Update on Tilapia and Vegetable Production in the UVI Aquaponic System James E. Rakocy, Donald S. Bailey, R. Charlie Shultz and Eric S. Thoman University of the Virgin Islands Agricultural Experiment Station RR 2, Box 10,000 Kingshill, VI 00850, USA [email protected] ABTRACT The UVI commercial-scale aquaponic system has produced Nile and red tilapia continuously for 4 years. During that time, two trials have been conducted to evaluate the production of basil and okra. Tilapia were harvested every 6 weeks from one of four 7.8-m3 rearing tanks. Nile and red tilapia were stocked at 77 and 154 fish/m3, respectively. During the last 20 harvests, production of Nile and red tilapia averaged 61.5 and 70.7 kg/m3, respectively. Mean harvest weight was 813.8 g for Nile tilapia and 512.5 g for red tilapia. Nile tilapia attained a higher survival rate (98.3%) and a lower red conversion ratio (1.7) than red tilapia (89.9% and 1.8, respectively). Projected annual production is 4.16 mt for Nile tilapia and 4.78 mt for red tilapia. Batch and staggered production of basil in the aquaponic system was compared to field production of basil using a staggered production technique. Annual projected yield of basil is 25.0, 23.4 and 7.7 kg/m3 for batch, staggered and field production, respectively. Annual projected yield of basil for the aquaponic system is 5.34 mt for batch production and 5.01 mt for staggered production. However, batch production was not sustainable with the current fish output because nutrient deficiencies occurred. The okra trial compared the production from three varieties (Clemson, Annie Oakley and North South) and two planting densities (2.7 and 4.0 plants/m2) in the aquaponic system. One variety (Clemson) was cultivated in a field plot at the low planting density. The highest production (3.04 kg/m2) was attained by the variety ‘North South’ at the high density. Projected annual production of ‘North South’ is 13.37 kg/m2 and 2.86 mt per system. Field okra grew slowly and produced only 0.15 kg/m2. The aquaponic system performed well over a sustained period of time. Aquaponic production of basil and okra was dramatically higher than field production. 1 INTRODUCTION Aquaponics is the combined culture of fish and plants in recirculating systems. Nutrients, which are excreted directly by the fish or generated by the microbial breakdown of organic wastes, are absorbed by plants cultured hydroponically (without soil). Fish feed provides most of the nutrients required for plant growth. As the aquaculture effluent flows through the hydroponic component of the recirculating system, fish waste metabolites are removed by nitrification and direct uptake by the plants, thereby treating the water, which flows back to the fish-rearing component for reuse. Aquaponics has several advantages over other recirculating aquaculture systems and hydroponic systems that use inorganic nutrient solutions. The hydroponic component serves as a biofilter, and therefore a separate biofilter is not needed as in other recirculating systems. Aquaponic systems have the only biofilter that generates income, which is obtained from the sale of hydroponic produce such as vegetables, herbs and flowers. In the UVI system, which employs raft hydroponics, only calcium, potassium and iron are supplemented. The nutrients provided by the fish would normally be discharged and could contribute to pollution. Removal of nutrients by plants prolongs water use and minimizes discharge. Aquaponic systems require less water quality monitoring than individual recirculating systems for fish or hydroponic plant production. Aquaponics increases profit potential due to free nutrients for plants, lower water requirements, elimination of a separate biofilter, less water quality monitoring and shared costs for operation and infrastructure. A commercial-scale aquaponic system was developed at the University of the Virgin Islands in St. Croix. The status of the system was reported in the proceedings of ISTA 4 and 5 (Rakocy et al. 1997: Rakocy et al. 2000). The development of the system initially required many design changes. There have been no major changes in system design since ISTA 5. The system has produced tilapia continuously since that time. During the continuous production of tilapia, two short-term trials were conducted to determine the production capacity of basil and okra. This paper will report the highlights of this work. METHODS The design of the UVI aquaponic system is shown in Figure 1. The water pump, which is located on the left side