zyxwvut zyx zyx zyxwvut zyxwvuts Vol. 25, No. 2 June, 1994 JOURNAL OF THE WORLD AQUACULTURE SOCIETY The Nitrogen Budget of a Tropical Semi-Intensive Freshwater Fish Culture Pond MARCOv. ACOSTA-NASSAR,' JULIO M. MORELL AND JORGE E. CORREDOR' University of Puerto Rico. Department of Marine Sciences, Mayagiiez. Puerto Rico 00680 Abstract Nitrogen inputs, outputs and compartamentalization were quantified in a freshwater fish pond stocked with hybrid Oreochromis throughout a production cycle. The budget accounts for 91% of the nitrogen added to the system. Feed addition accounted for 87% of the nitrogen input and an additional 11% was attributable to nitrogen fixation, mainly in the water column. The balance of the nitrogen input was contained in the source water for the pond. Commercial-size fish accumulated 17.5% of the nitrogen added to the system. Most of the nitrogen was eventually deposited in the sediments. Nitrification constituted a major pathway for nitrogen transformation, but only 1% of the nitrogen input was lost through denitrification. Optimization of nitrogen conversion to fish protein in intensive aquaculture requires a detailed knowledge of the dynamics of microbial nitrogen transformations. High fish yields in current aquacultural practice are obtained through the addition of large quantities of high-protein feed. However, such management schemes run the risk of exceeding the carrying capacity of the system resulting in degradation of the water quality required for successful fish production. Rational management procedures, therefore, require detailed knowledge of the capacity and limitations of sediment and water column microbial communities to either make excess nitrogen entering the system available to the fish or to remove toxic species such as ammonium and nitrites. Recently, work has begun on the establishment of detailed nitrogen budgets for aquaculture (Parnas and Lahav 1972; Avnimelech and Lacher 1979; Boyd 1985; Krom et al. 1985; Schroeder 1987; Daniels and Boyd 1989). In many instances, however, little attention has been given to the direct experimental assessment of microbial nitrogen transformations, principally nitro- gen fixation, nitrification and denitrification. Experimental determinations for a tropical semi-intensive culture pond in Puerto Rico are reported here and applied to an overall nitrogen budget for the pond. Materials and Methods The experimental earthen fishpond studied is located at the Aquaculture Experimental Station, University of Puerto Rico, Department of Marine Sciences, in Lajas, Puerto Rico. The surface area of the pond is 1,354 m2. Its depth averaged 1.02 m and ranged from 1.20 m at the standing drain pipe edge to 0.89 m at the inflow water pipe edge. The pond was emptied on 29 May 1991 and left exposed to the air for 10 d. The pond was then refilled with water and stocked with 1,696 red tilapia hybrid fingerlings (Oreochromis mossambicus x Oreochromis niloticus) of 9.4 g average weight. Fish were fed a diet of sinking pellets certified by the manufacturer to contain 30% crude protein 6 d w k at approximately 3% of body weight adjusted every month. Pond water level was maintained constant and the water inflow flow rate was 22.7 Wmin which resulted in a replacement rate of 0.70 by month. Pond overflow was avoided at all times. The earthen pond was divided into three zones located at the standing drain pipe sec- zyxwv zyxwvutsrq ' Present Address: Laboratorio de Oceanografia y Manejo Costero, San Jost 1200, Costa R i a . Corresponding author. 