Abstract.–We performed four experiments to optimize fertilization of carp aquaculture ponds by basing its rate upon weekly analysis of the N and P in ponds. For the Hubei Traditional Method, however, a constant amount of calcium superphosphate (260 mg P/L) and urea (1157 mg N/L) (N:P ratio 4.5:1 by weight) was dissolved in water and ladled into each pond every week (3 April-1 June 1999). For the Ohio Method, inorganic phosphate, ammonia, and nitrate concentrations were also measured weekly, but the ponds were fertilized with only enough phosphoric acid (42.5% H3PO4) and urea to restore phosphate to 30 mg P/L and inorganic N to 600 mg N/L (N:P =20:1). These fertilizers were dissolved in a large volume of water and sprayed evenly over the pond. Juvenile silver carp Hypophthalmichthys molitrix and bighead carp Aristichthys nobilis were stocked at an average of 6400 and 1500/ha (total of 600 kg/ha) in each pond. For Experiments 2 and 3, we stocked bigmouth buffalo Ictiobus cyprinellus juveniles and, for Experiment 4, yellow perch Perca flavescens.

Hubei Traditional Method ponds had somewhat higher average nutrient concentrations than those using the Ohio Method because more fertilizer was applied. The total cost of fertilizer was also higher: Hubei Method: 540 kg/ha superphosphate and 270 kg/ha urea (cost = Y810 (US$97.83)). The Ohio Method fertilizer cost was Y212/ha (US$25.61). For Experiment 2, the Hubei cost was again Y810, the Ohio cost was for 1.2 L/ha phosphoric acid (Y26) and 53.9 kg/ha urea (Y108) for a total cost of Y134 (US$16.18). Algal blooms and low oxygen in Hubei ponds killed carp and bigmouth buffalo, requiring algaecide and aeration treatments, but not in Ohio ponds. Overall cyprinid survival was slightly higher for Ohio ponds (Exp. 1: 94% vs. 93%; Exp. 2: 97% vs. 95%). Yield was 23% and 100% higher for the Hubei method (Exp. 1 (carp): 1065 vs. 867 kg/ha; Exp. 2 (buffalo): 1239 vs. 588 kg/ha), respectively. However, survival of juveniles of the low oxygen-intolerant yellow perch was 97% in the Ohio Method vs. 7.5% using the Hubei Traditional Method, and yields were also much higher,106 kg/ha vs. 6 kg/ha.

Introduction

The ultimate goals of pond fertilization in fish aquaculture, whether raising fingerlings for stocking or growing fish for harvesting, are to increase growth rates and to get higher production in terms of number of surviving fish and weight of those fish (McLarney 1998). Pond fertilization is meant to achieve these goals by supplementing the provision of nutrients needed for bacterial and algal growth, and hence zooplankton and benthos production. Bacteria, algae, and zooplankton form the basis of the food-web structure necessary to support growth in fish that are planktonic feeders during larval and juvenile stages of growth or as adults (Wu et al. 1997). While there is general agreement among aquaculturists about the goals of increasing fish growth rate and production through fertilization (often in combination with feed supplementation), there is little real consensus on what type of fertilization regimen is best (Mazid et al. 1997). Low N:P ratios in the fertilization regimen can encourage the growth of toxic Cyanobacteria (blue-green algae). Furthermore, there is increasing concern over the negative effects of nutrient-rich discharges from aquaculture ponds on the environment in general (Tucker et al. 1996).

Aquaculture fertilization strategies typically involve adding organic (natural) or inorganic (synthetic) fertilizers, or some combination of the two, to ponds on a regular basis. In many cases a fertilization regimen results from tradition or common usage without regard to the nitrogen and phosphorus content of the pond and whether the amounts of these elements in the fertilizer are of the appropriate amounts or ratios. Nitrogen and phosphorus content is especially difficult to control with the organic fertilization practices that dominate in carp polyculture because hay, bran, and manure nutrient content is quite variable depending on the plant or animal source and/or the animal's feed (Culver 1991), and release rates of individual nutrient elements from organic materials are not constant over time.