of the sump, pumps water a short distance to the fish rearing tanks. The flow to individual tanks is regulated by ball valves. As the carrying capacity of a fish- rearing tank is reached, a greater portion of the flow (378 L/min.) is diverted to that tank. Water flows from the fish-rearing tanks through the rest of the system by gravity and returns to the sump, which is the lowest point in the system. Each fish-rearing tank has 22 air diffusers (14 L/min), which are cleaned weekly. There are four air diffusers in the degassing tank and one in the base addition tank. Each hydroponic tank has 24 air diffusers (10 L/min), which are positioned every 1.2 m in the center of the tank. 2 Base addition tank Hydroponic tanks Rearing tanks Degassing tank Sump Clarifier Filter tanks Tank Dimensions Pipe Sizes Rearing tanks: Diameter: 3 m, Height: 1.2 m, Water Pump to rearing tanks: 7.6 cm volume: 7,800 L Rearing tanks to clarifier: 10 cm Clarifiers: Diameter: 1.8, Height of cylinder: 1.2 m, Clarifiers to filter tanks: 10 cm Depth of cone: 1.1 m, Slope: 45º, Water volume: 3,785 L Between filter tanks: 15 cm Filter and degassing tanks: Length: 1.8 m, Width: 0.76 Filter tank to degassing tank: 10 cm m, Depth: 0.61 m, Water volume: 700 L Degassing to hydroponic tanks: 15 cm Hydroponic tanks: Length: 30.5 m, Width: 1.2 m, Between hydroponic tanks: 15 cm Depth: 41 cm, Water volume: 11,356 L Hydroponic tanks to sump: 15 cm Sump: Diameter: 1.2 m, Height: 0.9 m, Water volume: Sump to pump: 7.6 cm 606 L Pipe to base addition tank: 1.9 cm Base addition tank: Diameter: 0.6 m, Height: 0.9 m, Base addition tank to sump: 3.2 cm Water volume: 189 L Total system water volume: 111,196 L Flow rate: 378 Lpm, Pump: ½ hp, Blowers: 1½ hp (fish) and 1 hp (plants) Total land area: 0.05 ha. Figure 1. Layout of UVI Aquaponic System Settleable solids are removed from the clarifiers three times daily by opening a ball valve. Fine solids collect on orchard netting in the filter tanks and are removed one or two times weekly by draining the tank and washing the netting with a high pressure water spray. Effluent from the filter tanks passes through a fine-meshed plastic screens as it enters the degassing tank. The screens, which prevent tilapia fry from reaching the hydroponic tanks, are washed daily. The pH is monitored daily and maintained at 7.0-7.5 by alternately adding calcium hydroxide and potassium hydroxide to the bass addition tank, where it dissolves and slowly enters the system. In the process of adding base, calcium and potassium ions are supplemented. The only other nutrient requiring supplementation is iron, which is added in a chelated form at a concentration of 2 mg/L once every three weeks. 3 Water lost through evaporation, transpiration and sludge removal is replenished with rainwater in the sump. Influent water is regulated by a float valve and measured by a water meter. Tilapia are stocked in the rearing tanks at a rate of 77 fish/m3 for Nile tilapia (Oreochromis niloticus) or 154 fish/m3 for red tilapia and cultured for 24 weeks. Production is staggered so that one tank is harvested every 6 weeks, at which time they are weighed and counted. After harvest, the rearing tank is immediately restocked. The fish are fed three times daily with a complete, floating pellet containing 32% protein. The fish are fed ad libitum to satiation over a 30-minute feeding period. Tilapia are produced continuously in this system to maintain stable bacterial populations. Hydroponic experiments are limited in duration. Between hydroponic experiments, the system is operated with a variety of demonstration crops or no crops at all. The hydroponic component has waste treatment capacity in excess of the amount generated at the recommended feeding rate and maintains acceptable ammonia and nitrite concentrations without the presence of plants. Snails were introduced to the system several years ago, probably by birds. Snails consumed the nitrifying bacteria, which resulted in higher concentrations of ammonia and nitrite. Therefore, the hydroponic tanks were stocked with red ear sunfish fingerlings (Lepomis microlophus) to control the snail populations. Hydroponic trials were conducted to evaluate the production of basil and okra. Basil trials occurred in 2002, and an okra trial occurred in 2003. Two basil production trials were conducted during the periods of January 28 - May 20 and June 18 - September 20, 2002. Hydroponic basil was planted into a fully established and functioning system with a reservoir of nutrients generated from fish