Study of the interactions among shrimp farming, estuarine water quality and climate would yield practical results for the shrimp producer and environmental manager, and information for the systems dynamics modeler. This information would have world-wide applicability.

Replicate pond studies will focus on increasing production efficiency and reducing nutrient effluents. Efficient use of feed directly impacts environmental quality and farm profitability. Over-feeding or unnecessary use of high-protein feeds results in wasteful nutrient discharges.

3

HONDURAS PROJECT STUDY 1: EFFECTS OF DIET PROTEIN ON FOOD CONVERSION AND NITROGEN DISCHARGE DURING SEMI-INTENSIVE PRODUCTION OF PENAEUS VANNAMEI

Objective: Test the effect of diet protein level on food conversion and nitrogen during the production of Penaeus vannamei at semi-intensive stocking levels during the warm (wet) season and cool (dry) season.

Significance: Previous studies demonstrated that shrimp production is similar at protein levels between 20% and 40%, when shrimp are stocked at densities between 5 and 11/m2. Feeding rates in these studies were such that feed efficiency was relatively low. Another trial conducted in Choluteca with shrimp stocked at 7.5/m2 demonstrated that production during the dry season was not significantly affected by a 50% reduction in feeding rate. Wet season production was impacted significantly by the 50% reduction in feeding, but feed efficiency was improved. These results indicated that too much feed was applied during the dry season, and that wet season feeding rates might be reduced, but not by half. It is possible that a comparatively high protein diet might improve shrimp growth and feed conversion, if used at a feeding rate that is low compared with that employed with lower protein diets.

Pond chemical budgets indicate that nitrogen discharge increases with both feeding rate and diet protein level. Primary productivity in estuaries appears to be limited by nitrogen. It is therefore economically and ecologically important to feed at appropriate rates with an appropriate protein level.

Null Hypothesis: 1) Shrimp growth, yield and feed conversion during each season will be independent of feeding rate and dietary protein level. 2) Nitrogen discharge from shrimp ponds during each season will be independent of shrimp feeding rate and dietary protein level.

Experimental Design: A completely randomized design in 2x2 factorial arrangement will be used to test two feeding rates each at two levels of protein.

1. Rate= 75% of feeding curve; protein= 20% 2. Rate= 75% of feeding curve; protein= 30% 3. Rate= 50% of feeding curve; protein= 20% 4. Rate= 50% of feeding curve; protein= 30%

Each treatment will be replicated at least 3 times. Feeding curve is described by the relationship: Y = lQA(-0.899 - 0.561Log10(X)), where Y= feed rate as a percent of shrimp biomass and X= mean individual shrimp weight.

Stocking Rate: Penaeus vannamei will be stocked at 20/m2(200,000/ha). A survival rate of 50% will be assumed because of Taura Syndrome. Mortality is assumed to decrease logarithmically with time. Shrimp growth will be sampled monthly.

Feeding: Shrimp smaller than 1 g will be fed at 13% of biomass. Daily ration will be divided into two meals.

4

Water Quality: Water samples will be taken at the entrances and exits of ponds (water actually entering and leaving ponds).

Daily: Temperature and dissolved oxygen Weekly: Salinity, Secchi disk, total nitrogen, nitrate, total ammonia, BOD2, pH Bi-weekly: Total phosphorus, filterable phosphorus Three times per cycle: Total alkalinity, primary productivity, pH and temperature (taken in the field at 0600, 1000, 1430, 1800, and 0600 hours). Harvest: total settleable solids and complete chemical analyses of water at 100%, 10%, and 0% of pond volume.

Statistical Methods: ANOVA

Schedule: Warm season study data collection, June to October 1995. Cool season study data collection, November 1995 to March 1996.

5

HONDURAS PROJECT STUDY 2: ESTUARINE WATER QUALITY

Objective: To assess impact of shrimp farm effluent on water quality of estuaries.

Significance: Frequent analysis of estuarine water quality over the past two years has resulted in the establishment of a baseline of information on selected chemical, biological and physical characteristics of water at points along major estuaries that supply water and receive effluents from shrimp farms. It now is possible to initiate research to assess the environmental impact of shrimp farming on the estuaries in southern Honduras. This data will provide a means to classify estuaries by type, level of pollution and potential for degradation. These data may help explain observed yield differences among farms in the area.

Experimental Design: Water samples will be collected every two weeks from 12 samplings sites distributed over six estuaries in the shrimp farming area of southern Honduras. An additional sample site, which serves as a reference point, is located on the Choluteca River upstream from any discharge from the city of Choluteca. Samples of estuarine influent will be collected at the pumping station at about the middle of the daily pumping cycle (high tide).

Variables Measured: Total settleable solids, total nitrogen, total ammonia nitrogen, total phosphorus, chlorophyll a salinity, temperature, and dissolved oxygen. River flow, rainfall and evaporation data will be obtained from the Directorate of Hydrology, Ministry of Natural Resources, Honduras.

Statistical Methods: Stratification of estuaries by type; monthly averages by location and estuarine type.

Schedule: Data collection, April 1995 - April 1996.

6

HONDURAS PROJECT STUDY 3: TIDAL EFFECTS ON NUTRIENT, OXYGEN, TEMPERATURE, AND SALINITY PROFILES IN ESTUARIES OF TWO MAJOR SHRIMP PRODUCING AREAS IN SOUTHERN HONDURAS

Objective: Determine effects of daily tidal fluctuation during a lunar cycle on nutrient, oxygen, temperature and salinity profiles in estuaries of two major shrimp producing areas.

Significance: Carrying capacity of an estuary depends on tidal fluctuations and water exchange with the Gulf of Fonseca. The water quality of the major estuaries in southern Honduras has been characterized, but water exchange rates and effect of depth on water quality has not been quantified well. Results of this study would support efforts to determine the dynamics of estuaries in major shrimp farming areas and the impact of shrimp farming on these estuaries.

Experimental Design: El Pedregal and San Bernardo estuaries will be sampled weekly during a 5 week period that spans a lunar cycle. Water column profiles at 50-cm intervals will be taken for dissolved oxygen, temperature and salinity at five locations along the lenth of the El Pedregal estuary and at 6 to 8 locations along the San Bernardo estuary. Profiles will be measured during low and high tide on each sample date. Water samples also will be collected from surface, mid-depth and bottom at three locations during low and high tides. Water samples will be analyzed for BODS, total phosphorus, total nitrogen, chlorophyll a, and total suspended solids. Inorganic forms of nitrogen phosphorus will be determined if time permits.

Schedule: Data collection, September-October 1995

7

HONDURAS PROJECT STUDY 4: SEX REVERSAL OF TILAPIA: 17 ALPHA METHYL TESTOSTERONE DOSE RATE BY ENVIRONMENT, AND EFFICACY OF BULL TESTES

Objective: The objectives of this research were to determine the efficacy of sex reversal of different dosage rates of MT to fish treated in different environments, and to evaluate the potential of potential of freeze-dried bull testes as a dietary source of testosterone for tilapia sex reversal.

Significance: Sex reversal of newly hatched tilapia generally is accomplished via oral administration of 17-alpha methyltestosterone (MT), which has been incorporated into a starter fish feed at 60 mg MT /kg feed. While use of the 60 mg MT /kg feed dose consistently yields populations comprised of less than 5% females (i.e., >95% males), this has not been shown to be the optimal dose. Other investigators have reported sex reversal of tilapia at dose rates less than 60 mg MT /kg feed. However, results form some of these studies are inconsistent, and it is difficult to separate treatment environment effects. Thus, it is necessary to identify the optimal dose of MT for consistent, successful sex reversal in a variety of treatment environments.

Naturally occurring sources of testosterone may be an alternative to using a synthetic androgen, which also is an anabolic steroid, for tilapia sex reversal. Successful tilapia sex reversal using ram's testes as a source of dietary testosterone has been reported. Bull testes are a by-product of beef industry in the US, and are a potential source of dietary testosterone for tilapia sex reversal.

Null Hypotheses: 1) Sex reversal of tilapia is not affected by dietary 17-a methyltestosterone level. 2)Inclusion of freeze-dried bull testes in diet does not induce sex reversal of tilapia fry.

Experimental Design: A completely randomized design in 2x5 factorial arrangement will be used to test five rates of dietary MT (0,15,30,45,60 mg/MT feed) each in two different treatment environments (indoor aquaria, outdoor tanks). Each treatment will be replicated 3 times. Aquaria (80 L) will be set up inside the hatchery building, and hapas (45 L water volume) will be suspended in outdoor 20-m3 concrete tanks.

In addition, incorporation of two levels of freeze-dried bull testes (25% or 50% of diet) into the fish diet will be tested for efficacy of sex reversal of tilapia fry in indoor aquaria.

Stocking Rate: Nile tilapia (Oreochromis niloticus) fry will be stocked at 8 fry /L.

Feeding: 42% protein trout chow will be used as the carrier for MT. Fry in each treatment will be fed at 20% body weight during week 1; the daily ration will be divided into four meals. Feed rate will be decreased by 2.5%/wk during weeks 2-4. Feed rate will be adjusted weekly based on results of weekly population samples.

Frozen bull testes will be obtained from a meat packing plant in Montgomery, AL. Individual testes will be skinned, sliced, freeze-dried and ground, and mixed with trout chow either in a 1:1 or 1:3 freeze-dried testes: trout chow ratio. Mixed feed will be refrigerated until fed and feed will be offered as described above.

Water Quality: NI A

8

Statistical Methods: ANOVA

Schedule: Data collection, August to December 1995; data analysis and report preparation, January to February 1996.

