a finishing system for large tilapia

113

Southeast Asia

Increasing the biomass of caged tilapia shouldincrease the nutrients available for the growth ofopen-pond tilapia. However, increasing the stockingdensities within single cages has been found tolower overall fish production (Yi et al., 1996). Analternative strategy is to increase the number ofcages within a pond. The current experiment testedthe use of multiple cages in ponds, and the use ofwater aeration to reduce the risk of oxygen depletion(Boyd, 1990). The objectives of this study were todetermine the appropriate biomass of large tilapia incages required to support maximum production ofsmall tilapia in open water, and also to investigatethe effects of aeration on water quality and growthperformance of both caged and open-pond tilapia.

METHODS AND MATERIALS

Two experiments were conducted for 90 days(January-April 1985) and 84 days (July-September1995) at the Asian Institute of Technology (AIT) inThailand. Large tilapia (91 ± 5.2 to 103 ± 4.6 g) werestocked in 4-m3 net cages (50 fish m-3) suspended inearthen ponds. One, two, three, or four cages weresuspended in each earthen pond as experimentaltreatments. Tilapia fingerlings (13 ± 0.3 to 16 ± 1.3 g)were stocked at two fish m-3 in the open water ofall ponds four days after the cages were stocked.Both caged and open-pond Nile tilapia werehormone-treated, sex-reversed males that hadreceived methyltestosterone treatment in the frystage.

The first experimental trial was conducted ina randomized complete block design in twelveponds. Eight ponds measured 335 m2 surface areawith a water depth of 1.2 m, and four pondsmeasured 394 m2 surface area with a water depthof 1 m. The water volume of all ponds was similar

INTRODUCTION

The major production system for Nile tilapia(Oreochromis niloticus) is semi-intensive withinorganic or organic fertilizer inputs in earthenponds. However, interest in the cage cultureof tilapia has increased, particularly in tropical,developing countries (Coche, 1982). Caged fishare commonly fed with high protein diets; wastesfrom the caged tilapia in the form of dissolvednutrients, uneaten feed, and metabolic products,either directly or indirectly released into thesurrounding pond environment, cause acceleratedeutrophication (Beveridge, 1984; Ackefors, 1986;Lin et al., 1989). Fish-livestock integrated systemshave been practiced widely for centuries (Pillay,1992). Therefore, it has been suggested that wastesfrom cage culture could also serve as a valuableresource in an integrated aquaculture system bygenerating natural foods for filter-feeding speciessuch as Nile tilapia (Lin et al., 1989; Lin, 1990;McGinty, 1991; Yi et al., 1996).

A series of experiments was designed to developa tilapia-tilapia cage-cum-pond integrated rotationculture system. Large Nile tilapia were stocked incages suspended in earthen ponds while smallNile tilapia were stocked outside the cages in theopen pond. The open-pond fish utilize cage wastesand can be transferred from the open pond torestock the cages. Large Nile tilapia (> 500 g) canfetch a much higher price in some countries than thesmaller Nile tilapia (250-300 g) that are commonlyproduced in fertilized pond systems. CagedNile tilapia, in a tilapia-tilapia cage-cum-pondintegrated culture system, have been grown to500 g within 90 days (stocked at 50 fish m-3 with onecage per 335-m2 pond of 1.0 to 1.2-m water depth)(Yi et al., 1996). However, the wastes derivedfrom the single cages were insufficient to generateabundant natural foods for the growth of theopen-pond fish.

A FINISHING SYSTEM FOR LARGE TILAPIA

Interim Work Plan, Thailand Activity 4

Yang Yi and C. Kwei LinAgricultural and Aquatic Systems

Asian Institute of TechnologyPathum Thani, Thailand

Fifteenth Annual Technical Report

114

(approximately 330 m3). The ponds were dividedinto three blocks, with each block containing pondsof similar dimensions. One replication from eachtreatment was randomly assigned to one pondin each block. Metal frame cages (2 x 2 x 1.2 m)covered with 2-cm mesh nylon net were suspendedto a depth of 1 m in each pond. In the shallowerponds, an area of nine m2, below each cage, wasdeepened by 20 cm to keep the cage floors 20 cmfrom the pond bottom.

The cages were arranged within ponds asfollows: for the one-cage treatment each cage wasplaced in the center of a pond; for the two-cagetreatment, cages were placed at either end of eachpond along the pond central line, 2 m from the pondbottom edges; for the three-cage treatment, onecage was placed in the middle of each pond andthe other two were placed in the opposite cornersof the pond, 2 m from the pond bottom edges, andthe other two were placed in the opposite cornersof the pond, 2 m from central lines; for the four-cagetreatment, cages were arranged in the four cornersof each pond, 2 m from bottom lines. To confinefloating pellets within the cages, a fine meshpolyethylene net was fixed from 5 cm above to15 cm below the water surface on the outside ofeach cage. A wooden or bamboo walkway connectedeach cage to the pond bank. To prevent fish fromjumping between the cages and the pond, and toprotect fish from predation by birds, the cages werecovered with nylon nets.

Water was added weekly to replace lossesdue to seepage and evaporation. No fertilizerwas applied to any experimental pond so that thegrowth of open-pond tilapia was solely dependenton natural foods derived from cage wastes. Cagedtilapia were fed with commercial floating pellets(30% crude protein, Charoen Pokphand Co., Ltd.)at 0900 and 1600 h, six days per week. Feedingrates were 3%, 2.5%, and 2% body weight per day(BWD) during the first, second, and third month,respectively. The feed ration was adjusted dailybased on mortality and the biweekly sample weightof caged tilapia. Small tilapia stocked in open waterwere not given artificial feed.