8 Copyrishi by the World Aquaculture Society 1994 26 1 262 zyxwvuts zyxwvu ACOSTA-NASSAR tion, the central section and the inflow water pipe section. The water column was sampled at the top 10 cm layer and at the water/ sediment interface using a 5 L van Dorne bottle. Cylindrical cores (35 cm2 in area and 20 cm in depth) were taken from sediment using plastic core liners capped with a nonreturn valve in each of the three zones. Monthly data from each of the three zones were averaged in developing the budget. In the laboratory, sediment cores were sectioned vertically at 2 cm intervals to 10 cm depth and porewaters were extruded by centrifugation. Adsorbed ammonium sediment content was estimated by extracting sediment-bound ammonium with KCl. An amount of 3M KCl was mixed with equal weight of fresh sediment and allowed to stand for 0.25 to 0.50 h. The supernatant was used for ammonium analysis. Water column and porewater samples were analyzed for ammonium, nitrate and nitrite. Nitrite and nitrate (reduced to nitrite by passing through a cadmium column) were determined by diazotization through flow injection analysis (FIA) using the protocols established by Whitledge et al. (1981) and a LACHAT two-channel FIA instrument. Ammonium analysis was performed using the phenol/hypochlorite method via FIA. Porosity of replicate 2 cm sediment segments was estimated by the ratio of wet to dry weight. Nitrogen mineralization rates in the sediment were estimated from NH4+diffision in intact cores (Corredor and Morel1 1989). Duplicate cores were taken as described above. Water overlying the cores was adjusted to pre-determined volume (300-450 ml). Overlying waters were then monitored for ammonium accumulation using the FIA procedure at intervals of up to 24 h. Ammonium inputs due to fish excretion were calculated using the expression of Melard ( 1 986): monium-nitrogen excreted per unit time and P is the weight of the fish. Nitrogen fixation in the pond water, sedimentlwater interface, and sediment was measured using the acetylene reduction technique (Stewart et al. 1968). Rubbercapped test tubes (1 2 ml) were used for incubation of the samples. Samples from the water column and benthic zone were incubated in situ for approximately 72 h using 2 ml of acetylene and 6 ml of sample. Sediment nitrogen fixation assays were performed using 3 ml of sediment dispensed directly into the tubes. After flushing with gaseous nitrogen for 3 min, samples were treated with 6 ml of acetylene and incubated in darkness for 72 h. Headspace samples were withdrawn from the tubes by syringe (250 pl) and analyzed for ethylene accumulation using a Hewlett-Packard 5890A gas chromatograph equipped with hydrogen flame ionization detector and a Porapak R column. High purity helium served as the camer gas at a flow rate of 30 ml/min. The flow rates of compressed air and hydrogen were 300 and 30 ml/min. Temperatures were kept at 40 C in the column, 175 C in the injector and 300 C in the detector. Calibration was achieved with 100 ppm ethylene in nitrogen (Alltech Assoc.). Total gas evolved was estimated using the solubility coefficients of Wilheim et al. (1977). Calculations of nitrogen fixation were based on the relationship of three acetylene molecules reduced for every nitrogen molecule. Denitrification rates were quantified using the acetylene block technique (Taylor 1983; Corredor and Capone 1985). Samples for sediment denitrification assays were collected and manipulated as for sediment nitrogen fixation analysis. Sediment samples (3 ml) were treated with 6 ml of acetylene and incubated for 2 h in the darkness. Headspace samples were withdrawn from the tubes by syringe (500 pl) and analyzed for nitrous oxide accumulation using a HewP-NH4+-N(mgh) = 0.345 x P (g)0.557(1) lett-Packard 5890A gas chromatograph equipped with a Porapak Q column and a Where P-NH4+-N is the weight of am- 63Nielectron-capture detector (ECD). A 5% zyxwvu zyxwvu zyxwv 263 NITROGEN BUDGET OF A FISH POND methane/argon gas mix at a flow rate of 19 ml/min was used as camer gas. Calibration of the instrument was achieved with a 100 ppm nitrous oxide standard (Alltech Assoc.). Routine checks for ECD efficiency were performed using air at a nominal concentration of 3 10 ppb (Oudot et al. 1990). Acetylene contamination of the detector was precluded with a 6 port valve assembly