9

AFRICA PROJECT STUDY 1: AFRICAN SITE EVALUATION AND DEVELOPMENT PLANNING

Cooperating Institutions:

Oregon State University, Lead Institution Wayne Seim James Bowman

Auburn University TomPopma Karen Veverka

Introduction: On-site research in Rwanda was interrupted by war and civil unrest beginning April 6, 1994. The establishment of a new government in Rwanda and continued lack of security has prompted the relocation or revision of planned activities under the CRSP. In any event, the inclusion of new sites into the CRSP poses new opportunities to include new host country scientists and to include selected environmental characteristics advantageous to CRSP objectives. Activities under Work Plan 7 resulted in the development of site evaluation criteria and the gathering of background data on several potential sites. The Sagana Fish Culture Farm in Kenya was visited and contacts with Kenyan and Belgian operatives were established. Some initial on­site information was collected. Contacts with another potential site in Zimbabwe were also established, while difficulty in attaining Mission approval for visits to Malawi was encountered. These activities form a good basis for further investigation of potential CRSP sites in Africa and for constructing development plans and costs for likely sites. Problems uncovered at the Kenya site need investigation, while little is known of potential Zimbabwe sites. Other possible African sites not uncovered in earlier efforts may warrant investigation. The activities under the proposed Work Plan would concentrate specifically on the construction of evaluations and proposed development plans and estimated costs for potential African sites. These would be make available to the ME for use in the decision process. Activities under this Work Plan would contribute to the ME-administered process of selecting new sites for inclusion in the new proposal.

This study would be accomplished by both OSU and Auburn personnel with OSU as the lead. Karen Veverka would lead the development planning and costs area; Jim Bowman would coordinate site data collection and organize data as its accumulates; Tom Popma and Wayne Seim would be involved in site evaluation and final report preparation.

Objective: To provide the ME with evaluations including development plans and costs for potential CRSP African sites. Site gee-climatic data such as temperature range, soil characteristics, and water quality would be obtained, along with information relating to each site criterion category. A time-table for delivery of this information to the TC and ME will be developed in conjunction with the ME director. A secondary objective would be the timely reporting of information gathered on site criteria following any evaluation travel to keep CRSP members involved in the process.

Significance: Continuing safety and security factors prevent the return of the CRSP to Rwanda and point out the necessity of careful planning in site selection. The programmatic value to the overall CRSP of the research conducted at the selected site may reside in its associated geo-climatic and socioeconomic characteristics. Constraints to efficient conduct of research would be known

10

before substantial investment in site development began. Potential for establishing new operational ties with in-country entities would be identified. Recent communication with the Belgian (KUL) researchers operating in Kenya and other African nations is an example. Recognition of potential costs and other requirements of initiating a new site are necessary to allow satisfactory planning and budget allocation by the CRSP.

Experimental Design: Data will be collected to provide information according to the criteria for site selection developed under Work Plan 7 and discussed in the draft continuation proposal. In order to facilitate regular communication dissemination within the CRSP, trip reports by evaluators would include information related to selection criteria, with discussion of areas for further investigation or consideration. For sites likely to be evaluated by the CRSP, a plan for establishing a CRSP operational presence at the site will be developed, along with the estimated costs.

A draft proposal would be developed at a workshop for that purpose attended by the study operatives plus other representatives as determined by the ME.

Laboratory Facilities: NI A. Activities will be conducted from both OSU and Auburn University, along with travel to potential sites in Africa.

Culture Period: NI A

Stocking Density: NI A

Nutrient Inputs: NI A

Water Management: NI A

Sampling Schedule: Travel would be completed by December, 1995. Travel to Kenya would occur in September to synchronize with Belgian planning. Timing of travel to Zimbabwe or other sites would await arrangements with operatives in those countries. Completion of evaluation and development planning tasks would be coordinated with TC chair and ME director but a report of results would be available by the Annual Meeting at the latest. If additional African sites are identified beyond Kenya and Zimbabwe that require travel beyond that budgeted for in this study, the ME would be requested to undertake or fund that additional travel.

Schedule: Study 1: May 1-April 30 (preliminary report complete Jan 20, 1996)

11

AFRICA PROJECT STUDY 2: MASCULINIZATION OF TILAPIA

THROUGH IMMERSION IN 17a-METHYLTESTOSTERONE OR

17a-METHYLDIHYDROTESTOSTERONE

Cooperating Institution:

Oregon State University Martin Fitzpatrick, Wayne Seim, Carl Schreck, and Bill Gale

Introduction: The biotechnology component of the Work Plan for the CRSP Egypt project included the goal of maximizing the efficiency of the procedures used for inverting the sex of tilapia. We have conducted experiments to determine the capacity of immersion treatment in synthetic androgens to masculinize tilapia. One experiment indicated that two hour immersion of tilapia fry in 17a-methyltestosterone or 17a-methyldihydrotestosterone twice during early development resulted in significant masculinization in comparison to control fish. Since these results were extremely promising, the experiment needs replication.

Objective: To determine if immersion in 17a-methyltestosterone or 17a-methyldihydrotestosterone masculinizes tilapia with similar efficacy to dietary treatment with 17a-methyltestosterone.

Significance: Development of teclmiques for masculinization through immersion in steroid-containing solutions may provide aquaculturists with a safe and cost effective alternative to treating fry with food that contains steroid.

Null Hypothesis: The sex ratio of tilapia fry immersed in solutions that contain 17a-methyltestosterone or 17a­methyldihydrotestosterone will not differ from that of control fry.

Experimental Design: Immersion of newly hatched gynogenetic fry in 17 a-methyltestosterone or 17 a­methyldihydrotestosterone will be compared with feeding of this steroid for production of males. Immersion doses will be 500 and 100 µg/Liter of water (immersion will be for two hours twice during early development), feeding concentration will be 40 mg/kg of food for six weeks beginning at the onset of feeding; controls will be animals immersed in solutions containing the ethanol vehicle only and unimmersed animals. Optimum treatment will be that which allows maximum production of males.

Laboratory Facilities: Oregon State University--two aquaria containing a total of two males and six females for production of eggs, seven three-liter jars for hormone treatment of 50-100 fry each, and two flow-through troughs with partitions for raising offspring.

Culture Period: 120 days for offspring.

Statistical Methods: Chi-square for sex ratio.

Schedule: Data collection: 5 /95-9 /95; technical report 4/96.

12

AFRICA PROJECT STUDY 3: EXPERIMENTAL EVALUATION OF LIME REQUIREMENT ESTIMATORS FOR GLOBAL SITES: EXTENSION OF WP7 STUDYB

Cooperating Institution:

Oregon State University Jim Bowman Wayne Seim

Introduction: Study Bin Work Plan 7 examined under laboratory conditions, lime requirement estimators including PONDCLASS and three other "standard" methods. Reponse of the pond to the addition of lime at the estimated dose was simulated in 800 ml microcosms and the resulting water alkalinity monitored over 28-day experiments. This experiment would be extended to include soils from Rwanda (already collected) and additional soils from Honduras and Africa.

In an extension of this study, an investigation of the use of artificial enclosures ("isolation columns") as in-pond test units will be conducted. These enclosures will be used to test a selected liming rate in situ and compare the results with those obtained in the laboratory microcosms. The successful development and use of a structure that isolates a column of water together with the underlying portion of the pond bottom may be useful for future CRSP soil-water investigations, particularly in situations where additional replicates within individual ponds are desirable.

Objectives: 1. To evaluate several lime requirement estimators as outlined in WP 7, Study B, using additional soils. 2. To use isolation columns to compare the results of laboratory lime requirement experiments in microcosms with results in a larger and more complex model of the whole pond ecosystem.

Significance: Additional examination of lime requirements of international soils using both POND and standard estimators, followed by monitoring the water alkalinity response will provide additional basis for evaluation of POND. Proper liming methods contribute to sufficient carbon resources for alga production and reduce potential phosphorus losses to acidic soils. Pond enclosures provide an additional evaluation of liming response. Such enclosures may prove useful in future research to maximize replicates for limited ponds.

Null Hypothesis: 1) The lime requirement estimates of all tested methods provide equivalent results in terms of total alkalinity in the water column. 2) The lime requirement estimate of each method is equally effective for all soil types. 3) The results of lime applications in laboratory microcosms are the same as those in pond enclosures in terms of total alkalinity response of the water column.

Experimental Design: The laboratory microcosm studies will be conducted according to the same design as outlined in WP 7, Study B. For the isolation column experiment, six columns (30 cm in diameter) will be placed in a single pond (alkalinity <20 mg/l CaC03 ), with the columns reaching from the acid soil (pH < 6.5) approximately 1 meter up to the water surface. Two treatments will be used; a control treatment in which no lime is applied, and a single liming treatment in which lime is applied at a level estimated by the Pillai and Boyd method. There will be three replicates of

13

each treatment. A parallel treatment and sampling protocol will be followed using the same soil and water in microcosms in the laboratory. After 28 days, the pH of the soil in each column (pond) and microcosm (lab) will be measured.

Facilities: The laboratory studies will continue to be conducted at the Oak Creek Laboratory,· Department of Fisheries and Wildlife, and Oregon State University. The isolation column studies will be conducted at the Soap Creek Ponds, OSU Department of Fisheries and Wildlife.

Culture Period: NI A

Stocking Density: NI A

Nutrient Inputs: NI A

Water Management: No water additions other than rainfall will be made to the ponds, pond enclosures, or laboratory microcosms during the course of the experiment.

Sampling Schedule: Pond bottom samples will be taken from areas near the isolation columns for characterization including particle-size analysis, pH, %0.M., CBC, base saturation, acidity, for lime requirement determination and for use in the laboratory portion of the experiment. Pond water will be sampled for pH and alkalinity prior to the beginning of the experiment. Water samples will be taken from the columns for pH and alkalinity determinations at intervals of 1,3,7,14,21, and 28 days following start-up. A parallel treatment and sampling protocol will be followed using the same soil and water in laboratory microcosms. After 14 days, the pH of the soil in each column (pond) and microcosm (lab) will be measured. Data collection: June 6- July 4; July 11- August 15

Statistical Methods: Multiple comparison methods using the Student's Neman-Keules and Duncan's tests. Regression correlation tests by TableCurve curve-fitted software.

Schedule: The experimental part of these studies will be completed by August, 1995. Report complete: Jan 1

14

AFRICA PROJECT STUDY 4: CHARACTERIZATION OF SOILS FROM POTENTIAL NEW SITES

Cooperating Institution:

Oregon State University Jim Bowman Wayne Seim

Introduction: Study C of Work Plan 7 included the characterization of African soils in its objectives, and soils from Sagana Station and from Mombasa, Kenya were characterized. This study would provide soil characterization data for soils from Zimbabwe or other African sites evaluated and extended the characterization to additional Kenyan soils. Soils from other proposed sites would also be characterized in cooperation with those evaluating those sites.