waste. In the first trial basil was produced by batch culture. Basil seedlings were planted at a density of 8 plants/m2 in the entire 214-m2 growing area. The plants were harvested three times by cutting and allowed to re-grow before a fourth and final harvest. The production period averaged 28.3 days. The main stem was cut at a height of 15 cm, which left sufficient leaves for re-growth. In the second trial a staggered production method was used. Every week one fourth of the growing area (53.5 m2) was planted with basil seedlings at a density of 8 plants/m2 until the system was fully planted. After a 28-day growing period, the plants were harvested at a height of 15 cm, allowed to re-grow and harvested a second time. There were a total of eight harvests. During the second trial, the staggered production procedure was followed for basil seedlings that were planted in an adjoining field at a density of 8 plants/m2. The soil was prepared by applying dried composted cow manure (2-1-2) at a rate of 5.87 mt/ha. The plants were irrigated as needed with well water. Each of four plots consisted of three rows containing nine plants. Production data was collected from the seven inner plants from the middle row. Basil in the field trial was cultivated for 28 days. During all trials the plants were sprayed twice weekly with a commercial formulation of Bacillus thuringiensis to control caterpillars. 4 An okra production trial was conducted during the period of October 1 – December 22, 2003. Hydroponic okra was planted into a fully established and functioning system with a reservoir of nutrients generated from fish waste. Seedlings of three varieties of okra (North-South, Annie Oakley and Clemson Spineless) were transplanted into the system at two densities [2.7 plants/m2 (low density) and 4.0 plants/m2 (high density)] in six completely randomized blocks in a 3 by 2 factorial design. The low-density spacing was 33% less dense than the high-density spacing and is the recommended density for field production of okra. The plants grew in the aquaponic system for 33 days before the first harvest. Within each treatment replication a sample area of 2.1-m2 was delineated, encompassing nine plants in the high-density plots and six plants in the low-density plots. Pods that were 8-cm and longer were cut with hand-held pruners from the sample area, counted and weighed en mass. Pod harvests were conducted every Monday, Wednesday and Friday for 49 days. There were a total of 22 harvests. During the trial, okra seedlings were also transplanted into an adjoining field at a density of 2.7 plants/m2. To prepare the soil, gypsum was applied at a rate of 4 mt/ha, and inorganic fertilizer (N-P-K of 21-7-7) was applied at a rate of 100 kg/ha. During the trial, field okra received four foliar applications of micronutrients (iron, manganese and molybdenum). Six plots consisting of three rows were established. After transplanting, a layer of straw was placed over each plot to reduce the growth of weeds. The plots were irrigated as needed with well water using drip irrigation lines. Production data was collected from the six inner plants from the middle row. Field okra was cultivated for the same time period as okra in the aquaponic system. The first harvest of field okra occurred 54 days after transplanting and continued for 28 days. Harvests were conducted on Mondays, Wednesdays and Fridays of each week. A pesticide (Sevin) was applied twice to the field plots to control ants. Field and aquaponic okra plants were sprayed twice a week with Bacillus thuringiensis to control caterpillars. During the last 6 weeks of the trial, field and aquaponic okra were sprayed once or twice weekly with a commercial formulation of potassium bicarbonate to control mildew. In the basil trials standard methods were used to measure pH, total alkalinity, total dissolved solids (TDS), total ammonia-nitrogen (TAN), nitrite-nitrogen and nitrate-nitrogen once every 2 weeks at one location in the system. Dissolved oxygen (DO) and water temperature were measured periodically. In the okra trial standard methods were used to measure the following water quality parameters once every two weeks: DO, water temperature, pH, total alkalinity, total suspended solids (TSS), turbidity, chemical oxygen demand (COD), TDS, electrical conductivity (EC), TAN, nitrite-nitrogen, nitrate-nitrogen, total phosphorous (TP), orthophosphate, potassium, calcium, magnesium, sulfate, chloride, iron, manganese, zinc, copper, boron, molybdenum and sodium. Samples for water quality