Average weights of tilapia were determinedbiweekly by bulk weighing 10% of tilapia fromeach cage in addition to 40 open-pond tilapiaper pond. Caged tilapia were sampled by dipnet and open-pond tilapia by seine. Tilapia wereharvested, counted, and bulk weighed at the endof the experiments.

To estimate total nitrogen and phosphorusloading from cages to open water, resultingfrom caged-tilapia waste products, the nitrogen(N) and phosphorus (P) content of carcasses ofcaged tilapia (harvested and dead) was deductedfrom the N and P content of the feed input.

Water column samples were taken biweekly nearthe center of each pond at approximately 0900 hfor the analysis of pH, total ammonia-nitrogen,chlorophyll a, total suspended solids and totalvolatile solids (APHA, 1985). Values for pHand temperature were used to determine theamount of unionized ammonia-nitrogen in ponds(see conversion table in Boyd, 1990). Temperatureand DO were measured between 0600 and 0700,and between 1500 and 1600 h with an oxygenmeter (YSI model 54).

Due to high mortality (65.5-70.5%) of tilapia in allcages of one replicate of the four-cage treatment,and due to the low levels of DO recorded, a secondexperiment was conducted to examine the effect ofnighttime aeration on the growth performanceof tilapia in this cage-cum-pond integrated culturesystem. An aeration trial was conducted fromJuly through September 1995 in three ponds,similar to the ponds of the first experiment. Fourcages, similar in dimensions and stocking densityto the first experiment, were placed in each pond.A paddle-wheel aerator (0.75 kW and 1,400 rpm)was mounted at the mid-point of the long sideof each pond. Ponds were aerated from 0300 to0800 h daily.

Due to unavailability of sufficient ponds,controls could not be run concurrently withthis second experiment. Data from the four-cagetreatment in the first experiment provided controldata (the two non-aerated ponds that did notexperience high mortality). The comparison ofdata collected at different times of the year can bejustified because the results of other experimentsconducted from August to November 1994, fromJanuary to April, and from July to September 1995showed no apparent effects of temperature on thegrowth performance of either caged or open-pondtilapia. Water temperature in each culture cycleonly briefly fell outside the optimum range fortilapia growth (Yi, 1997).

Data were analyzed statistically by analysis ofvariance and regression (Steele and Torrie, 1980)using the Statgraphics 7 statistical softwarepackage. Differences were considered significant

115

Southeast Asia

at an alpha of 0.05. All means are presented with± 1 standard error (SE).

RESULTS

Effects of the Biomass of Caged Nile Tilapia

Mean individual weight of harvested caged tilapiadecreased significantly (P < 0.05) from 478 ± 34.6to 280 ± 32.0 g as the number of cages in a pondincreased (Table 1). During grow-out, the weight oftilapia increased steadily with a mean daily weightgain ranging from 2.10 ± 0.39 to 4.27 ± 0.32 g fish-1

(Figure 1); mean daily weight gain decreased asthe number of cages in ponds increased . Themean daily weight gain in the treatment withone cage was significantly higher than the othertreatments with two or three cages. Survival ofcaged tilapia also decreased significantly from100 ± 0.0% to 76.8 ± 15.9% with an increase in thenumber of cages. The net yield of tilapia in a cage

decreased significantly (P < 0.05) from 76.9 ± 5.8to 25.4 ± 13.2 kg cage-1 crop-1 with an increase in thenumber of cages. However, there were no significant(P > 0.05) differences in total net yield, whichranged from 76.9 ± 5.8 to 96.9 ± 2.7 kg pond-1 crop-1

for all treatments with caged tilapia. The highesttotal net yield was achieved in the two-cagetreatment. Feed conversion ratios in theone- and two-cage treatments were significantlylower (1.22 ± 0.06 and 1.64 ± 0.05) than in thethree-cage treatment (3.01 ± 1.27).

The biomass of caged tilapia also had significant(P < 0.05) effects on the growth, net yield, andsurvival of open-pond tilapia. There were nosignificant differences in survival of open-pondtilapia between the two- and three-cage treatments;however, survival in the two- and three-cagetreatments was significantly higher than in theone-cage treatment. Mean individual weight andnet yield increased with an increase in the numberof cages (Table 2). The mean weight of individual

Performance Measures Treatments (cages per pond)

1 2 3

STOCKING

�Total Wt. (kg cage-1) 18.7 ± 1.4 18.6 ± 0.7 18.2 ± 1.0�Mean Wt. (g fish-1) 94 ± 6.8 93 ± 3.7 91 ± 5.2