Objective: To characterize soils at potential new sites to contribute quantitative description to new site evaluation activities.

Significance: Soil characteristics are significant factors in the success of pond culture of fish and directly impact on-site research efforts. Soils are determinants of such factors as pond water retention, lime requirement, turgidity and nutrient dynamics. These results would contribute to new site evaluation activities by providing quantitative descriptions.

Experimental Design: For each new site evaluated including Sagana Station, at least two samples each with a combined weight of at least 2 kg (dry weight) would be collected. Each sample would be composed of at least 3 individual sample cores. Sample cores would be 15 cm deep. Soils would be characterized for: lime requirement, pH, particle size, CBC, organic matter, acidity, exchangeable bases, base saturation and general type classification.

Laboratory Facilities: Soil samples will be analyzed either at the Oak Creek Laboratory (OSU Department of Fisheries and Wildlife) or the Central Analytical Laboratory (OSU Department of Crop and Soil Science).

Culture Period: NI A

Stocking Density: NI A

Nutrient Inputs: NI A

Water Management: NI A

Sampling Schedule: Samples would be collected during site visits conducted under Study 1 above. Characterization of soils will be completed by the end of 1995, for consideration along with other criteria in the evaluation of potential new sites.

Statistical Methods: NI A

Schedule: Data collection: Sept 1- Dec 31, 1995i report complete: April 30, 1996

15

AFRICA PROJECT STUDY 5: RELATIONSHIPS BETWEEN CONCENTRATIONS OF CHEMICALLY-EXTRACTED SOIL NUTRIENTS AND EQUILIBRIUM CONCENTRATIONS OF NUTRIENTS IN SOIL­WATER MICROCOSMS.

Cooperating Institution: Auburn University

Objectives: To determine if nutrient concentration in dilute acid and ammonium acetate extracts of soils can be used to predict equilibrium concentrations of nutrients in soil-water microcosms.

Significance: Pond soils often are extracted with dilute acid or ammonium acetate to remove mineral nutrients for analysis. These extractants were developed for agricultural soils, and they indicate amounts of plant-available nutrients in soils. In ponds we are interested in the equilibrium concentrations of mineral nutrients that are attained in water in contact with bottom soil. We do not have data to show that these two extractants remove nutrient fractions that are strongly correlated with the water soluble fractions. This study will reveal if the two exractants can be used to accurately predict the equilibrium concentrations of nutrients in soil­water microcosms. It also will reveal if one of the extractants is better than the other in this regard. Of course, if neither extract provides reliable estimates, pond soil samples should be extracted with water.

Null Hypotheses: The two exractants remove the same concentrations of nutrients and these concentrations are not related to equilibrium concentrations in soil-water microcosms.

Experimental Design: A series of 73 pond soil samples collected from locations in Thailand in 1994 will be used in this study. Soil samples have been air-dried and pulverized to pass a 20-mesh screen. Subsamples will be exracted with both dilute acid (0.025 N H2 S04 + 0.05 N HCI) and neutral, 1 N ammonium acetate. Extracts will be analyzed for phosphorus, calcium, magnesium, sodium, potassium, iron, zinc, copper, manganese, and boron by plasma-induced spectrophotometry (ICAP). Samples also will be analyzed for pH, particle-size distribution, cation exchange capacity, and base saturation. Soil-water microcosms will consist of 5-g soil in 100 ml distilled water. After agitation for 1-week on a rotating shaker, the water will be removed, filtered, and analyzed for mineral nutrients by ICAP, and for total alkalinity, total hardness and pH by routing procedures. Data will be analyzed by regression analyses to determine if soil variables are correlated with concentration of nutrients, alkalinity, and hardness in soil-water microcosms.

Pond Facilities: Soil samples have already been collected from ponds in Thailand. Laboratory studies will be conducted at an AU water quality laboratory.

Culture Period: NI A

Fish Stocking Rate: N/ A

16

AFRICA PROJECT STUDY 6: REPRODUCTIVE EFFICIENCY OF NILE TILAPIA AND "RED" TILAPIA AND FRY GROWTH AND RESPONSE TO SEX REVERSAL

Cooperating Institution:

Auburn University Ronald Phelps Edwin Smith

Objectives: Compare the Nile tilapia (0. niloticus) and a "Red" tilapia (Oreochromis spp.) to determine differences in 1) reproductive efficiency of adults and 2) growth and response of fry to sex reversal treatments of two, three, and four weeks duration. (Note: the "red" tilapia in this case will be further identified by electrophoresis).

Significance: In many countries shortage of tilapia fingerlings, especially monosex fingerlings, is a major constraint to wider adoption of aquaculture technologies. 0. niloticus, the common test animal across all PD I A CRSP sites, is preferred in some regions, but strains of the 'red" tilapia are preferred in other markets. The reproductive efficiency and efficacy of sex reversal has been widely demonstrated for both fish, but there have been few controlled comparisons of the two. Production of all-male tilapia by sex reversal is a two-stage process: fry production and hormone treatment of the fry. This study will evaluate biological and economic aspects of both stages.

Null Hypothesis: The reproductive efficiency of the Nile tilapia and the red tilapia are similar in open ponds; growth and response to sex reversal of fry of both groups are similar at all treatment duration's.

Experimental Design: The study will be implemented in two stages. Stage 1 will compare the reproductive efficiency of both fish in open ponds; ponds will be drained two to three weeks after stocking broodfish; fry will be collected and the number and size distribution of fry will be determined; both fish groups will be tested concurrently, with two replicates per species. Each fry production trial will be repeated three times. In Stage 2 fry from each test group will be stocked at a single density in nylon hapas and hormone-treated feed will be administered for two, three, or four weeks, with four replicates per treatment. A control feed (with no MT) will also be given to each fish group for four weeks. Growth rate of fry will be measured weekly during treatment. After suspending hormone treatment, fry will be reared to at least 5 g and preserved for later microscopic examination to determine phenotypic sex.

Facilities: Research will be conducted at El Carao Fish Culture Station in Honduras. At least four earthen ponds will be used to study reproductive efficiency. At least 24 fine mesh nylon hapas will be used during the hormone treatment phase of the study. Microscopic examination of gonads to determine phenotypic sex by microscopic examination.

Culture Period: Fry production cycles (from stocking to fry harvest) will be approximately 20 days. The hormone treatment phase will then begin on day 20 and continue for two, three, or four weeks. Rearing after sex reversal will be four to six weeks to a minimum size of 5 g to permit determination of phenotypic sex will be done by microscopic examination.

17

Stocking Rate: Broodfish will be stocked at a rate of 2000 kg/ha, 2 females per male. For sex reversal, fry with TL less than 11 mm will be stocked at a density of approximately 4000/m2.

Nutrient Inputs: Broodfish in fry production ponds will be fed daily with commercial fish feed at a rate of 1.5% BW. During sex reversal fry will be fed with commercial sex reversal feed at· the rate prescribed in the protocol of INAD 8479 (attached).

Water Management: All ponds will receive only sufficient water to replace evaporation and seepage.

Sampling Schedule: Maximum and minimum water temperature and dissolved oxygen will be measured daily in all ponds. Secchi disk will be measured at least weekly. Broodfish will be counted and weighed at the beginning and end of each trial. For sex reversal, fry will be measured and counted at the beginning and end of treatment.

Statistical Methods: Factorial ANOVA, regression analysis.

Schedule: This study was initiated in January 95 but will continue to August 95.

Logistic Considerations: Research will be conducted by a US RA, currently a MS candidate at Auburn University and/ or by the Rwandan RA (providing he gains entry to the US). Transport and housing in Honduras present no serious constraint to this study or to El Carao station management.

18

AFRICA PROJECT STUDY 7: GONADAL DIFFERENTIATION IN NILE TILAPIA FRY AS A FUNCTION OF TEMPERATURE.

Cooperating Institution:

Auburn University TomPopma Joyce Newman

Objectives: Time of gonadal differentiation will be determined in Nile tilapia fry grown at a range of temperatures commonly found in sex-reverse culture situations. A degree-day relationship between temperature and timing of gonadal differentiation will be sought.

Significance: In addition to investigating the effect of water temperature on tilapia growth and production, the Rwanda/ Africa PD I A CRSP has also been an active participant in the INAD for clearance of 17cx-methyltestosterone in the production of all-male tilapia fingerlings. In particular, the relationship between growth rate and efficacy of sex reversal has been a main focus. A clear understanding of the exact time of gonadal differentiation in tilapia grown at different temperatures is critical to more closely pinpoint the appropriate time to best administer hormone-treated feed. The ability to obtain a high percentage of sex-reversed tilapia fry with the application of a minimum amount of hormone, applied only at the time it is effective, would clearly be a boon. American tilapia farmers, marketing to consumers increasingly concerned with food safety, would particularly benefit. While the histological events in "normal" gonadal development of 0. niloticus fry with very slow growth have been reported, the time of gonadal differentiation in Nile tilapia fry reared at more common growth rates has not been investigated.

Null Hypothesis: Timing of gonadal differentiation in Nile tilapia fry is not affected by temperature.

Experimental Design: Tilapia fry will be reared at each of three temperatures- 30°C, 26°C and 22°C. Samples will be taken daily and analyzed histologically to determine when gonadal differentiation occurs.

Facilities: Research will be conducted at Auburn University. Experimental units are 45-L aquaria which receive water from a temperature-controlled reservoir, supplied by well water. Histological work will be carried out in the Fisheries Histology Lab at Auburn University.

Culture Period: Feeding and sampling period will be 28 days.

Stocking Rate: Recent swim-up fry, 9-12 mm TL, will be stocked into 4 aquaria per temperature at a rate of 330 per 45-L aquarium.

Nutrient Inputs: Diet will be Zeigler's control-diet tilapia feed. Fish will be fed the maximum feed consumption rate determined in Study 4A for each temperature. Daily rations will be divided into four approximately equal meals offered during daylight hours.