activated pneumatically to vent column effluent after elution of nitrous oxide. Total gas evolved was estimated using the coefficients of Weiss and Price (1980). Water column denitrification rate determinations were performed when oxygen in the water column was equal to or lower than 1 mg/L. This situation occurred only at dawn in the last month of culture. Oxygen concentration was measured using an oxygen electrode (YSI). For these analyses, samples were incubated in situ until oxygen in the water column increased above 1 mg/L (2 h approximately). These analyses were performed using 6 ml of sample and 3 ml of acetylene. A chlorate block (Belser and Mays 1980; Taylor 1983), in which the oxidation of nitrite to nitrate is inhibited by added potassium chlorate and nitrite is accumulated, was applied to slurry and water sample incubations to determine nitrification rates (Henriksen et al. 1981; Corredor and Capone 1985). For the sedimedwater interface, the overlying sediments from sample cores down to approximately 0.5 cm were removed by gentle aspiration and transferred to 125 ml Erlenmeyer flasks. These slumes were then sparged with air for 15 min to eliminate volatile reducing compounds and establish oxic conditions. The sedimentswere allowed to settle in the flasks, the overlying water was removed by aspiration and replaced with 50 ml pond water amended with potassium chlorate at 20 mM. The flasks were covered with perforated stoppers to allow gas exchange and incubated in the dark in a reciprocating shaker at ambient temperature. Samples of the overlying water were taken periodically and analyzed for nitrite. A similar procedure was implemented for water column rate determinations except that water samples were directly added to the flasks, amended with potassium chlorate and incubated as described. Secchi disk readings were taken at monthly intervals. This information was used to estimate chlorophyll contents using the Secchi depth to chlorophyll a relationship established by Ayarza (1988) for these ponds. The nitrogen content of the phytoplankton was estimated using the chlorophyll values thus derived and the chlorophyll to nitrogen conversion values of Parsons et al. (1984). Total organic nitrogen in the fish tissue was estimated by acid titration following Kjeldahl digestion and ammonia distillation. zy zyx Results Gross fish yield during the production cycle amounted to 2,786 kg/ha. Fish nitrogen content was 10.3% (n = 6) at the beginning and 9.1% (n = 6) at the end of the production cycle. Mortality during the production cycle was of 50%. The bulk of this mortality is attributable to predation by cattle egrets. Variations in the water quality parameters throughout the study period are presented in Fig. 1. Maximum concentrations of dissolved inorganic nitrogen in the pond were well below those reported as toxic to commercial fish species (Meade 1985; Lewis and Morris 1986; Daud et al. 1988; Russo and Thurston 1991). Only a moderate variation was observed during the final months of culture. Maximum ammonium values were on the order of 5 pM. Nitrate and nitrite were consistently below 3 pM. A transient peak of nitrate enrichment in the supply water was observed during the fourth month which coincided with a period of heavy rainfall. Ammonium concentrations in porewaters were consistently one order of magnitude or greater than those of nitrate and nitrite (Fig. 2). Temporal and spatial variations in porewater ammonium were minimal. Nitrate and nitrite concentrations, on 264 I zyxwv zyxwvu zyx zy ACOSTA-NASSAR -.-NH4 -.