HARVEST

�Total Wt. (kg cage-1) 95.6 ± 6.9a 67.1 ± 1.8b 43.7 ± 12.7c

�Mean Wt. (g fish-1) 478 ± 34.6a 341 ± 10.6b 280 ± 32.0c

�Total Wt. Gain (kg cage-1) 76.9 ± 5.8a 48.5 ± 1.7b 25.5 ± 13.2c

�Mean Wt. Gain (g fish-1) 384 ± 28.9a 248 ± 9.6b 189 ± 34.7c

(g fish-1 d-1) 4.27 ± 0.32a 2.75 ± 0.11b 2.10 ± 0.39c

�Net Yield (g m-3 d-1) 213.6 ± 16.0a 134.6 ± 4.6b 70.7 ± 36.7c

(kg cage-1 crop-1) 76.9 ± 5.8a 48.5 ± 1.7b 25.4 ± 13.2c

(kg pond-1 crop-1) 76.9 ± 5.8 96.9 ± 2.7 76.3 ± 34.7(t ha-1 crop-1) 2.30 ± 0.17 2.89 ± 0.08 2.34 ± 1.19

�Survival (%) 100.0 ± 0.0a 98.5 ± 2.1a 76.8 ± 15.9b

� F C R 1.22 ± 0.06a 1.64 ± 0.05ab 3.01 ± 1.27c

�Gross Yield (kg pond-1 crop-1) 95.6 ± 6.9 134.2 ± 1.7 131.1 ± 32.2(t ha-1 crop-1) 2.86 ± 0.21 4.00 ± 0.05 3.97 ± 1.11

Table 1. Growth performance of male Nile tilapia stocked in cages at 50 fish m-3 in treatments with one,two, and three cages per pond.

abc Mean values with different superscript letters in the same row were significantly different (P < 0.05).


116

Figure 1. Growth of caged and open-pond Nile tilapia stocked in treatments with one, two, and threecages per pond.

fish in the one-cage treatment (75 ± 3.6 g) wassignificantly lower than in the two- and three-cagetreatments (114 ± 5.0 and 123 ± 6.5 g, respectively;see Table 2 and Figure 1); however, there wasno significant difference in mean individualweight between the two- and three-cagetreatments. The daily weight gain of open-pondtilapia was significantly different amongtreatments; weight gain increased with the

number of cages per treatment. Total weight gainof open-pond tilapia was positively correlated(r = 0.95, P < 0.01) with total feed input to cages(Figure 2 ). Cage wastes fertilized the ponds atrates of 0.92, 1.76, and 2.36 kg N ha-1 d-1 givingN:P ratios of 5.01, 4.64, and 4.47 for treatmentswith one, two, and three cages, respectively.The extrapolated net yield of open-pond tilapiaincreased significantly from 1.10 ± 0.05 t ha-1 crop-1

117

Southeast Asia

Figure 2. Relationship between total feed given to caged Nile tilapia and total weight gain of open-pondNile tilapia.

for the one-cage treatment to 1.93 ± 0.06 and2.14 ± 0.12 t ha-1 crop-1 in the two- and three-cagetreatments, respectively.

The extrapolated net yield from the two-cagetreatment for caged and open-pond tilapiacombined was 4.83 ± 0.03 t ha-1 crop-1; this yieldwas not significantly higher (P > 0.05) than theyields from the one- and three-cage treatments(3.39 ± 0.22 and 4.41 ± 1.10 t ha-1 crop-1, respectively)(Table 3). The overall FCRs were 0.83 ± 0.03 and1.00 ± 0.03 for the one- and two-cage treatments,respectively. These FCRs were not significantlydifferent from each other, but they were significantlybetter than the FCR for the three-cage treatment(1.43 ± 0.19).

Water temperature averaged 28.7°C and pH rangedfrom 6.8 to 7.9 throughout the experimental period.The measured DO concentrations at dawn for alltreatments decreased sharply over the first month

and then gradually to approximately 1 mg l-1. TheDO concentrations tended to be higher, but were notsignificantly different, in the treatments with a lowerbiomass of caged tilapia (Figure 3). The unionizedNH3-N concentration in the one-cage treatment wassignificantly lower than in the two- and three-cagetreatments (Figure 3). The phytoplankton standingcrop, as indicated by chlorophyll a concentrations,was generally low, but chlorophyll a concentrationwas significantly higher in the three-cage treatmentthan in the one- and two-cage treatments (Figure 3).

Effects of Aeration

Aeration improved the growth performanceof caged tilapia significantly (P < 0.05) (Table 4).The survival of caged tilapia increased from93.9 ± 2.6% in non-aerated ponds to 97.3 ± 2.2% inthe aerated ponds. The mean weight of cagedtilapia harvested from aerated ponds (403 ± 36.3 g)was significantly greater than that from non-aerated


118

Table 2. Growth performance of open-pond, male Nile tilapia stocked in ponds at two fish m-3 witheither one, two, or three cages per pond.

ab Mean values with different superscript letters in the same row were significantlydifferent (P < 0.05).

ponds (261 ± 19.5 g) (Table 4 and Figure 4). Thenet yield of caged tilapia in aerated ponds reached232.0 ± 25.7 kg pond-1 crop-1 compared with the yieldin non-aerated ponds which reached 122.0 ± 10.6 kgpond-1 crop-1. Feed conversion ratios were 1.30 ± 0.12and 2.08 ± 0.02, respectively.

The growth rate of open-pond tilapia was,however, significantly lower (P < 0.05) in aeratedponds (1.21 ± 0.08 g fish-1 d-1) than in non-aeratedponds (1.50 ± 0.14 g fish-1 d-1 ). The final meanindividual weights for aerated and non-aeratedponds were 113 ± 7.1 and 142 ± 11.3 g, respectively(Figure 4). The survival of open-pond tilapiain aerated ponds (90.4 ± 2.2%) was significantlylower than the survival of open-pond tilapia innon-aerated ponds (96.2 ± 1.3%). The net yieldof open-pond tilapia in aerated ponds was alsosignificantly lower than in non-aerated ponds(Table 5).