A better understanding of factors involved in plankton production and fish growth has been provided by recent studies of larval fish grown in Ohio hatchery ponds (Culver 1991; Culver et al.1993; Qin and Culver 1995; Qin, Culver, et al. 1995; Qin, Madon, et al. 1995; Culver 1996) where timing of pond filling, N:P ratios and application rates of fertilizers, stocking densities and organic fertilizer additions all become important factors influencing plankton dynamics and thus fish growth. Briefly, we developed a fertilization and stocking regimen based on: 1. Filling ponds immediately before stocking, 2. Spraying dilute solutions of liquid inorganic fertilizer only with no organics, 3. Fertilizing weekly based on weekly inorganic nitrogen and phosphorus analysis in each pond such that its phosphate concentration was raised to 30 :g PO4-P/l and Ammonia-N + Nitrate-N were raised to 600 :g N/l, and 4. Stocking sufficient fry to harvest 300,000 fish/ha. For these conditions and species (Percidae: Stizostedion vitreum), this meant stocking 450,000 fry/ha.

Current practices in Asian carp polyculture must be understood in a socioeconomic context to combine the advantages of both methods to provide higher fish yields at reasonable costs while minimizing the impact of aquaculture on water quality. Traditional Asian polyculture systems consist of multiple carp species that represent different, sometimes interdependent, trophic levels and are usually highly integrated with farming through use of animal, and sometimes human, waste as fertilizer. After fish are harvested, the pond sediments often serve as fertilizer for animal fodder crops or even grasses to feed carp (Li 1987). Some polyculture practices deliberately integrate other animals, such as ducks into the system. Experiments with incorporation of ducks into polyculture have resulted in increased production of fish and duck eggs where the population of ducks was 500 ha-1 (Latif et al. 1993). In the eastern provinces of China, carp polyculture with manure fertilization (e.g., Hassan, et al. 1997), often incorporating some form of animal production, is the main type of aquaculture practiced and provides a great deal of protein-rich food to many people.

A recent bioeconomic analysis of carp polyculture in seven provinces of eastern China (Chen et al. 1995) revealed that variation in inputs of feed, fertilizer and in the fish stocking model used, in combination with some geographical and socioeconomic factors, could be associated with low, medium and high productivity classes of fish yield. Stocked fish and feeds comprised the greatest proportion of costs in their study regardless of the productivity class, while fertilization costs remained relatively low. Among 1013 ponds surveyed, the greatest revenues from fish in the low and medium yield ponds were from grass carp Ctenopharyngodon idella and filter feeders, such as silver Hypophthalmichthys molotrix and bighead carp Aristichthys nobilis. Grass carp, omnivores such as common carp Cyprinus carpio, and filter feeders earned most in the high yield ponds. In socioeconomic terms, the fish stocking ratios vary such that pond yields reflect both regional consumer preferences and relatively lower input costs associated with raising grass carp and filter feeders in poorer regions as opposed to the higher cost of adding feed inputs for omnivorous and black carp Mylopharyngodon piceus in regions of higher income. Assuming that it is desirable to increase fish yields in the low to medium yield ponds mainly located in the lower income regions (Chen et al. 1995), a fertilization regimen that increases plankton production to feed filter feeders would be useful.

Recent research has examined the effects of incorporating both organic and inorganic fertilization strategies in Israeli and Asian Tilapia and carp polyculture (Schroeder et al. 1990; Milstein et al. 1995; Hassan et al. 1997; Jana and Chakrabarty 1997). Varying manure additions has also been investigated as a means to increase production in polyculture (Zhu et al. 1990; Milstein et al. 1991; Latif et al. 1993). These studies have had widely varying results, attributed to a variety of factors. However, despite the focus on varying fertilization regimens to increase fish production, none of these researchers measured variation in the levels of nitrogen and phosphorus (or N:P ratios) in pond water, nor did nutrient levels play any part in the decisions about how much of what kind of fertilizer to add. Some Asian aquaculture facilities use fixed additions of inorganic nitrogen and phosphorus fertilizers, but because aquaculture ponds typically have been used for a varying number of years with heavy fertilization, nutrient concentrations (particularly phosphorus) vary widely between ponds even without any fertilizer addition.