- T v c 20- .cI 15- 0 c Q) 0 c 10- 5 5- zyx NH4 h zyxwvutsrqpon zyxwv - cJ Jul Aug Sept Oct Nov FIGURE 1. Temporal variations of inorganic nitrogenous species concentrations in the water column (a) and in the source water (b). Bars represent standard error; n = 6. the other hand, exhibited a consistent decrease during the initial months and remained at low levels thereafter. KC1-extractable ammonium exhibited a moderate increase over the initial 3 mo after which the values stabilized at about 100 mmole N/m2, possibly as a consequence of saturation of the cation exchange capacity of the sediment clays (Mikkelsen 1987). Levels of KCl-extractable ammonium were at all times approximately one order of magnitude greater than the respective porewater values (Fig. 2). Organic nitrogen in the sediment increased consistentlythroughout the experiment and was, by far, the most common nitrogen species present in the sediment (Fig. 2). Whole-pond values for these variables, integrated to 10 cm depth, are presented in Table 1. Nitrification in the water column was uniformly high during the first 4 mo but no nitrification occurred during the final month. Nitrification at the sediment/water interface was substantially lower than that in the water column but followed a similar pattern (Table 2). These estimates are in the range of values reported for summertime activity by Berounsky and Nixon (1990) in an eutrophic estuary and are well below those reported by Diab et al. (1 992) in fish culture ponds. Diffusionof mineralized ammonium fiom the sediments was low during the initial months and increased thereafter as the sediments became enriched in organic matter (Table 3). Time courses for nitrogen fixation and denitrification are presented in Fig. 3. Nitrogen fixation in the water column was found to be highly variable. Highest rates (4 mmole N/m2/d) were observed to occur during the fourth month. Interestingly, this activity peak coincides with peaks in dissolved ammonium and nitrate concentrations suggesting that feedback inhibition on nitrogenase is not active at these levels. Benthic rates were substantially lower than those observed in the water column. Denitrification rates in the sediment were high (200300 pmole N/m2/d) during the first month of the study and decreased logarithmically thereafter coincident with the disappearance of nitrate and nitrite from the porewaters to a final rate of 5 pmole N/m2/d. Denitrification was observed in the water column only during the fifth month and then only during sporadic nocturnal episodes of anoxia. Rates at these times were of 5 pmole N/m2/d. zyxw zyxwvut zyxwv zyxw zyxwvutsrqp 265 NITROGEN BUDGET OF A FISH POND - NH4 NO3 NO2 6 (4 T (c) 0.M 125100- T 2 2 0.15 b 0.10 zyxw zyxwv zyxwvu zyxwvutsrq .-0c 75. 0 50- 0.05 25- 1 0 Jul Aug Sept Oct 0.W 1 Nov FIGURE 2. Temporal variations of inorganic nitrogenous species concentrations in the porewaters (a), adsorbed ammonium in the sediments (b) and particulate organic nitrogen in the sediments (c). Bars represent standard error; n = 6. Phytoplankton biomass, as reflected by reduction in Secchi disk visibility, increased throughout the experimental period (Fig, 4). Computed phytoplankton nitrogen content ranged from 10.3 mmole N/m2 at the beginning of the study to 28.2 mmole N/m2 at the end. A budget for nitrogen inputs to the pond over the course of the experiment is presented in Table 4. A total input to the pond of 3287 mole N was observed ofwhich 87% was attributable to the feed supply. Similar results have been reported elsewhere (Parnas and Lahav 1972; Avnimelech and Lacher 1979; Boyd 1985; Daniels and Boyd 1989; Krom and Neori 1989). Nitrogen fixation accounted for 1 1Yo of the input. Rates reported here are lower than those observed in manure-fertilized ponds (Kwei Lin et al. 1988). In contrast to these observations, other authors have reported no nitrogenase activity in the sediments of fish culture ponds (El Samra and 016h 1979) attributing this lack of activity to feedback inhibition. Nevertheless, studies in adjacent sedimentary marine environments indicate that effective inhibition occurs only in the presence of porewater ammonium con- centrations greater than 300 pM (Nieves and Corredor 1987); well above those reported here. The activity values reported here are similar to those quoted