Water Management: The reservoir for each battery of aquaria is approximately equal in volume to the aquaria it supplies. Water flow to each aquarium will be maintained at a minimum of two exchanges per hour. Temperatures will be recorded, and adjusted if necessary to + / - 1°C, at least four times daily. All aquaria will be siphoned regularly to remove uneaten

19

feed and accumulated debris. Dissolved oxygen, total ammonia nitrogen and pH will be measured weekly, and more often if necessary. Total alkalinity and total hardness will be measured at the beginning and end of each trial.

Sampling Schedule: Fish will be sampled every four days to determine total biomass for calculation of proper feed rate. Ten fish per day will be randomly removed from one aquarium at each temperature and preserved. Preserved fish will be embedded in paraffin, sectioned, stained, and mounted for microscopic examination of the gonads.

Statistical Methods: ANOV A and regression analysis.

Schedule: September 1995 to February 1996

20

AFRICA PROJECT STUDY 8: THE EFFECT OF FEED STORAGE TIME AND STORAGE TEMPERATURE ON GROWTH RATE OF TILAPIA FRY AND EFFICACY OF SEX REVERSAL TREATMENT

Cooperating Institution:

Auburn University Ronald Phelps Edwin Smith

Objectives: The effects of storage temperature on the shelf life of a commercial methyltestosterone-treated feed will be studied. Indicators of shelf life will be fry growth and efficacy of sex reversal treatment.

Significance: Hormone-treated feed used for sex reversal of tilapia fry is stored a short time under refrigeration until use. For most large-scale operations this practice is not inconvenient, but for many small-scale, isolated producers the need for refrigeration is a serious constraint to sex reversal. An extended shelf-life under refrigeration followed by an additional shelf life at tropical ambient temperatures would make sex reversal of tilapia a viable option for more producers, thus reducing one of the bottlenecks to expanded production in many regions of the developing world.

Null Hypothesis: Sex reversal and growth rate of tilapia fry during treatment are not effected by storage time of feed at 4°C and subsequent maintenance at ambient temperatures.

Experimental Design: Standard sex reversal conditions for all treatments will be 0. niloticus fry less than 11 mmTL stocked in nylon hapas suspended in open ponds at a fry density of approximately 4000 fry /m2; other treatment protocol will be as indicated in the methyltestosterone INAD authorized by FDA. Treatment variables are storage conditions: two storage times in a refrigerator (0 and 2 months), each at three storage times in sealed plastic bags at tropical ambient temperature (maxima of 0,7, and 28 days). Maxima of 7 and 28 days means that the feed required for 7- or 28 day treatment periods will be removed from freezer and stored at environmental temperatures for those treatment periods. Each treatment will have four replicates. Growth rate of fry will be measured weekly during treatment. After suspending hormone treatment, fry will be reared to at least Sg and preserved for later microscopic examination to determine phenotypic sex.

Facilities: Research will be conducted at El Carao Fish Culture Station in Honduras, where at least two earthen ponds will be used. At least 24 fine mesh nylon hapas will be used during the hormone treatment phase of the study. Microscopic examination of gonads to determine phenotypic sex will be done at El Carao and at Auburn University.

Culture Period: Hormone treatment will be 28 days. Rearing after sex reversal will be four to six weeks.

Stocking Rate: Fry with TL not exceeding 11 mm will be stocked at a density of approximately 4000m2 .

Nutrient Inputs: During sex reversal fry will be fed with a FDA-authorized commercial sex reversal feed and at the rate prescribed in the methyltestosterone INAD authorized by FDA.

21

Water Management: Ponds will receive only sufficient water to replace evaporation and seepage.

Sampling Schedule: Maximum and minimum water temperatures and dissolved oxygen will be measured daily in all ponds. Secchi disk depth will be measured at least weekly. Fry will be measured and counted at the beginning and end of each treatment. Test feeds will be analyzed for degree of lipid oxidation by the peroxide number technique and TBA method.

Statistical Methods: Factorial ANOVA, regression analysis.

Schedule: This study was initiated under the last workplan but will continue to August 1995.

Logistic Considerations: See previous study. (study 6)

22

AFRICA PROJECT STUDY 9: FEED CONSUMPTION, GROWTH, CONVERSION EFFICIENCY OF TILAPIA FRY AND EFFICACY OF SEX REVERSAL TREATMENT AS A FUNCTION OF WATER TEMPERATURE.

Cooperating Institution:

Auburn University

Objectives: Effects of temperature and feeding rate on growth, maximum feed consumption, feed conversion and efficacy of sex-reversal will be determined for 0. niloticus, from recent swim-up stage for a minimum period of 28 days.

Significance: A central focus of the Rwanda PD I A CRSP has been the effect of water temperature on tilapia production. Earlier controlled studies were implemented at UAPB, and a study on the effect of fluctuating and constant temperatures on appetite and growth of juveniles has been completed as a scheduled part of this work plan. Appetite and growth response of tilapia fry to different temperature regimes has not yet been evaluated. Factors affecting growth rate of fry are especially critical in the production of monosex seed, often the primary constraint to broad adoption of food fish production strategies evaluated through the CRSP. Information on maximum feed consumption of fry of different sizes as a function of temperature is required to enable proper feeding of the hormone-treated feed during sex reversal. Likewise, the growth rate of fry during treatment likely influences the efficacy of sex reversal, but little information is currently available.

Null Hypothesis: Study A: Maximum feed consumption of tilapia fry is not affected by temperature. Study B: Growth, feed conversion efficiency and efficacy of sex reversal are not affected by feeding rate and temperature.

Experimental Design: Study A: This experiment will be conducted as three trials, one at each of the following temperatures: 30°C, 26°C, and 22°C. To determine maximum growth rate and approximate maximum feed consumption (expressed as % BW Id) of fry, fish at each temperature will be fed at seven different feed rates for a period of 28 days. At each temperature, maximum feed consumption will be defined as the lowest feeding rate at which maximum growth occurs.

Study B: This experiment will determine the efficacy of sex reversal of fry fed for 28 days at 30%, 60%, and 100% of satiation with feed containing 60 mg methyltestosterone/kg at temperatures of 30°C, 26°C, and 30°C. After suspending hormone treatment, fry will be reared to at least 5 g and preserved for later microscopic examination to determine phenotypic sex.

Facilities: Research will be conducted at Auburn University. Experimental units are 45-L aquaria which receive water from a temperature-controlled reservoir, supplied by well water.

Culture Period: Feeding period will be 28 days. In the 22°C trial of Study A, if fish have not reached an average size of 200 mg (21 mm TL), the trial will be continued to a maximum of 40 days.

Stocking Rate: Recent swim-up fry, 9-12 mm TL, will be stocked into four replicate aquaria per treatment at a rate of 300 per 45-L aquarium.

23

Nutrient Inputs: Basal diet will be Zeigler's control-diet tilapia feed. In study A, fish will be fed at the following feed rates: 30C- 10%, 13%, 16%, 19%, 22%, 25% and 28% BW / d; 26°C- 7% 10%, 13%, 16%, 19%, 22% and 25% BW Id; 22C- 4%, 7%, 10%, 13%, 16%, 19% and 22% BW Id. In Study B, approximately 200 ml of ethyl alcohol containing 60 mg of 17 a-methyltestosterone will be added to each kg of diet and allowed to volatilize at room temperature. Study B fish -will be fed 30%, 60% or 100% of the maximum feed consumption rate determined in Study A fro each temperature. Daily rations will be divided into four approximately equal meals offered during daylight hours.

Water Management: The reservoir for each battery of aquaria is approximately equal in volume to the aquaria it supplies. Water flow to each aquarium will be maintained at a minimum of two exchanges per hour. Temperatures will be recorded, and adjusted if necessary to + /- lcC, at least four times daily. All aquaria will be siphoned regularly to remove uneaten feed and accumulated debris. Dissolved oxygen, total ammonia nitrogen and pH will be measured weekly, and more often if necessary. Total alkalinity and total hardness will be measured at the beginning and end of each trial.

Sampling Schedule: Fish will be sampled every four days by counting all fish in each aquarium and weighing them as a group to determine total biomass. At harvest, all fish in each aquarium will be weighed as a group and counted, and a minimum of 20% of the remaining fish will be measured for length.

Statistical Methods: ANOVA and regression analysis.

Schedule: WP7 Study 4 was not completed by 30 April 1995 due to mortality problems which caused a several-month delay. The first two trials of Study 4A have been completed (trial 1 completed in October 1994; trial 2 completed on 27 May 1995). The third trial is scheduled to begin 1 June 1995, with study 4B remaining to be conducted. September 1994 to September 1995.

24

AFRICA PROJECT STUDY 10- ECONOMICS

Cooperating Institution:

University of Arkansas at Pine Bluff Carole Engle

Objective: To evaluate impact of regional cost difference on optimal product mix and fish farms in Rwanda.

Significance: Previous work in Rwanda developed mathematical programming models to asses optimal product mix on fish farms in Rwanda under a variety of conditions, including risks. However, each of these models used a single set of prices, when in fact prices of various crops vary widely from region to region. The relative differences in prices also are not consistent from region to region and will affect decisions made on the best crops to raise in different regions.

Experimental Design: This study will utilize the models developed in previous work plans those with and without risks parameters. A multi- year plans both those with and without risk parameters. A multi­year price data will be used to model regional price differentials and assess its effect on optimal product mix. Analyses will be performed primarily on the campus of the University of Arkansas at Pine Bluff.

Schedule: June 1-May 1996

25

THAILAND PROJECT

Cooperating Institutions:

University of Michigan Dr. James Diana Dr. C. Kwei Lin

University of Hawaii Dr. James Szyper Dr. Kevin Hopkins

Asian Institute of Technology, Bangkok, Thailand Prof. Peter Edwards

Royal Thai Department of Fisheries, Bangkok Mr. Chaninthom Sritongsuk

Overview: During the interim year, the institutions listed above will continue to collaborate as the Thailand Project. Field work will consist of completion of unfinished activities from Work Plan 7, and performance of the Global Experiment. Other activities consist of compilation of data for transmission to the data base, production of manuscripts reporting research results, and examinations of the Central Data Base for background and trends pertinent to issues described in specific activities below.