In this study, therefore, our study objective was to compare the growth and survival of three cyprinid species and one percid species in ponds to which a fixed amount of inorganic fertilizer was added each week with those for which nutrient concentration targets were determined ahead of time and the amounts of N and P to be added were based on weekly direct measurement of inorganic N and P present in each pond. This objective was addressed via a series of four sets of experiments involving: 1) two filter-feeding carp species grown together in clay-bottomed ponds, 2) a single filter-feeding cyprinid grown in clay-bottomed ponds, 3) a single filter-feeding cyprinid grown in concrete ponds, and 4) a single percid species sensitive to low oxygen concentration grown in concrete ponds.

Methods

We performed two experiments in a series of four clay-bottomed ponds at a fish farm in Dunkou, Hubei Province, People's Republic of China, and two additional experiments in concrete pools at Hubei Province Fisheries Research Institute’s East Lake fish farm, Wuhan, P. R. C. The Dunkou ponds are 1.2 m deep and were filled with water from the Swan River. The areas of ponds 2, 3, 5, and 9, are 733, 667, 667, and 600 m2, respectively. We measured inorganic phosphate, ammonia, and nitrate concentrations weekly in each pond/pool using standard methods (APHA et al. 1980), including the acid molybdate spectrophotometric method for phosphate, and ion-specific electrodes for ammonia (Orion Model #9512BN) and nitrate (Orion Model #9507BN). For the Hubei Traditional Fertilization Method in Experiment 1, a constant amount of fertilizer was added each week irrespective of the amount of nitrogen and phosphorus present in the water. Specifically, we dissolved calcium superphosphate (60 kg/ha, equivalent to 260 mg P/L) and urea (30 kg/ha, equiv. to 1157 mg N/L) (N:P ratio 4.5:1 by weight) in water and ladled it into ponds #2 and #3 weekly (3 April - 1 June 1999). For the Ohio Method (Culver 1996), we used the results of each week’s nutrient analyses to determine how much phosphoric acid (42.5% H3PO4) and urea to add to restore the phosphate concentrations in ponds #5 and #9 to 30 mg P/L and the inorganic N to 600 mg N/L (N:P =20:1 by weight). For this method, we dissolved the appropriate amount of each fertilizer in a large volume of pond water (100:1) and sprayed it evenly over each pond. The ponds do not stay at an inorganic N:P ratio of 20:1, of course, because algae and bacteria begin assimilating nutrients (especially phosphate) as soon as they are added. We are simply restoring the phosphate to 30 mg P/L and the N:P ratio to at least 20:1 once per week.

For the first experiment, we stocked juvenile silver carp Hypophthalmichthys molitrix and bighead carp Aristichthys nobilis at an average density of 6,400 and 1,500/ha, respectively (total of 600 kg/ha) (Table 1). For Experiment 2 (26 July - 4 August), we reversed the ponds used for the Hubei Traditional Method with those used for the Ohio Method, and stocked them with bigmouth buffalo Ictiobus cyprinellus juveniles (Total Length=2.7 cm; weight = 0.32g; 22,500/ha). Experiment 3 (6 July - 30 August 1999) was performed in four concrete pools (area = 167 m2) at the East Lake fish farm, stocking 500 bigmouth buffalo juveniles (1.08 g wet weight) per pool (30,000/ha). Experiment 4 (25 July - 25 August 2000) was performed in slightly smaller concrete pools (area = 146 m2) at East Lake fish farm, and involved stocking 200 juvenile yellow perch Perca flavescens (1.01 g wet weight) per pool (13,700/ha). Juvenile yellow perch have been successfully cultured for seven years at the St. Mary’s State Fish Hatchery, St. Mary’s, Ohio, using the Ohio Method of pond fertilization. The first three Chinese experiments thus involved comparing the growth and survival of three cyprinid species (two as polyculture) using the Hubei Traditional Method of fertilization with those using the much lower amount of fertilization used in Ohio. We then completed the comparisons of the fertilizer treatments by measuring the survival and growth of yellow perch, a member of the family Percidae, a taxon much less tolerant of low water quality.