by Howarth et al. (1 988) from sediments of eutrophic lakes. Only 2% of the nitrogen input is attributable to the water supply, an observation similar to that reported for other semi-intensive culture ponds (Daniels and Boyd 1989). Fate of the nitrogen added to the pond is detailed in Table 5. Major nitrogen reservoirs considered are commercial-size fish, young recruits, dissolved inorganic nitrogen in the water column, phytoplankton biomass, adsorbed (KC1-extractable) ammonium in the sediments, porewater inorganic nitrogen, and sediment organic nitrogen. In addition to the reservoirs, major losses are considered. The budget accounts for 3003 moles of nitrogen or 9 19'0 of the gross input. TABLE 1. Temporaldistribution of nitrogenous species (mrnolesN/m2)integrated to 10 cm depth in the pond sediments (N = 6; SE = standard error). Jul. Aua. Seot. Oct. Nov. 8.60 4.40 0.24 0.02 0.32 0.02 6.00 2.70 0.08 0.01 0.14 0.04 7.70 3.70 0.06 0.01 0.04 0.02 7.80 1.90 0.04 0.01 0.07 0.02 6.50 1.14 0.07 0.02 0.08 0.05 Adsorbed ammonium (KCI extractable) Total N 61 75 80 111 SE 36 24 28 12 107 10 Reservoir Porewater Ammonium SE Nitrite SE Nitrate SE Organic nitrogen Total N 4,277 3,728 3,463 2,364 5,906 SE 1,137 877 3,463 556 899 266 zyxwv zyxwvuts zyxwv zyxwvu zyxwvutsrq zyxwvut zyxwvu zy ACOSTA-NASSAR 150 0 ' N E 100 250 f 1.o 0.5 0.0 &Jul Aug oSept Oct Nov FIGURE 3. Temporal variations of process rates for nitrogen firation in the water column (a); nitrogen jixation in the sediments (b) and denitrification in the sediment (c). Bars represent standard error: n = 6. Additional losses that might account for the balance include predation by egrets, ammonia volatilization and seepage. Denitrification rates reported here are lower than those observed in eutrophic lakes (Seitzinger 1988). Low overall denitrification rates observed in this study are probably the result of the high organic content and consequent reducing nature of the sediments which impede effective benthic nitrification. Tight coupling between nitrification and denitrification assuring an adequate supply of nitrate is a prerequisite for active benthic denitrification (Knowles 1982; Mark and Kemp 1984). Discussion Commercial-size fish incorporated 17.5% of the gross nitrogen input; a value comparable to those reported by other authors (Krom et al. 1985; Krom and Neon 1989; Boyd 1985; Daniels and Boyd 1989). Should the young recruits be of collateral benefit, the recoverable fraction would increase to 20% of the nitrogen input. The water column and plankton communities did not constitute a significant nitrogen reservoir in the experiment. In contrast, Krom et al. (1 985) have reported accumulation of between 20 and 70% of total nitrogen inputs in this compartment of experimental marine fish ponds. Sediment clays accumulated approximately 2% of the total gross nitrogen input in the form of adsorbed ammonium (KC1-extractable). These values are substantial but are nevertheless lower than those cited by van Rijn et al. (1984) and Diab and Shilo (1986). Organic nitrogen accumulated in the sediment constituted the greatest single nitrogen reservoir in the experimental system having accumulated 65% of the gross input. Although Schroeder (1987) remarked upon the role of fish pond sediments as a significant nitrogen reservoir, his results do not allow for discrimination between organic nitrogen on the one hand and inorganic pore water nitrogen and adsorbed ammonium on the other. A major question consequently arises as to the nature of the organic matter TABLE 2. Nitrification rates in the frrhpond (mmole N/m2/d; N = 6). Water column SE Waterkdiment interface SE Jul. Aug. Sept. Oct. Nov. 10.8 4.1 10.6 5.3 0.00 50 26 26 21 6.8 2.8 66 45 12.8 5.6 34 23 TABLE 3. Rates of diffiion of ammonium (pmole N/m2/d) from the sediment to the water column of thefihpond (N = 3: n.a. = not available). ~______ 34 23 SE Jul. Aug. Sept. Oct. Nov. 506 163 445 152 4,610 255 n.a. n.a. 