Pooled growth performance measures of bothcaged and open-pond tilapia were significantlyhigher (P < 0.05) for aerated ponds than fornon-aerated ponds (Table 6). The extrapolated netyield was significantly higher for aerated ponds

(8.62 ± 0.78 t ha-1 crop-1) than for non-aerated ponds(6.19 ± 0.53 t ha-1 crop-1), giving overall FCRs of1.04 ± 0.06 and 1.25 ± 0.01, respectively.

Water temperature averaged 29.0°C and pH rangedfrom 6.8 to 8.0 throughout the experimental period.DO concentration at dawn in non-aerated pondsdecreased sharply from an initial level of 5.27 mg l-1

to less than 1.00 mg l-1 over the first two months;DO was less than 1.00 mg l-1 for the remainder ofthe experiment. DO concentration in the aeratedponds was 1.47 mg l-1 at the beginning of theexperiment prior to aeration, but once aerationwas initiated DO ranged from 3.30 to 4.83 mg l-1

throughout the experiment (Figure 5 ). UnionizedNH3-N concentrations in aerated ponds increasedfrom 0.00 to 0.10 mg l-1 over 90 days, while theconcentrations in non-aerated ponds fluctuatedfrom 0.01 to 0.05 mg l-1 throughout the experiment(Figure 5). The phytoplankton standing crop wasnearly the same in both aerated and non-aeratedtreatments; chlorophyll a concentrations increasedin both treatments but were higher in non-aeratedponds than in aerated ponds at the end of theexperiment (Figure 5). The levels of total suspendedsolids increased gradually in both aerated and


1 2 3

STOCKING

Total Wt. (kg pond-1) 8.6 ± 0.2 9.0 ± 0.5 0.9 ± 0.4Mean Wt. (g fish-1) 13 ± 0.3 14 ± 0.8 14 ± 0.6

HARVEST

Total Wt. (kg pond-1) 45.5 ± 1.8b 73.8 ± 1.7a 80.5 ± 3.9a

Mean Wt. (g fish-1) 75 ± 3.6b 114 ± 5.0a 123 ± 6.5a

Total Wt. Gain (kg pond-1) 36.9 ± 1.6b 64.8 ± 2.0a 71.6 ± 3.7a

Mean Wt. Gain (g fish-1) 62 ± 3.4b 100 ± 5.2a 109 ± 6.1a

(g fish-1 d-1) 0.72 ± 0.03b 1.16 ± 0.06a 1.27 ± 0.07a

Net Yield (t ha-1 crop-1) 1.10 ± 0.05b 1.93 ± 0.06a 2.14 ± 0.12a

Survival (%) 89.7 ± 2.2b 94.6 ± 2.4a 95.5 ± 1.1a

Gross Yield (t ha-1 crop-1) 1.36 ± 0.06b 2.20 ± 0.05a 2.40 ± 0.12a

119

Southeast Asia

Figure 3. Fluctuations in DO at dawn, unionized NH3 (0900 h), and chlorophyll a (0900 h) in treatmentswith one, two, and three cages per pond.


120


1 2 3

Initial Fish Biomass (kg) 27.3 ± 1.5 27.3 ± 1.5 63.6 ± 2.3Final Fish Biomass (kg) 141.1 ± 8.5 208.0 ± 0.9 211.5 ± 33.7Fish Biomass Gain (kg) 113.8 ± 7.2 161.8 ± 1.1 147.9 ± 35.8Net Fish Yield (g m-3 d-1) 3.83 ± 0.25 5.45 ± 0.04 4.98 ± 1.22

(t ha-1 crop-1) 3.39 ± 0.22 4.83 ± 0.03 4.41 ± 1.10Gross Fish Yield (t ha-1 crop-1) 4.21 ± 0.26 6.21 ± 0.03 6.31 ± 1.02Overall FCR 0.83 ± 0.03a 1.00 ± 0.03a 1.43 ± 0.19b

Table 3. Growth performance of male Nile tilapia cultured in ponds stocked with two fish m-3 fromtreatments with either one, two, or three cages per pond.

ab Mean values with different superscript letters in the same row were significantly different (P < 0.05).

Performance Measures Treatments

Aerated Non-aerated

STOCKING

Total Wt. (kg cage-1) 20.6 ± 0.9 18.5 ± 0.5Mean Wt. (g fish-1) 103 ± 4.6 92 ± 2.8

HARVEST

Total Wt. (kg cage-1) 78.6 ± 8.6a 49.0 ± 4.3b

Mean Wt. (g fish-1) 403 ± 36.3a 261 ± 19.5b

Total Wt. Gain (kg cage-1) 58.0 ± 8.0a 30.5 ± 4.0b

Mean Wt. Gain (g fish-1) 300 ± 33.6a 169 ± 17.9b

(g fish-1 d-1) 3.57 ± 0.40a 1.88 ± 0.20b

Net Yield (g m-3 d-1) 172.6 ± 24.1a 84.7 ± 11.0b

(kg pond-1 crop-1) 232.0 ± 25.7a 122.0 ± 10.6b

(t ha-1 crop-1) 6.93 ± 1.03a 3.65 ± 0.30b

Survival (%) 97.3 ± 2.2 93.9 ± 2.6FCR 1.30 ± 0.12a 2.08 ± 0.02b

Gross Yield (kg pond-1 crop-1) 314.4 ± 31.4a 196.0 ± 9.6b

(t ha-1 crop-1) 9.39 ± 1.10 5.85 ± 0.28

Table 4. Growth performance of male Nile tilapia stocked in four cages in aerated and non-aerated ponds.

ab Mean values with different superscript letters in the same row were significantly different(P < 0.05).