Results and Discussion

Phosphate concentrations during Experiment 1 were typically higher in the Hubei Traditional Method ponds (Mean for pond #2 = 9.9 :g P/L, #3 = 8.6 :g P/L) than those fertilized using the Ohio Method (Mean for pond #5 = 5.7 :g P/L, #9 = 5.5 :g P/L) (Figure 1), although the differences were not statistically significant. A constant amount of calcium superphosphate fertilizer was added every week irrespective of the phosphate concentration in the pond for the Hubei Traditional Method (60 kg?ha-1?wk-1, equivalent to 262 :g P?L-1?wk-1) for all four experiments, whereas the Ohio Method simply returns the concentration to 30 :g P/L weekly as required. Hence the Hubei Traditional Method added at least 262/30 = 8.7 times as much phosphate as in the Ohio Method. Phosphate concentrations varied widely, with the highest values occurring in the fourth and fifth weeks of the experiment (Figure 1), so less phosphate fertilizer was added these weeks for the Ohio Method ponds (Table 1). We have observed similar variation in phosphate in Ohio (Qin and Culver 1995) and Australia (Culver and Geddes 1993). Algal blooms occurred in the Hubei Traditional Method ponds for Experiment 1, requiring aeration and algaecide (0.5 mg/L CuSO4) treatment, but not in ponds treated with the Ohio Method.

Nitrate and ammonia concentrations were also typically higher in Experiment 1 in the Hubei Traditional Method ponds (Figures 2 and 3), although analytical problems prevented us from obtaining accurate nitrate concentrations for the ponds for 4 of 9 weeks. Ammonia concentrations occasionally were high enough to harm or kill fish if the pH were high enough (>9.5) to convert ammonium ions to free ammonia. We collected water in the morning and measured pH values upon return to the laboratory (2 hr away), occasionally observing high values, particularly in pond 2 (Figure 4). Local pH values in the field (particularly in late afternoon) may have greatly exceeded those we measured in the lab.

The amounts of urea added in the Hubei Traditional Method (30 kg?ha-1?wk-1, equivalent to 1,166 :g N?L-1?wk-1) was 1.6 times the average amount added in the Ohio Method (16.15 kg?ha-1?wk-1, equivalent to 753 :g N?L-1?wk-1), but note that for Experiment 1, pond #9 received, on average, 1.8 times as much as pond #5 (Table 1), because chemical analyses indicated that it occasionally had less inorganic nitrogen than pond #5. The objective of the Ohio Method is to bring all of the ponds to a common standard concentration once per week, although uptake by some organisms (aquatic vascular plants, algae, and bacteria) and excretion by others (fish, plankton, benthos, and bacteria) rapidly altered the concentrations immediately after each week’s fertilization. Our approach could have been made even more pond-specific by carefully measuring the actual volume of water in each pond at the water level used in the experiment and incorporating it into the calculation of fertilizer added to each pond each week. Instead, we took the average depth of all ponds for Experiments 1 and 2 to be 1.2 m and the average area to be 667 m2.

The ratio of N:P added in the Hubei Traditional Method was 1166/262 = 4.45:1 by weight (9.85:1 by moles), but the actual N:P ratio in the pond after fertilization depended upon the initial inorganic nitrogen and phosphate concentrations, and was as low as 3.7:1 on 17 May in pond #3 (Figure 5). In contrast, the Ohio Method restores the inorganic N:P ratio to 20:1 by weight (44.3:1 by moles) once per week. Maintaining a high N:P ratio in fish ponds is advantageous, because a low N:P ratio favors nitrogen-fixing Cyanobacteria, many of which are toxic to zooplankton, fish, benthic invertebrates, and humans.