5,331 355 z zyxwvu zyxwvutsrqpo -zyx zyxwvu zyxw 267 NITROGEN BUDGET OF A FISH POND TABLE 4. Estimated nitrogen inputs to the fishpond. Inout source mole N Percent of total 281 26 - E 0 24 - W Feed Water supply Ammonium Nitrite Nitrate Nitrogen fixation Water column Water-sediment interface Benthic zone Total 2,865.4 15.76 3.35 42.83 342.19 0.07 I 17.37 3.286.97 87. I 0.5 0. I 1.3 .a.gE' 2220- 18- zyxwvu zyx 10.4 co. 1 0.5 100 deposited in the sediments. Ammonium input to the water column amounted to 2,586 mole N; 687 moles diffusing from the sediments, 15.9 moles added in the water supply and an estimated 1,883 moles from fish excretion. Additional unquantified ammonium sources include dry atmospheric deposition, microbial ammonification in the water column and excretion by zooplankton, seed fish, egrets and snails. High nitrification rates were observed in the water column. The resulting nitrate was not denitrified to any great extent in the sediment and did not accumulate in the water column or the porewaters. A total of 1,744 moles of nitrate appeared in the system; 43 added through the piped input and 1,701 arising from nitrifying activity. As 22 moles were lost through denitrification, a balance of 1,722 moles is left. This, plus the 885 moles of ammonium not nitrified, amount to a total of 2,607 moles dissolved inorganic N appearing in the water column throughout the experiment. As only 24.2 moles N accumulated in the form of phytoplankton biomass at the end of the experiment and accumulation of dissolved inorganic nitrogen was negligible, 2,583 moles N must have cycled through this compartment during the course of the experiment and appeared in the sediment as detrital organic nitrogen. The observed accumulation of organic matter in the sediment (2,145 moles N) was in fact quite close to this value. To meet this ' 0 v) 14 1 2 1 Jul Aug Sept Oct Nov FIGURE 4. Temporal variations of Secchi disk visibility. Bars represent standard error; n = 6. cycling rate, a net phytoplankton doubling rate of0.6/d would be required; 2,583 moles divided by the 159 days of operation divided by in turn by the 26 moles N mean phytoplankton biomass in the pond. This value is well within the range of those reported by Colman and Edwards (1987) for common pond phytoplankton species. Perhaps the most significant result of this study is the observation of low nitrogen loss through denitrification. Although the rate of nitrification in the water column was substantial, that at the sedimedwater interface was low. Sediment denitrification is usually maintained by the nitrate input from benthic nitrifiers in natural systems (Blackburn 1987). Given the low rates of nitrification observed at the sediment/water interface, it is not surprising that denitrifying communities in the pond exhibit low activity. Nitrification at the sediment/water interface averaged 35.6 pmole N/m2/d, while denitrification in the sediment averaged 139.8 pmole N/m2/d. The balance of nitrate necessary to supply the observed denitrification rate is most probably derived from nitrification in the water column and subsequent diffusion to the sediment. These results contrast sharply with previously postulated schemes wherein denitrification constitutes 268 zyxwvuts zyxwvu zyxwvutsrq zyxwvut ACOSTA-NASSAR TABLE5 . Mass balance of nitrogen in thehhpond (moles). Reservoir Initial Fish Commerical size New recruits 26.40 n.a. Water column Ammonium Nitrite Nitrate 2.13 0.89 0.15 Final 599.70 166.3 2.12 1.18 1.27 Sediment Adsorbed ammonium (KCI extractable) Organic nitrogen 82.1 5,698 144.9 7,842 Phytoplankton biomass 13.9 38.2 Loss to atmosphere through denitrification Water column Benthic zone Total accounted n.a. n.a. n.a. n.a. Net difference Percent of total inout zyxwv zyxw zy 573.3 166.3 --0.01 17.45 4.38 0.29 1.12 <0.01 0.0 I 0.03 62.82 2,144 1.91 65.22 24.3 0.20 30.07 3,002.5 0.74 <o.o 1 0.92 90.7 zyx de tilapias. Master's thesis. University of Puerto a substantial nitrogen sink in comparable Rico, Mayagiiez, Puerto Rico. ponds Operated in a environment Belser, L,W. and E. A. L. Mays. 1980. Specific in(Boyd 9 8 5 ; and Boyd 989) and hibition of nitrite oxidation by chlorate and its use coincide with the observations of Kaspar et in assessing nitrification in soils and sediments. Applied and Environmental Microbiology 39505al. (1988) for seacage salmon farms where 510. denitrification directly under the cages is seV. M. and S. W. Nixon. 