analysis were collected at the influent and effluent of the hydroponic tanks to determine changes that occurred in water quality as culture water passed through the hydroponic component and the fish rearing and solids removal components. Water temperature, pH and total alkalinity were only 5 measured at the influent to the hydroponic tanks because these parameters generally remain constant throughout the system. RESULTS AND DISCUSSION Tilapia Production Tilapia production data for 20 harvests are given in Table 1. The total harvest weight per tank for Nile and red tilapia averaged 480 and 551 kg, respectively. Based on these means, projected annual production would be 4.16 mt for Nile tilapia and 4.78 mt for red tilapia. Total harvest weight ranged from 387 to 632 kg for Nile tilapia and 448 to 688 kg for red tilapia. Some variability in total harvest weight was likely due to seasonal temperature fluctuation and the weight of fingerlings at stocking. There is a slight growth depression of tilapia from mid-December through mid-April when water temperatures decrease in the Virgin Islands, which are located at 18° North latitude. The mean weight of fingerlings stocked ranged from 43 to 138 g for Nile tilapia and 23 to 85 g for red tilapia. Feeding is another factor that may have contributed to variability in total harvest weight and slight underproduction. The fish were always fed three times daily, but the system was not managed as intensely as a commercial system would be, especially during periods between hydroponic experiments. It is possible that the fish were slightly underfed at times. With optimal ad libitum feeding, production may have been greater and less variable. Red tilapia were stocked 154 fish/m3 to produce smaller fish (512.5 g) for the West Indian market, which prefers a colorful whole fish that is served with its head on. At this density production averaged 70.7 kg/m3, and the growth rate averaged 2.69 g/day. Nile tilapia were stocked at 77 fish/m3 to produce a larger fish (813.8 g) for the fillet market. At this density production averaged 61.5 kg/m3, and the growth rate averaged 4.40 g/day. The stocking rates appeared to be nearly optimal for the desired product size. Nile tilapia attained a higher survival rate (98.3%) and a lower feed conversion ratio (1.7) than red tilapia (89.9% and 1.8, respectively). To achieve a desired minimum production of 5 mt, more research is needed on types of feed (e.g., higher protein levels) and the delivery of the feed. To achieve an annual harvest of 5 mt for Nile tilapia, the average harvest weight must be 978 g, an increase of 164 g over the current harvest weight. In addition to better feed and feed delivery, it may be necessary to stock larger fingerlings or increase the stocking rate slightly. Basil Production Batch production of basil averaged 2.0 kg/m2 per harvest (Figure 2). There was some mortality after each harvest, and final survival was 84.7%. Harvest by cutting weakened the plants and their roots became infected with Pythium. Before the first harvest, the roots appeared to be healthy. There was no indication of nutrient deficiency during the initial harvests. However, by the fourth harvest nutrient deficiencies were evident, especially in the second hydroponic tank of each set, indicating that some nutrient or nutrients became limiting as water traveled a distance of 61 m through each set of two hydroponic tanks. The 6 deficiency was manifested as chlorosis (yellowing) of the leaves. Initially there was a large reservoir of nutrients and no deficiencies appeared early in the trial. However, during the production of four consecutive batches of basil, nutrient depletion occurred. During this period, the ratio between daily feed input and plant growing area was 81.4 g/day/m2. Batch production of basil exceeded the nutrient generation capacity of the system. The cropping system was therefore changed to staggered production to moderate nutrient uptake. In the staggered production trial, the plants were cut once and allowed to re-grow for a final second harvest. Production was two times higher in the second harvest (2.4 kg/m2) than in the first harvest (1.2 kg/m2). The average weight/plant was 167.1 g in the first harvest compared to 327.1 g in the second harvest. Basil exhibited slow growth after transplanting Table 1. Production of Nile and Red tilapia in the UVI aquaponic system. Nile tilapia are stocked at 77 fish/m3 and red tilapia are stocked at 154/m3 Harvest Date Tilapia Harvest Harvest Initial Final Growth Survival