Calcium superphosphate fertilizer is relatively insoluble, and the fertilizer was ladled into the ponds, so much of the phosphorus used in the Hubei Traditional Method may have ended up sitting on the bottom of the pond. For this reason, the Ohio Method calls for use of concentrated phosphoric acid (85% H3PO4 by weight, 453 g P/L, specific gravity = 1.685 kg/L), diluted 1:100 with pond water and sprayed all over the pond. In China, we used 50% strength phosphoric acid (268.8 g P/L), which cost Y22/L (US$2.66/L). At the time of our experiments, calcium superphosphate (5.2% P) cost Y0.5/kg (US$0.06/kg). The cost for phosphorus for the Hubei Traditional Method, was thus 9 applications x 60 kg/ha/wk x Y0.5/kg = Y270 (US$32.61). Phosphoric acid applications averaged 8.33 L/ha x Y22./L ?= Y183 over the season. Urea (46.3% N) is very soluble, and relatively inexpensive (Y2.0/kg or US$0.24/kg). The Hubei Traditional Method used 30 kg urea?ha-1?wk-1 x 9 wk x Y2.0/kg = Y540 ($65.22). The Ohio Method used an average of 16.16 kg urea/ha over the whole season at a cost of Y32.3 (US$3.90). Total fertilizer cost per hectare for Experiment 1 for the Hubei Traditional Method was Y810 (US$97.83). The Ohio Method fertilizer cost for Experiment 1 was Y215/ha (US$25.97). For Experiment 2, the Hubei Method cost was again Y810/ha, the Ohio Method cost was for 1.2 L/ha phosphoric acid (Y26) and 53.9 kg/ha urea (Y108), for a total cost of Y134 (US$16.18). Relative costs for the other two experiments were similar.

For experiment 1, we stocked a combination of silver and bighead carp on 14 April 1999 at a density of 7,500 to 8,184 fish/ha in the ponds and harvested them after 48 days (1 June 1999). Survival was excellent in both treatments (Table 2), as was growth, with the Hubei Traditional Method producing higher biomass yield. For experiment 2, juvenile (0.32 g) bigmouth buffalo stocked on 21 June grew rapidly, reaching an average of 55 g (Hubei Method) and 28 g (Ohio Method) by 18 August (58 days) (Table 3). In this case, there was a large disparity in growth rates between the two methods, but the experiment was complicated by a flood that overflowed the ponds on 19 July, causing mortality in some ponds, particularly in pond #5 (Hubei Method). We counted 60, 2, 1, and 3 dead fish in ponds #5, #9, #2, and #3, respectively. Therefore, we restocked all ponds on 26 July without draining them. Therefore, we do not know whether the differences in mean sizes between Hubei Method and Ohio Method ponds on 18 August were due to differences in overall growth rate or to greater survival of smaller fish in the Ohio Method after the second stocking on 26 July.

In the two additional experiments performed in concrete pools, we studied growth and survival in bigmouth buffalo and yellow perch Perca flavescens. In the first concrete pool experiment (Experiment 3), bigmouth buffalo size at harvest for the Hubei Traditional Method averaged almost twice as much for the Ohio Method (Table 4), similar to what was seen in Experiment 2. Unfortunately, we again do not have data on the number of fish alive in each pool at the end of the experiment, so we cannot determine whether the growth rates of bigmouth buffalo were indeed twice as fast in the Traditional Method or whether the difference in individual final fish weights between the two treatments was in part due to a larger number of fish surviving in the Ohio Method.

The final experiment (Experiment 4) was performed with juvenile yellow perch, a species that is more sensitive to low oxygen concentrations than are the cyprinids used in experiments 1-3. Survival in pools using the Ohio Method was much higher than in those fertilized according to the Hubei Traditional Method (Table 5). No fish in Pool #2 survived the 31 day culture period, and survival in Pool #3 was only 15%. Survival and growth were excellent in pools treated with the Ohio Method.