121

Southeast Asia

Figure 4. Growth of caged and open-pond Nile tilapia in aerated and non-aerated treatments with fourcages per pond.


122


Aerated Non-aerated

STOCKING

Total Wt. (kg pond-1) 10.5 ± 0.9 8.7 ± 0.4Mean Wt. (g fish-1) 16 ± 1.3 13 ± 0.6

HARVEST

Total Wt. (kg pond-1) 67.3 ± 2.6b 94.0 ± 6.3a

Mean Wt. (g fish-1) 113 ± 7.1b 142 ± 11.3a

Total Wt. Gain (kg pond-1) 56.8 ± 2.1b 85.3 ± 6.7a

Mean Wt. Gain (g fish-1) 97 ± 6.3b 129 ± 12.0a

(g fish-1 d-1) 1.21 ± 0.08b 1.50 ± 0.14a

Net Yield (g m-3 d-1) 2.15 ± 0.08b 3.08 ± 0.23a

(t ha-1 crop-1) 1.70 ± 0.07b 2.55 ± 0.21a

Survival (%) 90.4 ± 2.2b 96.2 ± 1.3a

Gross Yield (t ha-1 crop-1) 2.01 ± 0.09b 2.81 ± 0.19a

Table 5. Growth performance of male Nile tilapia stocked at two fish m-3 in treatments with four cages, inaerated and non-aerated ponds.

ab Mean values with different superscript letters in the same row were significantlydifferent (P < 0.05).


Aerated Non-aerated

Initial Fish Biomass (kg) 92.9 ± 2.9 82.7 ± 1.4Final Fish Biomass (kg) 381.7 ± 34.5 290.2 ± 15.9Fish Biomass Gain (kg) 288.8 ± 32.7a 207.5 ± 17.3b

Net Fish Yield (g m-3 d-1) 10.42 ± 1.18a 7.49 ± 0.62b

(t ha-1 crop-1) 8.62 ± 0.78a 6.19 ± 0.53b

Gross Yield (t ha-1 crop-1) 11.39 ± 0.84a 8.66 ± 0.48b

Overall FCR 1.04 ± 0.06a 1.25 ± 0.01b

ab Mean values with different superscript letters in the same row are significantlydifferent (P < 0.05).

Table 6. Growth performance of male Nile tilapia cultured in ponds stocked at two fish m-3 in treatmentswith four cages, in aerated and non-aerated ponds.

123

Southeast Asia

Figure 5. Fluctuations in DO at dawn, unionized NH3 (0900 h), and chlorophyll a (0900 h) in aeratedand non-aerated treatments with four cages per pond.


124

non-aerated ponds prior to aeration. The levelsof total suspended solids increased dramaticallyin aerated ponds, after aeration, compared tonon-aerated ponds, and there was no significantdifference in total volatile solids between aeratedand non-aerated ponds (Figure 6).

DISCUSSION

The mean weight (478 ± 34.6 g) of caged tilapiain this experiment did not reach the desiredmarket size (> 500 g). This may have been due tothe initial stocking size of caged tilapia (94 ± 6.8 g)in the one-cage treatment which was smaller thanthe initial stocking size in an experiment thatdid produce market-size tilapia (Yi et al., 1996)(148 ± 2.5 g). The mean daily weight gain in thepresent experiment was similar to the results ofYi et al. (1996). The lower daily weight gains inthe two- and three-cage treatments were due tothe higher biomass of caged tilapia. Daily weightgains in this study, which ranged from 2.10 ± 0.39to 2.75 ± 0.11 g fish-1, were higher than those of cagestudies carried out by Guerrero (1979, 1980) or fortilapia reared in thermal effluent by Philippart et al.(1979, cited by Coche, 1982) where daily weightgains ranged from 0.56 to 1.60 g fish-1. The dailyweight gain in the present study is comparable

to growth in ponds obtained by McGinty (1991)(2.09 to 2.49 g fish-1) or to growth in intensively-cultured cages in lakes (1.05 to 2.33 g fish-1)(Coche, 1977 and Campbell, 1978, cited by Coche,1982).

The FCR in three-cage treatment was higher thanthe FCR for the treatments with fewer cages as aresult of relatively higher mortality and lowergrowth rates during the grow-out period. FCRsof the one- and two-cage treatments were similarto the one-cage treatment with the same stockingdensities in an earlier experiment (Yi et al., 1996).FCRs obtained in this study were also similar toFCRs reported by Carro-Anzalotta and McGinty(1986) for intensive cage culture in ponds, butwere much lower than FCRs reported by Guerrero(1979, 1980) and McGinty (1991) for culture inponds and by Coche (1977, cited by Coche, 1982)and Campbell (1978, cited by Coche, 1982) for lakes.The above results indicate that a higher biomassof caged tilapia in a pond negatively affects growthperformance.