1990. Temperverely inhibited in comparison with adja- Berounsky, ature and the annual cycle of nitrification in waters cent sediments. Blackburn et al. (1 988) have of Narragansett Bay. Limnology and Oceanograsimilarly reported low denitrification rates phy 37(7): 161G1617. in the sediments of earthen marine fish- Blackburn,T. H. 1987. Role and impact of anaerobic microbial processes in aquatic systems. Pages 32ponds. ' ' Acknowledgments Support for this work was provided through a grant to J. Corredor and J. Morel1 from the University of Puerto Rico Sea Grant Program. Organic analyses were carried out in the laboratory of Dr. E. Riquelme. We thank Dr. G. Schroeder and two anonymous reviewers for their substantial contribution to the quality of this paper. Literature Cited Avnimelech, Y. and M. Lacher. 1979. A tentative nutrient balance for intensive fishponds. Bamidgeh 31(1):3-9. Ayarza, C. A. Calidad de agua en estanque intensivo 53 in D. J. W. and R. S. V. Pullin, editors. Detritus and microbial ecology in aquaculture. ICLARM Conference Proceedings 14. International Center for Living Aquatic Resources Management, Manila, Phillipines. Blackburn, T. H., B. A. Lund and M. D. Krom. 1988. C- and N-mineralization in the sediments ofearthen marine fishponds. Marine Ecology Progress Series 44:221-227. Boyd, C. E. 1985. Chemical budgets for channel catfish ponds. Transactions ofthe American Fisheries Society 114:291-298. Colman, J. A. and P. Edwards. 1987. Feeding pathways and environmental constraints in waste-fed aquaculture: balance and optimization. Pages 24Ck 218 in D. J. W. and R. S. V. Pullin, editors. Detritus and microbial ecology in aquaculture. ICLARM Conference Proceedings 14. International Center for Living Aquatic Resources Management, Manila, Phillipines. Corredor, J. E. and D. G. Capone. 1985. Studies on z zyxwv zyxwvut zyxwv NITROGEN BUDGET OF A FISH POND nitrogen diagenesis in coral reef sands. In E. D. Gomez et al., editors. Proceedings of the Vth International Coral Reef Symposium 3:395-399. Manila, Phillipines. Corredor, J. E. and J. M. Morell. 1989. Assessment of inorganic nitrogen flux across the sediment-water interface in a tropical lagoon. Estuarine Coastal and Shelf Science 28:339-345. Daniels, H. V. and C. E. Boyd. 1989. Chemical budgets for polyethylene-lined, brackishwater ponds. Journal of the World Aquaculture Society 20(2): 269 Department of Fisheries, Bangkok, Thailand and International Center for Living and Aquatic Resources Management, Manila, Philippines. Lewis, W. M. and D. P. Moms. 1986. Toxicity of nitrite to fish: a review. Transactions of the American Fisheries Society 115(2): 183-1 95. Mark, J. and M. Kemp. 1984. The coupling of nitrification and denitrification in two estuarine ecosystems. Limnology and Oceanography 29(3):609- 619. Meade, J. W. 1985. Allowable ammonia for fish culture. The ProgressiveFish-Culturist47(3):135-145. 53-60. Daud, S. K., D. Hasbollah and A. T. Law. 1988. Ef- Melard, C. 1986. Les bases biologiques de I'elevage fects of unionized ammonia on red tilapia (Oreointensif du tilapia du nil. Cahiers d'Ethologie Apchromis mossarnbicusl0. niloticus) fry. Pages 4 1 1pliquk 6(3). 413 in R. S. V. Pullin, K. Tonguthai and J. L. Mikkelsen, D. S. 1987. Nitrogen budgets in flooded Maclean, editors. The Second International Symsoils used for rice production. Plant and Soil 100: posium on Tilapia in Aquaculture. ICLARM Con7 1-97. ference Proceedings 15. Department of Fisheries, Nieves, F. A. and J. E. Corredor. 