Conclusions

The Hubei Traditional Method dates from the 1970s, when the price of carp was very high due to insufficient supply relative to demand. The cost of fertilizer was not a significant issue at the time, and the potential for pollution of the environment from excessive fertilization was not a consideration. Currently, the market value of silver carp and bighead carp is lower than before, because supplies exceed demand, and protection of the environment has a much higher priority. Unfortunately, the Hubei Traditional Method adds excessive fertilizer (over eight times as much phosphorus and almost twice as much urea as in the Ohio Method) and ignores nutrients already in the pond water and those coming from the sediments and fish excreta, probably leaving photosynthesis and algal growth more limited by light than by nutrients. With added aeration and algaecides, cyprinids can usually survive these conditions and grow well. In fact, cyprinids consistently grew to a larger size with the higher fertilizer levels present with the Hubei Traditional Method than ponds fertilized by the Ohio Method, suggesting they were able to use the higher algal biomass present in the Hubei Traditional Method ponds. Nevertheless, aeration and algaecide applications were required, suggesting that too much fertilizer was applied, at a cost of four to six times that of the Ohio Method. Therefore, subsequent research should experimentally identify a higher fertilizer target (still adjusted to the current nitrogen and phosphorus concentrations in individual ponds) that would optimize carp growth without creating the nuisance blooms of Microcystis and other toxic algae and the night-time oxygen concentration declines characteristic of over-fertilized ponds. We anticipate that there would be a large benefit from decreasing the amount of phosphate added, thus raising the N:P ratio in the ponds. We are not suggesting that every fish farmer needs a spectrophotometer and ion-specific electrodes for nutrient analysis in order to grow cyprinids, but the farmer does need a more effective fertilizer regimen to use. Development of a Hubei New Method will provide significant savings in the cost of fertilizers, while achieving more consistent survival and growth of cultured fish. Moreover, the Hubei Traditional Method is often applied to large reservoirs and lakes, requiring a very large amount of fertilizer, trucks, and manpower for each fertilizer application. The Ohio Method requires much less fertilizer, while producing only slightly lower concentrations of nutrients in the water (Figures 1-3). The fertilizer is applied with a sprayer that more conveniently and effectively distributes the nitrogen and fertilizer in the productive zone of the lake, minimizing the amount that sinks to the bottom.

Polyculture has always been advocated in China, but the focus now has moved from increasing the mass of low-value cyprinid fish produced to production of non-traditional species that have a high economic value. Unfortunately, these species typically have much lower tolerance for the low oxygen and high pH often associated with silver carp and bighead carp production. The nutrient targets for the Ohio Method, on the other hand, were developed for culturing juvenile stages of just such sensitive species, namely the walleye Stizostedion vitreum vitreum, a species closely related to yellow perch and equally sensitive to low water quality. The results of Experiment 4 illustrate the low tolerance of such species for the high algal density associated with the Hubei Traditional Method. Furthermore, discharge of culture water from a Hubei New Method ponds will contribute less to eutrophication of lakes and rivers because the nutrient concentrations in the ponds would be kept at the minimum level needed to maintain appropriate algal growth. Should culture of juveniles of high-value, non-cyprinid fish (e.g. Serranidae: Siniperca chuatsi) be desired, the Hubei New Method may also provide significant benefits through its maintenance of higher water quality in ponds at low cost.

References

APHA et al. (American Public Health Association, American Water Works Association, and Water Pollution Control Federation). 1980. Standard methods for the examination of water and waste water, 15th ed. American Public Health Association, Washington , D.C.

Buck, H. D., R. J. Baur, and C. R. Rose. 1978. Utilization of swine manure in a polyculture of Asian and North American Fishes. Transactions of the American Fisheries Society 107(1): 216-222.

Bukvic, I., M. Kerovec, A. Plenkovic, and M. Mrakovcic. ?1998. Impact of silver and bighead carp (Cyprinidae) on plankton and water quality in fish ponds. Biologia 53(2): 145-157.

Chen, H., B. Hu, and A. T. Charles. 1995. Chinese integrated fish farming: a comparative bioeconomic analysis. Aquaculture Research 26: 81-94.

Culver, D. A. 1991. Effects of the N:P ratio in fertilizer for fish hatchery ponds. Verhandlungen Internationale Vereinigung für Theoretische und Angewandte Limnologie 24:1503-1507.

Culver, D. A. 1996. Fertilization procedures for pond culture of walleye and saugeye. Pages 115-122 In R. C. Summerfelt, ed. Walleye Culture Manual. NCRAC Culture Series 101. North Central Regional Aquaculture Center, USDA, Iowa State University, Ames.

Culver, D. A. and M. C. Geddes. 1993. Limnology of Australian larval fish culture ponds: relationships among water quality, phytoplankton, zooplankton, and the growth of larval fish. Australian Journal of Marine and Freshwater Research 44:537-551.

Culver, D. A., S. P. Madon, and J. Qin. 1993. Percid pond production techniques: Timing, enrichment, and stocking density manipulation. Journal of Applied Aquaculture 2: 9-31.