The reason for significantly lower survival ofopen-pond tilapia in the one-cage treatment wasnot clear. The survival of open-pond tilapia in thetwo-, three- and four-cage treatments in non-aeratedponds was greater than 90%. Survival and growth of

Figure 6. Comparison between concentrations of TSS and TVS in aerated and non-aerated ponds.

125

Southeast Asia

open-pond tilapia was not influenced by poor waterquality as was the growth and survival of cagedtilapia. This may have been due to the differingrequirements and tolerances to water quality. Fishcontained at high densities in cages, in contrast tofree-swimming fish in earthen ponds, are unableto seek out zones of favorable water quality andmay be more susceptible to fluctuations in pondwater quality, particularly to low concentrations ofDO (Hargreaves et al., 1991). Instantaneous changesin water quality in cages, especially after feedingin late afternoon, in addition to accumulated cagewastes may cause a sudden decline in water qualityaround cages.

Daily weight gains of open-pond tilapia, whichincreased with the increased biomass of cagedtilapia in this experiment, were much higher thandaily weight gains from a previous experiment(Yi et al., 1996), but were still lower than daily weightgains obtained from catfish-tilapia cage-cum-pondintegrated culture (2.4 to 2.7 g fish-1) (Lin, 1990)and tilapia-tilapia cage-cum-pond integrated culture(2.31 g fish-1) (McGinty, 1991). This is probablydue to the lower feed input to cages and a higherstocking density of open-pond tilapia used in thepresent experiment. The poor growth performanceof open-pond tilapia in Yi et al.’s (1996) experimentmay have been caused by the relatively large initialstocking size of tilapia. Growth may have beenstunted when the tilapia were held in fertilizedponds at high density for about three monthsprior to the experiment. In the present experimenttreatments with a higher biomass of caged tilapiaproduced more wastes which improved the growthof open-pond tilapia.

The nitrogen loading rates derived fromcaged tilapia, ranging from 0.92 to 2.36 kg ha-1 d-1,were lower than the optimal fertilization rates(approximately 4 kg N ha-1 d-1) recommendedfor semi-intensive culture of tilapia in tropicalponds (Knud-Hansen et al., 1991). However, theextrapolated gross yields of open-pond tilapiafrom this experiment increased significantly from5.44 ± 0.18 t ha-1 yr-1 in the one-cage treatment to8.80 ± 0.14 and 9.60 ± 0.35 t ha-1 yr-1 in the two- andthree-cage treatments, respectively. These yieldssurpassed yields obtained from catfish-tilapiacage-cum-pond integrated culture systems(Lin, 1990), conventional integrated fish-livestocksystems (AIT, 1986), and systems optimally fertilizedwith either chicken manure (Diana et al., 1988) orchemical fertilizers (Diana et al., 1991). In the presentexperiment, open-pond tilapia depended solely

on the wastes (dissolved nutrients from feed,metabolic wastes and uneaten feed) derived fromcages. Uneaten feed, knocked out of the cages asa result of the vigorous swimming activity of cagedtilapia (Collins, 1971; Coche, 1979; McGinty, 1991)provided open-pond tilapia with some supplementalfeed. This may explain why open-pond tilapiain this experiment demonstrated better growthperformance than tilapia cultured in traditionallyfertilized ponds, despite the lower fertilizationrates. However, as evidenced by the very lowFCR of caged tilapia, the amount of uneaten feedwas probably limited. This may imply that dissolvednutrients from feed and metabolic wastes of cagedtilapia are in a more utilizable form or are in moreefficient supply than from organic manure orchemical fertilizers.

The combined extrapolated net yieldsof caged and open-pond tilapia ranged from13.56 ± 0.41 to 19.32 ± 0.10 t ha-1 yr-1 in theone-, two-, and three-cage treatments. They weremuch higher than yields achieved in the earlierexperiment (6.5 to 9.7 t ha-1 yr-1) (Yi et al., 1996),primarily because the growth performance ofopen-pond tilapia in this experiment was better.The higher stocking density of open-pond tilapiain the present experiment also resulted in higheryields (approximately 5.2 to 6.6 t ha-1 yr-1) thanin the tilapia-tilapia cage-cum-pond integratedculture system reported by McGinty (1991).

The higher mortality in treatments with a higherbiomass of caged tilapia may have been causedby an extended period of exposure to low DOconcentrations. Extended periods of hypoxia mayreduce growth (Chervinski, 1982) and causemortality (Coche, 1982). Net yield of all cagedtilapia leveled off when the number of cages inponds increased or the biomass of caged tilapiaincreased, which indicates that the carryingcapacity of caged tilapia, in the earthen pondsstocked with small tilapia at two fish m-3, mayhave been exceeded. High numbers of open-pondtilapia caused a decrease in the growth of cagedtilapia (McGinty, 1991), therefore lowering thestocking density of open-pond tilapia may bethe best way to increase the harvest size of bothcaged and open-pond tilapia.