26

THAILAND PROJECT ACTIVITY 1: NET PHOTOSYNTHETIC OXYGEN PRODUCTION AND FISH PRODUCTION DURING PD/A CRSP POND EXPERIMENTS

Objectives: 1. To synthesize and analyze pertinent information contained in the CRSP Central Data Base on the relationship between oxygen production and fish production. 2. To relate the analysis in 1. above to the outcomes of Pondclass trials in Thailand and the Philippines.

Significance: This activity is a coherent unit of work which is mentioned in the continuation proposal. This unit, however, will be competed during the interim year. The proposed work will synthesize from the CRSP Data Base and from the other work program participants, the relationships under various conditions between photosynthetic oxygen production in ponds and production of fish biomass under various conditions during CRSP experiments. These relationships have not been compiled on the CRSP-global scale. Understanding currently available information on the connection between photosynthetic oxygen production and fish production will facilitate interpretation of existing data on fertilization by documenting the second part of the fertilization-photosynthesis-fish production sequence, and may assist with formulation of new questions. CRSP study of pond ecosystems has focused on primary productivity. This focus was natural and appropriate consequence of the principle, and some evidence that photosynthesis as the source of fixed carbon to food chain must be critical to fish production in fertilized ponds. There were earlier efforts by two DAST components which examined productivity from the Central Data Base, but no recent or fully global syntheses of the relationships, as noted above.

Hypotheses: 1. Fish production and dNPP will be significantly correlated on at least some scales of space and time. 2. Site, season, and nature of fertilizers (organic or inorganic) will have significant influence on correlations.

Activity Plan: The Central Data Base will be examined by correlation analysis and other techniques for preliminary information on the nature and strength of relationships. Further analysis will be indicated by first results. "Photosynthetic oxygen production" for this purpose will mean the daytime net increase in dissolved oxygen (DO) concentration on a water-column or areal basis. This is termed "daytime net primary production (dNPP)," and is, in practical terms, close to the "DO at 1600 hours" quantity which is available from the diel sampling regimes. This quantity will be averaged or integrated over time intervals similar to those on which fish growth is measured. "Production of fish biomass" will mean the net increment in total weight of the stock in a pond during a given period, generally to exclude the weight of fry produced after stocking. Reproduction will be assessed secondarily, as appropriate. Global analyses by standard statistical methods (simple and multiple regression, ANOVA) will be made, but it is likely that the most definitive and useful relationships will be found within large subsets of the total data set, for example, using a division between bright dry seasons compared with darker rainy seasons, or treatments involving or not involving organic inputs. This work is not intended as a general examination of factors which may be controlling the observed photosynthesis, except in broad terms as stated here.

Facilities: UH-Monoa, UH-Hilo

27

Schedule: 06/95-08/95

THAILAND PROJECT ACTIVITY 2: RELATIONSHIPS AMONG MEASURES OF SUSPENDED PARTICULATE MATERIAL IN THE CRSP DATA BASE

Objectives: 1. To synthesize and analyze pertinent information contained in CRSP Central Data Base on the relationship between chlorophyll concentration, Secchi depth, and total and volatile suspended solids. 2. To examine relationships of measures of suspended particulates with fish production and other parameters of water quality.

Significance: Mineral (clay) turbidity is widely recognized as a factor in pond-to-pond variation in support of blooms and fish production without feed inputs. There is a abundant theory and some practical information about the interaction of mineral turbidity with efficient photosynthetic utilization of light in culture ponds, which is the current subject of a DAST modeling effort. It would be valuable for several purposes to develop a convenient index of the percentage of total turbidity contained in mineral particles. CRSP monitoring protocols include Secchi disc depth and analysis of chlorophyll and of filter-collected suspended solids (total and volatile or organic) to index turbidity of various types. Limited (non-global) examination of the CRSP Central Data Base suggests 1) that these quantities may provide useful description of the mineral content of pond particulate matter under global analysis of the Central Data Base; and 2) they are not adequate outside the global context, and better indexing is needed. The proposed activity will compile and analyze information on Secchi depth, chlorophyll a concentrations, and concentrations of total suspended solids (TSS) and total volatile solids (TVS) from the Central Data Base, and will examine additional analyses pertinent to indexing the proportion of mineral solids in pond suspended particulate matter.

Hypotheses: 1. Significant correlations exist among measures of organic suspended particulate materials. 2. Mineral turbidity can be distinguished from organic turbidity in the Central Data Base.

Activity Plan: The Data Base (DB) will be examined by correlation analysis and other techniques for preliminary information on the nature and strength of relationships. Further analysis will be indicated by first results. If TSS and TVS analysis were sufficiently sensitive to the division under study, the mineral fraction could be obtained as [(TSS-TVS)/TSS]. This quantity will be compared with the indices developed by other methods, and traced through the Central Data Base with chlorophyll and Secchi depth as well.

Facilities: UH- Monoa, UH-Hilo

Schedule: 06/95-08/95

28

THAILAND PROJECT ACTIVITY 3: POL YCULTURE IN DEEP PONDS

Objective: 1. To perform an experiment from Work Plan 7 (Thailand Study 2) which has not yet been done due to the time required for the deep pond construction. 2. To assess the effect of other fish specie on the water quality and yield of tilapia and of all fish in deep, rainfed ponds.

Significance: These deep ponds may develop severe stratification and low oxygen in the bottom waters, as well as loss of nutrients to the deep hypolimnion. Tilapia alone cannot break up this stratification (although our earlier experiments have shown slightly decreased stratification during to fish in some treatments), and wind is often insufficient to do so. In the absence of rain, the ponds become nearly permanently stratified and develop anoxia. Addition of large bodied fish, like carp, which forage in the sediments and actively swim throughout the water column, may break down this stratification regularly, which will prevent anoxia and improve water quality. Carp may also utilize different food items and improve the yield compared to tilapia alone. The use of fish for mixing is preferred over mechanical devices in many of these low intensity and subsistence ponds.

Null Hypotheses: That the addition if common carp will not affect the stratification or total yield of fish from the ponds, compared to ponds with tilapia alone. That carp density will affect stratification or total yield.

Experimental Design: Three fish stocking treatments with four replicates per treatment.

Pond Facilities: 12 earthen ponds at Udom (if the newly constructed ponds are competed by then) or at AIT in the deepest ponds available.

Culture Period: Five months in the dry season.

Fish Stocking Rate: Sex-reversed Oreochromis niloticus at 2 fish/m2. Three treatments will differ in stocking density of common carp, with 0 carp in treatment one, 500 carp per hectare in treatment 2, and 1000 carp per hectare in treatment 3. Carp will be stocked at 500 g in size.

Nutrient Inputs: All treatments will receive fertilizer application at the optimum frequency and rate determined in study 1.

Water Management: Initial filling of the ponds, then addition only by rainfall.

Sampling Schedule: Standard protocols except as noted. Diurnal sampling every month at 4 depths (stratification can be quantified by top to bottom differentials in 02 or temperature, as well as depth of the thermocline if one develops). Include secchi disk depth in diurnals. Evaporation weekly. Weather data daily.

Statistical Methods: ANOV A

Schedule: 07 /95 to 12/95

29

THAILAND PROJECT ACTIVITY 4: FINISHING SYSTEM FOR LARGE TILAPIA: A CONTINUATION, PART 3, OF WP7 STUDY 3

Objective: To determine techniques for producing tilapia at 500 g market size with the best efficiency.

Significance: Tilapia fetch a much higher price per kilogram when sold at 500 g compared to 250-300 g common in our pond grow-outs. However, growth declines at such a large size in semi-intensive ponds, and growing these fish to large market size may take much too long and be too costly. The use of supplemental feeds can accelerate this growth, and caging of tilapia in ponds for the final grow outs can efficiently produce a number of large tilapia while still utilizing the nutrients produced from feeding to grow smaller tilapia in a semi-intensive system. Such a system, similar to our earlier work on catfish-tilapia polyculture, would allow small scale farmers to use one pond to efficiently grow young tilapia and still feed older tilapia for a larger final size. Cost/ benefit analyses will also allow a comparison of this caging technique to earlier at-large techniques of supplemental feeding.

Null Hypotheses: Cages in themselves have no effect on carrying capacity of ponds or fish yield. Density of small fish has no effect on yield.

Experimental Design: 4 treatments, 3 ponds per treatment. Tl: Large tilapia stocked in 2 cages/ pond at 200 fish/cage. Small tilapia free in pond at 2 fish persqm. T2: As in Tl, but small tilapia at 1.5 fish per sq m. T3: Large tilapia stocked in open water at 400 fish/pond. Small tilapia stocked in open water at 2 fish per sq m. T4: As in T3, but small tilapia at 1.5 per sq m

Pond Facilities: 12 earthen ponds at AIT

Culture Period: 3 months

Fish Stocking: Sex-reversed Oreochromis niloticus will be used. Tilapia stocked cages will average 250g in size. ·

Feeding Rate: All tilapia cages fed at 3% body weight per day during the first month, 2.5% during the second, and 2% during the third.

Water Management: Replace evaporation and seepage losses weekly.

Sampling Schedule: Standard protocols except as noted. Intensive measurement of physical, chemical, and biological parameters every two weeks. Nutrient bioassay every two weeks.

Statistical Methods: ANOV A

Schedule: 06/95-09/95

30

PHILIPPINES PROJECT

Cooperating Institutions:

University of Hawaii Dr. James Szyper Dr. Kevin Hopkins

Central Luzon State University, Munoz, Nueva Ecija, Philippines Freshwater Aquaculture Center Prof. Eduardo Lopez

Overview: During the transition year, the University of Hawaii will continue to collaborate with CLSU for work in the Philippines. Field work will consist of completion of unfinished activities from Work Plan 7, performance of the Global Experiment, and new field trials which will be completed during the transition period. Other activities consist of compilation of data for transmission to the data base, and production of manuscripts reporting results.

31

PHILIPPINES PROJECT ACTIVITY 1: STOCKING DENSITY IN RELATION TO ELEVATION IN THE PHILIPPINES

Objective: To determine the most appropriate stocking density for male Nile tilapia in enriched ponds located in distinct elevation zones.