Gu, B. h., D. M. Schell, X. H. Huang, and F. L. Yie. 1996. Stable isotope evidence for dietary overlap between two planktivorous fishes in aquaculture ponds. Canadian Journal of Fisheries and Aquatic Sciences 53: 2814-2818.

Hassan, S., P. Edwards and D.C. Little. 1997. Comparison of tilapia monoculture and carp polyculture in fertilized earthen ponds. Journal of the World Aquaculture Society 28(3): 268-274.

Jana, B. B. and D. Chakrabarty. 1997. Relative status and contribution of sediment phosphorus and nitrogen in carp culture system fertilized with various combinations of rock phosphate. Aquaculture Research 28: 247-255.

Li, S. 1987. Energy structure and efficiency of a typical Chinese integrated farm. Aquaculture 65: 105-118.

Mazid, M. A., M. Zaher, N. N. Begum, M. Z. Ali, and F. Nahar. 1997. Formulation of cost-effective feeds from locally available ingredients for carp polyculture system for increased production. Aquaculture 151(1-4): 71-78.

McLarney, W. 1998. Freshwater Aquaculture. Hartley and Marks Publishers, Point Roberts, WA.

Milstein, A., A. Alkon, Y. Avnimelech, M. Kochba, G. Hulata, and G. Schroeder. 1991. Effects of manuring rate on ecology and fish performance in polyculture ponds. Aquaculture 96: 119-138.

Milstein, A., A. Alkon, I. Karplus, M. Kochba, and Y. Avnimelech. 1995. Combined effects of fertilization rate, manuring and feed pellet application on fish performance and water quality in polyculture ponds. Aquaculture Research 26: 55-65.

Muhammad, A. L., J.A. Mohammad, and M.A. Rahman. 1993. Integrated duck-cum-fish farming in Bangladesh. Journal of the World Aquaculture Society 24(3): 402-409.

Qin, J. and D. A. Culver. 1995. Separate and combined effects of larval walleye and fertilization on plankton community structure in enclosures. Journal of Freshwater Ecology 10(4):385-391.

Qin, J., S. P. Madon, and D. A. Culver. 1995. Effect of larval walleye (Stizostedion vitreum) and fertilization on the plankton community: Implications for larval fish culture. Aquaculture 130:51-65.

Qin, J., D. A. Culver, and N. Yu. 1995. Effect of organic fertilizer on heterotrophs and autotrophs: Implications for water quality management. Aquaculture Research 26:911-920.

Schroeder, G. L., G. Wohlfarth, A. Alkon, A. Halevy, and H. Kreuger. 1990. The dominance of algal-based food webs in fish ponds receiving chemical fertilizers plus organic fertilizers. Aquaculture 86: 219-229.

Shei, P., W. L. Lin, S. M. Wang, and J. K. Liu. 1993. Plankton and seston structure in a shallow, eutrophic subtropical Chinese lake. Archiv für Hydrobiologie 129(2): 199-220.

Tucker, C. S., S. K. Kingsbury, J. W. Pote, and C. L. Wax. 1996. Effects of water management practices on discharge of nutrients and organic matter from channel catfish (Ictalurus punctatus) ponds. Aquaculture 147: 57-69.

Wu, L., P. Xie, M. Dai, and J. Wang. 1997. Effects of silver carp density on zooplankton and water quality: Implications for eutrophic lakes in China. Journal of Freshwarter Ecology 12(3):437-444.

Xie, P. 2001. Gut contents of bighead carp (Aristichthys nobilis) and the processing and digestion of algal cells in the alimentary canal. Aquaculture 195(1-2):149-161.

Xie, P. and Y. Yang. 2000. Long-term changes of Copepoda community (1957-1996) in a subtropical Chinese lake stocked densely with planktivorous filter-feeding silver and bighead carp. Journal of Plankton Research 22(9): 1757-1778.

Zhu, Y., Y. Yang, J. Wan, D. Hua, and J. A. Mathias. 1990. The effect of manure application rate and frequency upon fish yield in integrated fish ponds. Aquaculture 91:233-251.