Significantly higher gross yields and better feedefficiencies were found in aerated ponds comparedto non-aerated ponds. Similar results were obtainedin channel catfish ponds (Lai-Fa and Boyd, 1988).The survival of caged tilapia in aerated ponds was


126

also similar to survival rates in aerated channelcatfish ponds (Hollerman and Boyd, 1980). Growthperformance of caged tilapia was much better inaerated ponds than in non-aerated ponds. This wasprobably due to the higher DO concentrations ofaerated ponds at dawn throughout the experimentalperiod. However, daily weight gain of caged tilapiain aerated ponds was lower than in non-aeratedtreatments with one cage, and the final meanweight of caged tilapia in aerated ponds did notreach the desired size (> 500 g). This was probablydue to the smaller stocking size (103 ± 4.6 g) andshorter grow-out period (84 days) than in Yi et al.’s,(1996) experiment (141 ± 11.1 to 152 ± 2.1 g and90 days, respectively). The aeration experimentwas terminated 6 days earlier than plannedbecause of the threat of a flood. Another reasonfor the lower growth rates might be the three-weekdelay of aeration at the beginning of this experimentdue to technical problems, and a power outagethat lasted 24 hours near the end of the experiment.However, aeration for five hours each nightprevented nighttime DO concentrations fromfalling below 3 mg l-1. Since DO concentrationswere adequate (Boyd, 1990), some other waterquality parameters, probably ammonia, imposedlimits on production. Teichert-Coddington andGreen (1993) reported that total ammonia-nitrogenconcentrations were significantly higher in aeratedtreatments than in non-aerated controls. Boyd (1990)found that total ammonia-nitrogen concentrationsof 1 to 3 mg l-1 were common in aerated pondswith high feeding rates and that sublethal effectsof ammonia probably adversely affected fishgrowth. The total ammonia-nitrogen concentrationsin aerated ponds in this experiment averaged 1.83 to2.56 mg l-1 after the first month and probably causedthe lower growth of caged tilapia in aerated pondscompared to treatments with lower biomass of cagedtilapia in the present and previous experiments(Yi et al., 1996). Water exchange has been shownto improve high ammonia concentrations inchannel catfish ponds (Parker, 1979; Plemmonsand Avault, 1980); however, most farms do nothave sufficient water supply to permit high ratesof water exchange. Boyd (1990) suggested that itmight be possible in some locations to partially drainand quickly refill ponds with well water of lowammonia concentration when ammonia-nitrogenconcentrations become too high. This combination ofnightly aeration with partial water exchange mightbe the best method for eliminating low nighttimeDO concentrations and high concentrations of totalammonia, thereby increasing production in pondswith high feeding rates.

The quantities of total suspended solids were muchgreater in aerated ponds than in non-aerated ponds,while the total volatile solids values were similar.The aeration-induced mud turbidity interferedwith algal growth in the aerated ponds, andthus reduced the availability of natural foods foropen-pond tilapia. Lowered primary productivityin aerated ponds may also cause significantlyhigher concentrations of total ammonia (Teichert-Coddington and Green, 1993), which further reducesgrowth of open-pond tilapia. A combination of all ofthese factors probably caused the significantlylower growth rates of open-pond tilapia in theaerated ponds. Teichert-Coddington and Green(1993) demonstrated that aeration could enhanceyields but that it had little effect on water qualityother than to increase turbidity. Thus, anothermethod to increase nighttime DO levels may beto circulate water. Water circulation appears tostimulate phytoplankton growth (Sanares et al.,1986) which could increase DO production byphotosynthesis (Boyd, 1990).

ANTICIPATED BENEFITS

This study demonstrated the feasibility of thetilapia-tilapia cage-cum pond integrated culturesystem which entails the intensive cage culture oflarge tilapia in earthen ponds in which small tilapiaare semi-intensively cultured. Results also indicatedthat nighttime aeration for five hours improvedgrowth performance of caged tilapia stocked atthe highest biomass for use in the integrated culturesystem. This culture system will allow small-scalefarmers to maximize their profits. Since open-pondtilapia effectively recovered wastes, this culturesystem benefits both aquaculture production andthe environments receiving pond water discharge.

ACKNOWLEDGMENTS

The authors wish to thank Ms. Chintana B.,Mr. Manoj Y., and Mr. Supat P. for their field andlaboratory assistance and James Diana (Universityof Michigan) for reviewing this report. This studyforms part of Mr. Yang Yi’s doctoral dissertationsupported by the PD/A CRSP.

LITERATURE CITED

Ackefors, H., 1986. The impact on the environmentby cage farming in open water. J. Aqua.Trop., 1:25-33.

127

Southeast Asia

AIT, 1986. Buffalo/fish and duck/fish integratedsmall scale system at the family level.Research Report No. 198. Asian Institute ofTechnology, Bangkok, Thailand, 138 pp.

APHA, 1985. Standard Methods for the Examinationof Water and Waste water, 16th Edition.American Public Health Association,American Water Works Association,and Water Pollution Control Federation,Washington, D.C., 1268 pp.

Beveridge, C.M., 1984. Cage and pen fish farming:Carrying capacity models andenvironmental impacts. FAO FisheriesTechnical Paper No. 255. FIRI/222. Foodand Agriculture Organization of theUnited Nations, Rome, 131 pp.

Boyd, C.E., 1990. Water Quality in Ponds forAquaculture. Auburn University, Auburn,Alabama, 482 pp.

Carro-Anzalotta, A.E. and A.S. McGinty, 1986.Effects of stocking density on growth ofTilapia niloticus cultured in cages in ponds.Journal of the World Aquaculture Society,17:52-57.