Significance: This activity was adopted by the Philippines Project from the Rwanda Project. It was begun during WP7, and will be completed during the transition period. It was originally designed to "tie up previous and current research" and "refine pond management guidelines" (Topics 1 and 4 of suggested subject matter for W orkplan 7). As water temperature decreases, the physiological capability of tilapia to consume available food likely decreases more rapidly than does the capacity of the pond to produce natural food organisms. Consequently, optimum fish stocking density for a given nutrient input regime may vary as a function of elevation.

Experimental Design: Three elevation ranges (near 2000 m, 1500 m, and 1000 m), three fish stocking densities, with each treatment replicated in five farmer-managed ponds.

Pond Facilities: Forty five farmer-owned ponds, 15 at each of 3 elevation ranges.

Culture Period: Three to five months or until fish :i;each an average weight of 150 g.

Fish Stocking Rate: Juvenile male Oreochromis niloticus at 1, 2 and 3 per m2.

Nutrient Inputs: Inorganic fertilization at 14 kg N/ha/week and an N/P ratio of 5:1.

Water Management: Replace evaporation and seepage losses weekly.

Sampling Schedule: Standard on-farm sampling protocol.

Statistical Methods: ANOVA, regression analysis.

Schedule: Study ends on June 1995

32

PHILIPPINES PROJECT ACTIVITY 2: YIELD TRIALS WITH GENETICALLY SELECTED TILAPIA

Objective: To determine the growth and yield of genetically-improved tilapia in ponds fertilized according to the most current CRSP guidelines.

Significance: This is a continuation of the concept of WP7 Study 9 of the same name, with new comparisons following from results as they are obtained. CRSP research in the Philippines has completed preliminary yield trials of genetically-selected Oreochromis niloticus produced by the PAC/University of Wales Swansea Research Project on Genetic Improvement of Tilapia.

Experimental Design: 4 treatments, 3 replicate ponds/treatment Treatments: A. Stock juv. GMT Nile tilapia at 2 fish/m2, plus fingerling snakehead at 0.3 fish/m2.

B. Tilapia GMT as above plus Clarias fingerlings at 0.3 fish/m2.

C. Mixed sex tilapia fingerlings plus snakehead as above. D. Mixed sex tilapia plus Clarias as above.

Pond Facilities: Twelve 500 m2 ponds at F AC Culture Period: Five months.

Nutrient Inputs: Most current CRSP fertilization guidelines.

Schedule: 08/95- 01/96

33

PHILIPPINES PROJECT ACTIVITY 3: PHILIPPINES OUTREACH

Objective: To link our experimental results with outreach and extension programs in the Philippines.

Significance: This is an ongoing activity, as WP 7 ElO was. Results from the first round of lowland on-farm trials will be compiled and reported during the transition period. CRSP activities in the Philippines started in 1991 to conduct on-station trials of CRSP guidelines. By 1993, on-station trials were able to be complimented by on-farm trials. In this way, we hope to adapt the CRSP guidelines to the conditions facing Philippines farmers.

Experimental Design: The overall approach consists of three phases, two of which have been completed. First, a study was conducted to determine if the CRSP fertilization guidelines were socially acceptable and economically viable under Philippine conditions. Second, a mechanism to improve and disseminate CRSP guidelines was developed by the CRSP in cooperation with fish farmers, potential fish farmers and extension personnel. Currently, one round of field tests of the CRSP guidelines has been completed at selected private farms. The second round will be performed during the transition period.

Pond Facilities: At least 10 private fish ponds.

Culture Period: To be determined based on local fish size preferences and estimated returns.

Fish Stocking Rate: To be determined based on local fish size preferences and estimated returns.

Nutrient Inputs: Based on most current CRSP fertilization guidelines.

Sampling and Water Management: Standard on-farm protocols.

Schedule: 6/95 to 3/96; report due: 4/96

34

DATABASE MANAGEMENT

Cooperating Institution: University of Hawaii Dr. Kevin Hopkins Mr. John Wassell

Background/Significance: The CRSP has developed a database management system to store and retrieve most data collected during CRSP experiments since the start of the CRSP. The database is a resource of major importance for researchers and access is available upon request. By analyzing large amounts of data simultaneously, it has been possible to detect trends which are not readily apparent with smaller data sets.

Objectives: The basic objectives of this activity during the transition year are: -Complete entry of all "old and/ or outstanding" data; -Review and correct all incoming data; -Complete restructuring the database to allow entry of data from flexible forms; -Provide text capability so that textual information can be submitted together with quantitative data; -Preparation of new manual for database entry which facilitates entry of data in flexible forms that are more convenient to the field researchers; -Explore the possibility of providing access to the database using the World-Wide Web; -Data will be provided to the DAST and other interested persons in several different formats upon request. -Maintain the current relational database.

In addition to the basic objectives, several supplemental activities will be undertaken when time is available. These activities are: -Data from field stations will be reviewed for conformance to formats and outliers (>3 s.d. from the mean) values will be identified. This information will be transmitted to the field stations for verification and/ or correction. -Development of analytical routines including non-linear regression for growth curves; -Explore the possibility of distributing the CRSP database on CD-ROM with a user-friendly query interface; -Annual Meeting 1996 (and via e-mail thereafter)- Development and presentation of a training program for CRSP participants to encourage them to change from spreadsheet-based data management to FoxPro (both IBM-compatible and Macintosh) in order to facilitate data analysis in the field.

Schedule: Database management is an on-going activity which will be conducted for the duration of the program.

35

DATA ANALYSIS AND SYNTHESIS TEAM

Cooperating Institution:

Oregon State University Dr. John Bolte, Bioresource Engineering. Principal Investigator Shree Nath, Research Assistant Doug Ernst, Research Assistant Priscilla Darakjian, Research Assistant

Introduction: The revised proposal for continuation of the activities of the OSU Data Analysis and Synthesis Team (DAST) of the PD/ A CRSP during 1995-96 includes two components: 1) completion of Work Plan 7 studies previously approved by the CRSP, and 2) transitional-year studies that reflect long-term objectives of the DAST as identified in the continuation proposal. The transitional-year long-term objectives of the DAST as identified in the continuation proposal. The transitional-year studies emphasize continued development of POND as a tool for analysis and planning, and activities aimed at developing links with the US aquaculture industry and international aquaculture researchers. The latter activities will involve development and application of POND for decision support relevant to catfish and baitfish aquaculture in the US, and the potential use of POND to address regional- scale aquaculture developments and related environmental issues in developing countries.

Specific deliverables from the OSU-DAST during the period 1 May 1995 to 30 April 1996 include the following:

-POND Version 2.5 by 31August1995 (with the functionality indicated in Work Plan 7), and -POND Version 3.0 by 30 April 1996 (with the functionality indicated in the Transitional Year Studies 1-3 below).

Completion of Work Plan 7 Studies: Work that remains to be completed under Work Plan 7 includes analysis of data from the global experiment designed to test PONDCLASS at the PD I A CRSP sites, continued validation of POND models, development of optimization strategies for aquaculture facility economics, and completion of technical manual for POND. This manual will document the simulation models and economic analysis framework in POND. The manual will also describe use of these analysis tools to examine the effects of alternate pond management strategies on facility-level fish yields and economic returns. Management strategies that will be addressed in the manual include different rates of pond fertilization and

· supplementary feeding, various fish stocking densities and sizes (under both monoculture and polyculture situations), different stocking and harvest dates, and water quality.

Schedule: May to August 1995

36

DAST STUDY 1: DEVELOPMENT AND USE OF POND AS A TOOL FOR ANALYSIS AND PLANNING

Cooperating Institution:

Oregon State University

Background and Justification: POND provides a useful framework for synthesizing knowledge about pond aquaculture and presenting it in a form that can be used for decision support. The software currently provides capabilities for addressing planning and management decisions relating to the location and size of aquaculture facilities (including fish ponds and lots), species used, soil types, as well as fertilizer, feed, and lime sources used at a facility. Analyses are accomplished by the use of simulation models coupled with an economics analysis package. The more sophisticated models in POND can also be used as research tools to explore fundamental relationships between fish growth, the natural food composition of ponds, water quality and management practices. Future extension of POND as a tool for decision support and planning requires additional work in the areas of existing model development and validation analysis of the resulting models, and assessment of the requirements for future modeling efforts.

Although some components in the POND models have been validated, it is necessary to further refine and validate other components as knowledge of pond dynamics continues to increase. For example, although fish-plankton relationships in ponds have been studied for several years, there have been relatively few attempts to comprehensively models such relationships. The fish-plankton models that have been included in POND predict qualitative shifts in plankton pools that are consistent with available knowledge, but CRSP data to complete validation of the models are sparse. OSU/DAST researchers have initiated a collaborative effort with international researcher teams (headed by Dr. Mark Prein at ICLARM, Dr. Ana Milstein in Israel, and Dr. Malcolm Beveridge from the University of Stirling, U.K) to address this issue. We propose to work closely with these research teams to further calibrate and validate Level 1 and 2 POND models. This effort will enhance the ability of the software to better predict pond dynamics across global sites.

Interactions with CRSP researchers have indicated the importance of feed quality characterization on pond production and economics. Further, the water quality models in POND have not been completely validated. Because of the importance of environmental issues such as water use and the quality of effluents discharged from aquaculture facilities, validation of this aspect of POND is critical for its utilization in environmental decision-making. Inclusion of water balance and flow considerations in the POND models will allow these issues to be addressed. Practical benefits for farm management and planning are expected from this activity because users may be able to analyze the effects of additional production of costs imposed by increased water requirements under different scenarios (e.g., location of facilities on soils with higher seepage rates), and local water use and/or environmental regulations.

Additionally, it is becoming increasingly important to link aquacultural analyses with associated agricultural systems to identify potential synergies and conflicts in resource utilization between the two systems, and to develop guidelines for planning integrated farming systems. Initial work in this area is currently underway under the UCD /DAST. Additional work is needed to link the UCD terrestrial agriculture models with the POND decision support framework.

Objectives: 1) Further refine and validate models of fish-plankton relationships in ponds. 2) Characterize feed types used in domestic and international production and extend the bioenergetic model in POND to consider feed quality. 3) Further validate components of the water quality models in POND. 4) Incorporate preliminary considerations of water balance and flow in the POND models.