1987. Gradientes Bangkok, Thailand and International Center for de fijaci6n de nitr6geno en 10s sedimentos marinos Living and Aquatic Resources Management, Maasociados a arrecifes coralinos en el suroeste de nila, Philippines. Puerto Rico. Anales del Instituto de InvestigaDiab, S. and M. Shilo. 1986. Transformation of niciones Marinas 17:27-37. trogen in sediments of fish ponds in Israel. Bam- Oudot, C., C. Andrie and Y. Montel. 1990. Nitrous idgeh 38(3):67-88. oxide production in the tropical Atlantic Ocean. Diab, S., M. Kochba, D. Mires and Y. Avnimelech. Deep-sea Research 37:183-202. 1992. Combined intensive-extensive (CIE) pond Pamas, B. and N. Lahav. 1972. Movement ofnitrates system. A: inorganic nitrogen transformations. and others dissolved salts from a fishpond into Aquaculture I0 I :33-39. Lake Kinneret. Bamidgeh 24(4):99-133. El Samra, M. I. and J. Olah. 1979. Significance of Parsons, T. R., M. Takahashi and B. Hargrave. 1984. nitrogen fixation in fish ponds. Aquaculture 18: Biological oceanographic processes, 3rd edition. 367-37 3. Pergamon Press, Elmsford, New York, USA. Henriksen, K., J. 1. Hansen and T. H. Blackburn. 1981. Rates of nitrification, distribution of nitri- Russo, R. C. and R. V. Thurston. I99 1. Toxicity of ammonia, nitrite and nitrate to fishes, Pages 58fying bacteria and nitrate fluxes in different types 89 in D. E. Brune and J. R. Tomasso, editors. of Danish sediments. Marine Biology 6 1:299-304. Aquaculture and water quality. Advances in world Howarth, R. W., R. Marino and J. Lane. 1988. Niaquaculture, volume 3. The World Aquaculture trogen fixation in freshwater, estuarine, and maSociety, Baton Rouge, Louisiana, USA. rine ecosystems. I . Rates and importance. LimSchroeder, G. L. 1987. Carbon and nitrogen budgets nology and Oceanography 33(4): 669-687. in manured fish pond on Israel's coastal plain. Kaspar, H.F., G. H. Hall and A. J. Holland. 1988. Aquaculture 62:25 9-279. Effects of sea cage salmon farming on sediment nitrification and dissimilatory nitrate reductions. Seitzinger, S. P. 1988. Denitrification in freshwater and coastal marine ecosystems: ecological and Aquaculture 70:333-344. geochemical significance. Limnology and OceanKnowles, R. 1982. Denitrification. Microbiological ography 33(4):702-724. Reviews 46:43-70. Krom, M. D. and A. Neori. 1989. A total nutrient Stewart, W. D. P., G. P. Fitzgerald and R. H. Burris. 1968. Acetylene reduction by nitrogen fixing bluebudget for an experimental intensive fishpond with green algae. Archiv fur Mikrobiolgie 62:336-348. circulatory moving seawater. Aquaculture 83:345Taylor, B. E. 1983. Assays of microbial nitrogen 358. transformations. Pages 809-868 in E. J. Carpenter Krom, M. D., C. Porter and H. Gordin. 1985. Nuand D. G. Capone, editors. Nitrogen in the marine trient budget of a marine fish pond in Eliat, Israel. environment. Academic Press, New York, New Aquaculture 5 1:65-80. York, USA. Kwei Lin, C., V. Tansakul and C. Apinhapath. 1988. Biological fixation as a source of nitrogen input in van Rijn, J., S. Diab and M. Shilo. 1984. Mechanisms of ammonia transformation in fishponds. fishponds. Pages 53-58 in R. S . V. Pullin, K. TonPages 17-40 in H. Rosenthal and S. Sang, editors. guthai and J. L. Maclean, editors. The Second InResearch on aquaculture. European Mariculture ternational Symposium on Tilapia in AquaculSociety, Special Publication No 8. Belgium. ture. ICLARM Conference Proceedings 15. zyxwvutsrqpo zyxwvuts 270 zyxwvutsr zyxwvu ACQSTA-NASSAR Weiss, R. F. and B. A. Price. 1980. Nitrous oxide solubility in water and seawater. Marine Chemistry 8347-359. Whitledge, T. E., S. C. Malloy, C. J. Patton and C. D. Wirick. 198 1. Automated nutrient analysis in seawater. Brookhaven National Laboratory Technical Report No. BNLS 1398. Wilheim, E., R. Battino and R. J. W i l m k . 1977. Low pressure solubility of gases in liquid water. Chemical Reviews 77(2):2 19-262.