Chervinski, J., 1982. Environmental physiologyof tilapias. In: R.S.V. Pullin and R.H. Lowe-McConnell (Editors), The Biology andCulture of Tilapias. ICLARM ConferenceProceedings 7, 2-5 September 1980,Bellagio, Italy. International Center forLiving Aquatic Resources Management,Manila, Philippines, pp. 119-128.

Coche, A.G., 1979. A review of cage fish cultureand its application in Africa. In: T.V.R.Pillay and W.A. Dill (Editors), Advances inAquaculture. Fishing Book News, London,pp. 428-441.

Coche, A.G., 1982. Cage culture of tilapia. In: R.S.V.Pullin and R.H. Lowe-McConnell (Editors),The Biology and Culture of Tilapias.ICLARM Conference Proceedings 7,2-5 September 1980, Bellagio, Italy.International Center for Living AquaticResources Management, Manila,Philippines, pp. 205-246.

Collins, R.A., 1971. Cage culture of catfish inreservoir lakes. Proceedings of the Annual

Conference Southeastern Association ofGame and Fish Commissioners, 24:489-496.

Diana, J.S., P.J. Schneeberger and C.K. Lin, 1988.Relationship between primary productionand yield of tilapia in ponds. In: R.S.V.Pullin, T. Bhukaswan, K. Tonguthai, andJ. Maclean (Editors), The SecondInternational Symposium on Tilapia inAquaculture. ICLARM ConferenceProceedings 15, 16-20 March 1987, Bankok,Thailand. International Center for LivingAquatic Resources Management, Manila,Philippines, pp. 1-6.

Diana, J.S., C.K. Lin and P.J. Schneeberger, 1991.Relationship among nutrient inputs, waternutrient concentrations, primary productionand yield of Oreochromis niloticus in ponds.Aquaculture, 92:323-341.

Guerrero, R.D., 1979. Cage culture of tilapia in thePhilippines. Asian Aquaculture, 2(11):6.

Guerrero, R.D., 1980. Studies on the feeding ofTilapia nilotica in floating cages.Aquaculture, 20:169-175.

Hargreaves, J.A., J.E. Rakocy, and D.S. Bailey, 1991.Effects of diffused aeration and stockingdensity on growth, feed conversion, andproduction of Florida red tilapia in cages.Journal of the World Aquaculture Society,22:24-29.

Hollerman, W.D. and C.E. Boyd, 1980. Nightlyaeration to increase production of channelcatfish. Trans. Amer. Fish. Soc., 109:446-452.

Knud-Hansen, C.F., C.D. McNabb, and T.R.Batterson, 1991. Application of limnologyfor efficient nutrient utilization in tropicalpond aquaculture. Verh. Internatl. Verein.Limnol., 24:2541-2543.

Lai-Fa, Z. and C.E. Boyd, 1988. Nightly aeration toincrease the efficiency of channel catfishproduction. Prog. Fish-Cult., 50:237-242.

Lin, C.K., 1990. Integrated culture of walkingcatfish (Clarias macrocephalus) and tilapia(Oreochromis niloticus). In: R. Hirano andI. Hanyu (Editors), The Second AsianFisheries Forum, 17-22 April 1989, Tokyo,Japan. Asian Fisheries Society, Manila,Philippines, pp. 209-212.


128

Lin, C.K., K. Jaiyen and V. Muthuwan, 1989.Integration of intensive and semi-intensiveaquaculture: concept and example. ThaiFisheries Gazette, 43:425-430.

McGinty, A.S., 1991. Tilapia production in cages:Effects of cage size and number of noncagedfish. Prog. Fish-Cult., 53:246-249.

Parker, N.C., 1979. Channel catfish production incontinuously aerated ponds. In: ResearchWorkshop Summary of Papers, CatfishFarmers of America Annual Meeting,Jackson, MS, USA, 44 pp.

Pillay, T.V.R., 1992. Aquaculture and theEnvironment. Fishing Book News,London, 189 pp.

Plemmons, B. and J.W. Jr. Avault, 1980. Six tonsof catfish per acre with constant aeration.Louisiana Agriculture, 23:6-8.

Sanares, R.C., S.A. Katase, A.W. Fast, and K.E.Carpenter, 1986. Water quality dynamics inbrackishwater shrimp ponds with artificialaeration and circulation. In: J.L. Maclean,

L.B. Dizon and L.V. Hosillos (Editors), TheFirst Asian Fisheries Forum, 26-31 May 1986,Manila, Philippines. Asian Fish. Soc.,Manila, Philippines, pp. 83-86.

Steele, R.G.D. and J.H. Torrie, 1980. Principles andProcedures of Statistics, Second Edition.McGraw-Hill, New York, 633 pp.

Teichert-Coddington, D. and B.W. Green, 1993.Tilapia yield improvement throughmaintenance of minimal oxygenconcentrations in experimentalgrow-out ponds in Honduras. Aquaculture,118:63-71.

Yi, Y., 1997. A cage-cum-pond integrated rotationculture system for fattening large Niletilapia (Oreochromis niloticus) in cagesand nursing small Nile tilapia in openponds. PhD dissertation, Asian Instituteof Technology, Bangkok, Thailand, 169 pp.

Yi, Y., C.K. Lin, and J.S. Diana, 1996. Effects ofstocking densities on growth of caged adultNile tilapia (Oreochromis niloticus) and onyield of small Nile tilapia in open water inearthen ponds. Aquaculture, 146:205-215.

a finishing system for large tilapia

Documents