37

5) Incorporate terrestrial agriculture models into the POND framework. 6) Conduct a general sensitivity analysis of POND to identify key production variables.

Tasks: Modeling of fish-plankton relationships will be undertaken in collaboration with international researchers identified above. Such collaboration will include common use of data available from various researchers. The OSU /DAST will classify available plankton data by size class or some suitable approach based on preferences of different fish species for various plankton components. Simulation experiments will then be conducted to compare model output with the available data. If needed, existing POND models will be refined based on results of the simulation experiments.

The bioenergetic model in POND will be modified to consider feeds of different quality (e.g. protein, energy, and moisture contents), and quantitative changes in supplemental feed consumption by fish depending on feed quality. This will allow exploration of the effects of different feeds on fish performance and facility-level economics. The bioenergetic model will also be modified to enable users to specify typical feeding rates at their facilities on a percent body weight basis; this modification will help users fine-tune POND for their particular needs.

CRSP data will be used to further validate some of the components (e.g. alkalinity, nitrogen, and phosphorus pools, etc.) in the water quality models in POND. Finally, the models will be modified to consider factors such as rainfall, evaporation and seepage that influence pond water balance. Preliminary consideration of flow through individual ponds and the overall facility will also be included in the models.

We will work with the UC-Davis team to explore linkages between terrestrial and aquatic production models and develop a prototype decision support system that combines the models within the POND framework.

A general sensitivity analysis of POND will be conducted using representative baseline datasets from existing CRSP and US sites to determine which variables most impact production and economics of aquaculture facilities. The results of this analysis will be used to assess requirements for future model development and parameterization.

Schedule: September 1995 to April 1996

38

DAST STUDY 2: DECISION SUPPORT FOR POND PRODUCTION SYSTEMS IN THE UNITED STATES

Cooperating Institution:

Oregon State University

Background and Justification: Feedback from the interactions of the OSU /DAST members with US farmers and researchers suggests that the decision support system POND developed at Oregon State University has considerable potential for use in the US aquaculture industry. Two principal industries where the software can be used are catfish and baitfish aquaculture. POND is likely to be of use in these aquaculture industries both for farm management, as well as planning activities.

Objectives: 1) Examine current fertilization, feeding and water quality management practices in catfish and baitfish aquaculture facilities in the Southern US. 2) Consolidate existing datasets that can be used for calibration and validation of the models in POND for catfish and baitfish. 3) Generate production estimates and economic summaries by stimulating different management practices at selected sites.

Tasks: A survey of existing production and research farms in the Southern US will be undertaken. Datasets that include fish growth, management inputs, water quality and cost information will be collected from researchers familiar with catfish and baitfish aquaculture as well as farmers. These datasets will be used to parameterize the simulation models in POND. If warranted, appropriate changes to the POND models or existing software architecture will be made. The calibrated models will be used to simulate different management practices, and the resulting production estimates and economic summaries compared with existing data. The overall simulation results and conclusions will be presented to farmers and researchers. Feedback from these farmers and researchers will be used as a component of requirements assessment.

Schedule: September 1995 to April 1996

39

DAST STUDY 3: APPLICATION OF POND FOR DECISION SUPPORT AT OTHER INTERNATIONAL SITES

Cooperating Institution:

Oregon State University

Background and Justification: Researchers from other international development agencies (e.g. F AO) have expressed considerable interest in exploring the use of POND both for planning and managing pond aquaculture facilities. Interest has also been expressed on the part of FAO researchers to apply POND for regional-scale analysis, including environmental impact studies. However, POND in its current form has limitations in this regard. Therefore, we propose to conduct a meeting/workshop with FAO personnel (and other researchers, if possible) to discuss the possibility of a collaborative effort in the future development of POND. Involvement of development agencies with the activities of the OSU /DAST would be a powerful mechanism of promoting the use of POND and furthering the visibility of the PD I A CRSP at different international sites.

Objectives: 1) Demonstrate the capabilities of POND to FAO (and other) researchers. 2)Promote use of the current version of the software at some of their international sites. 3) Develop a framework for applying POND on a pilot basis for large-scale aquaculture planning. 4) Identifying changes that may be required in the software for large-scale pond aquaculture and environmental impact analyses.

Tasks: The OSU/DAST will organize a meeting with FAO researchers (and other workers, if possible) interested in POND. The meeting will involve a workshop designed to familiarize the participants with the capabilities of POND, and assess key areas for future POND development. We will also discuss the possibility of using POND for decision support at international sites. Specific areas where POND may be useful in this regard include the development or fine-tuning of farm management practices, environmental issues (e.g. water quality), and economics of pond aquaculture.

The possibility of using POND for regional-scale analysis will also be explored. Such a study will indicate future modifications that may be necessary in the software to address regional issues such as estimates of aquaculture potential, socioeconomic impacts, and the development of appropriate guidelines for effluent discharge from aquaculture facilities.

Schedule: September 1995 to April 1996

40

DAST STUDY 4: A WATER QUALITY/FISH YIELD MODEL WITH STOCHASTIC INPUTS

Cooperating Institution:

University of California at Davis

Objectives: 1) To develop a model for using stochastically generated solar radiation and wind values for estimating water temperature and dissolved oxygen concentration in stratified ponds over a growing season; 2) To develop a model that uses stochastically generated solar radiation and wind values, and temperature and water quality models to predict fish growth and fish yield.

Significance: Current models of water quality and fish production in aquaculture ponds are based on deterministic relationships between atmospheric, chemical and biological parameters within the ponds. The relationships and model proposed under this project would constitute a first attempt at the development of a probabilistic model that can be used to project water quality and fish production over a growing season. The model would be valuable in that risk or probability factors would be associated with particular fish yields under a given set of management practices. This information would then be available for decision making purposes. Work to date has resulted in computer tools for the use of PD/ A CRSP weather data to generate stochastic solar radiation and wind predictions. These stochastic predictions have been used to run PD/ A CRSP models of water temperature in stratified ponds. Work is under way to adapt previously developed models of dissolved oxygen so that they can be executed with stochastic inputs over a full growing season. The techniques used for stochastic temperature and dissolved oxygen models will be extended to modeling fish growth.

Methods: A procedure has been developed to use PD I A CRSP solar radiation and wind data to generate stochastic solar radiation and wind values. This procedure is unique in that it allows for the use of data sets that may not be complete and that may extend for time periods as short as five to ten years. A previously developed model to simulate water temperature in stratified ponds has been executed with stochastically generated solar radiation and wind data corresponding to the PD/ A CRSP Thailand site. Results of the simulations provide insight into the probable temperature fluctuations that can be expected over time given typical weather patterns for the site. Work is under way to adapt a previously developed dissolved oxygen model to long term execution (over a growing season) using stochastic solar radiation and wind values. The adaptation process requires substantial revisions of the estimation of oxygen production and consumption rates to reduce the cumulative error over long term simulations. Revision of the model for long term execution also includes provision for estimating primary production and plankton concentration over time. Plankton concentrations will then form the basis for estimating fish growth and fish yield. Fish growth will be modeled using an adaptation of a bioenergetic model included in POND© Programming will be carried out using STELLA', a modeling language available for both Windows' and Macintosh' operating systems. The model will be executed such that probability distributions of yields are obtained as an output. These distributions will be obtained by running the model many times (approximately 50 based on results to date) using stochastically generated solar radiation and wind values. Measured values of output parameters available in the PD I A CRSP Database will be compared to simulation results to determine if the measurements fall within the range of expected values obtained from the stochastic simulations. The model output will then be assembled in a new

41

probability distribution which can be analyzed to examine the likelihood of exceeding a particular level of production.

Schedule: The first version of a model to use stochastically generated solar radiation and wind values for estimating temperature has been developed. This model is being revised to improve the stochastic generation of solar radiation and wind values, and the revision should be complete by July 95. The first version of the oxygen and water quality simulation model will be completed and tested by August 1995. Integration of the fish growth and fish yield models with the water quality model will be complete, and stochastic fish growth and fish yield results will be available by April 30, 1996.

42

DAST STUDY 5: AQUACULTURE POND MODELING FOR THE ANALYSIS OF INTEGRATED AQUACULTURE/ AGRICULTURE SYSTEMS.

Cooperating Institution:

University of California at Davis

Objectives: 1) To develop a computer model which can be used to analyze and predict nitrogen and organic matter outputs from an aquaculture pond. 2) To incorporate the pond model into a model that simulates the flow of organic matter and of nitrogen through aquaculture and through conventional agricultural practices which may be integrated with aquacultural activities.

Significance: As the number of aquaculture ponds continues to expand around the world, and as the intensity of production in those ponds increases, so does the potential for limitations in the availability of supplies such as fertilizers and feeds, and for impacting the environment. Given this growth, it is essential that the efficiency of utilization of resources be optimized by modifying pond management practices as needed, and by integrating aquaculture with other activities such as agriculture. Computer models such as those proposed here are important tools in achieving these objectives. Models can serve to analyze field data and to suggest critical questions that may be examined experimentally in the field. Finally, the models can serve as management and planning tools.

Methods: The models to be developed will build on models which have been created and tested under the current PD/ A CRSP project. Work during the transition period will focus on the development of simple nutrient flux models for agricultural production which can be integrated with pond models, and on fish growth and fish feeding. Existing PD I A CRSP pond models will be modified to include considerations of nutrient cycling with a special emphasis on nitrogen: nitrogen release and uptake in the water column and sediments; nitrogen accumulation in sediments, and removal of sediments for use in agriculture, etc. The inclusion of the sediment terms will necessitate additional information on the interaction between sediments and the water column, which is currently not available. The first version of the sediment submode! will be based on literature information and on the limited information on sediments that is available in the current PD/ A CRSP database. Work during the transition period will provide a starting point for further developments as information is generated in field work to be carried out under the new grant. Field work being proposed for the next PD I A CRSP grant will greatly expand our knowledge base on sediments and sediment/water interactions, and will result in improved pond models and pond management practices.

Schedule: Modification of a bioenergetics fish growth model included in POND ©will be completed by December 1995. The modification will include techniques for dealing with multiple feed sources and with changes in their concentrations and in the preference of fish for a particular feed type. The next component of the model to be addressed is the accumulation of sediments. A first version of a sediment mass balance will be completed by April 30, 1996, and will consider nitrogen and organic matter.

43

SOCIAL SCIENCE PROJECT: SOCIOECONOMIC DIMENSIONS OF AQUACULTURE DEVELOPMENT: A COMPARATIVE ASSESSMENT OF FINANCIAL RETURNS, ADOPTION BARRIERS, AND IMPACTS OF TILAPIA PRODUCTION REGIMES

Cooperating Institution:

Auburn University JosephJ. Molnar Leonard L. Lovshin Terry R. Hanson

Introduction: The Pond Dynamics/ Aquaculture CRSP has been organized to generate basic science that may be used to advance aquacultural development. The socioeconomic component of the project has been designed to generate information to complement and clarify the impact of the PD I A CRSP in selected host countries.

Year 1 examined the socioeconomic appropriateness of technology being investigated by the CRSP Global Experiment in Honduras, Thailand, and the Philippines. The framework for identifying the relevant social and economic criteria used the results of previous work in Rwanda as a starting point for developing the more general perspective.

Year 1 was primarily devoted to data collection. Year 2 completed the data collection effort and began analysis. We addressed the feasibility of implementing CRSP strategies and will more fully attend to the ways that CRSP technologies might actually be extended to progressive farmers. We completed data collection. Presently we are analyes of CRSP technologies and production strategies.

In year 3 we will endeavor to integrate our results with the DAST simulation package to integrate our analysis with the larger biological and production data framework developed there.

Year 3 also may include an advance visit to Vietnam by Molnar to explore the difficulties with tilapia farming being experienced there. This visit may take place during the week or ten days in advance of the CRSP Annual Meeting. We will enter into discussions with CRSP staff members at the CRSP Annual Meeting and with host country scientists and officials in Vietnam.

Travel from Bangkok to Vietnam is contingent upon the availability of non-CRSP funds to support field work in Vietnam We will make a series of farm visits in localities characterized by higher rates of pond abandonment or undr-utilization of the pond enterprise. The finding of this visit may suggest alternative production strategies to be explored or research questions to be addressed by the CRSP in the context of the Global Experiment.

Objectives: 1. Report the results of surveys conducted in Years 1 and 2 profiling costs and returns associated with alternative production regimes specified by the CRSP workplans in Rwanda, Thailand, the Philippines, and Honduras. 2. To describe the practices, technical proficiency, and receptivity to the adoption of CRSP technologies and production regimes among tilapia farmers. 3. To explore the social and institutional feasibility of extending the PD I A CRSP paradigm to Vietnam through preliminary field work.

44

Significance: The multiple identities of tilapia stem from its role as an export commodity generating foreign exchange that simultaneously coexists with that of a subsistence crop. While industrial-scale operations produce tilapia for outside markets, farm ponds provide food security, nutrition, and occasional cash income for the rural poor. The paper to be presented to the Annual Meeting of -the Rural Sociological Society considers variations in the organization of tilapia production across the four countries. Rwanda has the least global connections and influence on its nascent industry whereas Thai and th Philippines feed companies have central roles in the evolution of tilapia production technology, processing for export, and promotion of the enterprise among farm operators. Honduras suffers from weak institutional commitment to the rural sector and uncertain industry leadership. In the context of shifting US. policy toward foreign assistance and agricultural development in general, tilapia has remained a multiple-front effort to build export domestic markets, export trade, and enhance village well-being. Some of the tensions and contradictions among these mandates are explored.

The extension of CRSP research efforts to Vietnam represents the opportunity to develop baseline information about the technology needs and production niches for fish farmers in Vietnam. This information may help guide the selection of the research strategies and types of technologies to be developed by CRSP research. Information detailing farmer perceptions of production and marketing problems should be useful for shaping the problems addressed in research trials and the formulation of realistic prescriptions for farm level practice.

Recent visitors to Vietnam report the street price of tilapia to be around $0.28 to $0.49 for lkg or larger fish. Vietnamese are reputed to each fish up to 4 times per week and per capita consumption is around 12 kg/year, the lowest in Asia. Most families have small fish ponds located near houses. In low-lying areas, the ponds are associated with excavations resulting from home site construction. Land tenure is based on government concession, but sufficient title security apparently is present to motivate land improvements such as pond construction. Rents or taxes are paid by some farmers. The poorest people are in the uplands. They are targeted for aquacultural development. Le Thanh Hung, Associate Fisheries Dean at the University of Agriculture and Forestry in Saigon is interested in doing socioeconomic assessments.

In Vietnam, the research will provide baseline information about the practice of aquaculture, the availability of inputs, and the structure of incentives and constraints that shape farm­level decision making. Understanding the role of fish culture in the household and the local economy is crucial for developing appropriate technology and selecting effective mechanisms for technology transfer. The field work will complement the visits made by other PD I A CRSP personnel using non-federal funds to Vietnamese institutions and production areas. This work leverages the expense and time commitment associated with participation in the PD/ A CRSP annual meeting to accomplish baseline socioeconomic research that will be funded by other sources.

Methods: Sample and Data Collection: Data have been collected from tilapia farmers in four CRSP countries: Rwanda, Honduras, Thailand, and the Philippines. The following sections detail the procedures employed in each country and the approach to use of the data for this paper.

Rwanda. Data were obtained from a sample of 121 active Rwanda fish farmers in eight local administrative districts (communes) during the winter and early spring of 1992. The 141 communes (or countries) are the basic units of administration in Rwanda. Several districts were chosen to represent diversity in the nation's regions; others were selected randomly. Interviews were conducted with 115 active fish farmers randomly selected form National Fish Culture Service (SPN) extension rolls. About 45 percent of the respondents were women.

45

Honduras. Data were obtained from a sample of 51 active Honduran fish farmers in five of 15 Honduras departments during the fall 1993. Tilapia farmers were identified through referrals made by Peace Corps volunteers working in fish culture, Honduran extension personnel, and by farmers identifying neighbors raising tilapia. The departments were chosen to represent the major tilapia production regions in the country. Interviews were conducted with 51 active fish­farmers.

Thailand. Data were obtained from a sample of 51 active Thai fish farmers in four of 75 Thai provinces during winter 1994. Tilapia farm operators in Ayutthaya, Nakhom Pathom, and Suphanburi provinces Central Thailand were interviewed. The survey was revised and adapted, then translated into the Thai language.

Tilapia farmers were identified through referrals made by Department of Fisheries extension personnel, knowledgeable local individuals, and by fish farmers identifying neighbors raising tilapia. The provinces were chosen to represent major tilapia production regions in south central Thailand, the major aquaculture region in the country. Interviews were conducted with 51 active fish farmers.

Philippines. Data were obtained from a sample of Philippine fish farmers in four of 15 provinces on the main island of Luzon during winter 1994. Tilapia farm operators in Bulacan, Nueva Ecija, Pampanga, and Tarlac provinces were interviewed. The survey was revised and adapted, then translated into the Tagalog language.

Tilapia farmers were identified by sampling lists of farmers purchasing fingerlings at the Freshwater Aquaculture Center at Central Luzon State University in Munoz. Sample farmers were asked to identify neighbors raising tilapia who also were approached for interviews. The provinces were chosen to represent the major tilapia production region in the country. Interview were conducted with 51 active fish farmers. Figure 4.4 shows the location of the sample provinces.

Vietnam. Vietnam is a prospective site for PD I A activities. If funding can be obtained from other sources, Joseph J. Molnar will visit Vietnam for approximately ten days in advance of the Annual Meeting to conduct field work in prospective target areas for PD I A CRSP technologies. Field work means unstructured interviews with farmers, researchers, extensionists, officials, fish sellers, and other informants knowledgeable about aquaculture. The purpose of the field work will be to identify the niche occupied by tilapia in the Vietnamese agricultural system. Molnar will collaborate with C.K. Lin to identify sites, institutional collaborators, and country-specific research issues. This work will be planned in consultation with AIT staff members Harvey Demaine and Peter Edwards who have established extensive contacts and relationships with Vietnamese institutions.

Analysis: In each country, the interview schedules were edited to reconcile missing data, ambiguous answers, and exceptional cases. The data were keypunched according to precoded numerical response categories on the printed questionnaire that did not require translation. Open-ended questions elicited verbatim comments or explanations were cumulated in a separate process and presented here in tabular form or summarized in the narrative as appropriate.

The data for each country sample are tabulated in parallel to facilitate comparisons across the four sites. The objective is to document the nature and circumstances of practicing tilapia farmers who might adopt CRSP technologies. The data establish the basic community, household, labor, farming system, and marketing environment for the conduct of fish culture at each site.

46

The field work in Vietnam will yield a project report that should lead to a published paper. It also will provide socioeconomic advisory information for the PD I A CRSP as it considers new project sites.

Schedule: June 1995 Continue writing, data analysis. Submit draft Final Report to CRSP management for review. Develop articles for refereed journals. Complete first year report; obtain review from CRSP scientists.

July 1995 Incorporate CRSP scientist comments in Final Report. Submit to Alabama Agricultural Experiment Station for approval as a Research and Development Series publication.

August 1995 Present paper based on project results to the Annual Meeting of the Rural Sociological Society "The multiple identities of tilapia as a farm enterprise: the growth of a global commodity in the context of development assistance." Revise paper for submission to World Development.

September 1995 Initiate planning for Vietnam visit.

October 1995 Develop strategy for Vietnam visit. Submit quarterly report on October 10. Submit 6 Copies of Final Report to CRSP management by October 15.

November 1995 Continue economic analysis of technological alternatives.

December 1995 Continue economic analysis of technological alternatives.

January 1996 Attend Annual Meeting; present overview of project results to the meeting; visit Vietnam tilapia farming area.

February 1996 Revision of draft report that evaluates impacts of CRSP technology.

Development of journal articles for submission to social science journal, applied aquaculture journal.

March 1996 Continue writing, data analysis. Attend CRSP Annual Meeting. Present paper to American Fisheries Society Annual Meeting.

April 1996 Continue writing, data analysis. Issue Final Report as AES R&D series.

May 1996 Develop journal articles for submission to referred journals.

47