FCR Weight Weight per Weight Weight Rate (%) per tank unit volume (g/fish) (g/fish) (g/day) (kg) (kg/m3) 02/07/2002 Nile 632 81.1 113 1070 5.70 98.8 1.8 03/21/2002 Nile 429 55.0 74 711 3.79 100.0 1.8 05/02/2002 Nile 417 53.4 76 701 3.72 99.2 1.7 06/13/2002 Red 528 67.7 23 476 2.52 92.5 1.8 07/25/2002 Nile 418 53.6 60 700 4.16 99.5 1.7 09/05/2002 Nile 461 59.1 76 781 4.22 98.3 1.9 10/17/2002 Nile 534 68.5 89 896 4.80 99.3 1.8 11/28/2002 Red 542 69.5 75 470 2.53 96.0 2.0 01/09/2003 Nile 460 59.0 67 769 4.18 99.7 1.7 02/20/2003 Nile 432 55.4 68 786 4.27 92.0 1.8 04/03/2003 Nile 387 49.6 61 654 3.53 98.5 1.7 05/14/2003 Red 448 57.4 52 441 2.32 84.6 1.9 06/24/2003 Nile 432 55.3 107 733 3.77 98.2 1.6 08/07/2003 Nile 480 61.6 71 825 4.49 97.0 1.6 09/18/2003 Nile 551 70.6 80 921 5.01 100.0 1.6 10/30/2003 Nile 443 56.8 59 768 4.25 96.2 1.7 12/11/2003 Red 688 88.2 85 663 3.40 86.5 1.6 01/22/2004 Nile 482 61.8 43 815 4.60 98.7 1.7 03/04/2004 Nile 551 70.6 85 924 4.99 99.3 1.7 04/15/2004 Nile 571 73.2 138 967 4.93 98.3 1.8 Mean Nile 480 61.5 79.2 813.8 4.40 98.3 1.7 Red 551 70.7 58.8 512.5 2.69 89.9 1.8 7 Figure 2. Basil and okra production in the UVI Aquaponic System while it became established. Cutting stimulated branching, and re-growth was faster. Mortality due to Pythium and aggravated by cutting was still a serious problem in this trial. After eight harvests there was no sign of nutrient deficiency. The ratio between daily feed input and growing area was 99.6 g/day/m2, 22% higher than in the batch trial. Fish consumed more feed during the staggered production trial because water temperatures were as much as 5°C higher (29°C vs. 24°C) during the summer compared to winter/spring when the batch trial was conducted. The higher feeding ratio and the even uptake of nutrients created a sustainable production system, although Pythium was more severe due to higher water temperatures. Increasing the feeding ratio during the cool season would require the cultivation of more fish. Field production of basil resulted in much lower yields (0.6 kg/m2) and average weight (104.4 g) compared to aquaponics, but survival was 100%, as the plants did not have Pythium and could easily recover from cutting. As with the staggered production, the plants grew slowly after transplanting and only attained an average yield and weight/plant of 0.3 kg/m2 and 49.8 g at the first harvest. Yield and weight/plant tripled to 1.0 kg/m2 and 159.1 g by the second harvest. A comparison of all three cropping systems showed that batch and staggered production of basil in an aquaponic system were comparable and both aquaponic crops were approximately three times more productive than field production (Table 2). Annual projected yield was 25.0, 23.4 and 7.8 kg/m2 for batch, staggered and field production, respectively. Annual projected yield for the system was 5,341 kg for batch production and 5,008 kg for staggered production. However, batch production was not sustainable with the current fish output, and nutrient deficiencies would render much of the harvest unmarketable. Fresh basil with stems sells for $22.00/kg in the U.S. Virgin Islands. Therefore, gross income from staggered production would be US$515/m2/year and US$110,210/system/year compared to field production with gross income of US$172/m2/year and US$36,808/year for the same production area. Compared to field production, the aquaponic system would save substantial labor associated with weeding but would require additional labor for seedling replacement due to mortality. Total income from the system would be US$134,245 when fish are included. The basil would generate 4.6 times more income than the fish. However, the fish contribute additional value to the system through consistent nutrient generation and the elimination of the need for excessive water quality monitoring or frequent water replacement. 8 Table 2. Comparison of basil yield, mean plant weight, survival and gross income with three production methods. (Rakocy et al., 2004) Production Annual Annual Mean Plant Survival Income Income Method Yield Yield Weight (g) (%) (US$/ (US$/214 (kg/m2) (kg/214 m2/yr) m2/yr) m2 /yr) Batch 25.0 5,341 286.5 84.7 550 117,700 Staggered 23.4 5,008 244.7 - 515 110,210 Field 7.8 1,669 104.4 100 172 36,808 Table 3. Means values and ranges of water quality variables during basil production trials. Units are in mg/L unless otherwise noted. (Rakocy et al., 2004) Variable Batch Culture Staggered Culture pH 7.4 (7.0-7.6) 7.1 (6.9-7.3) Total-ammonia-N 2.2 (1.6-2.9) 1.7 (1.1-2.4) Nitrite-N 0.7 (0.4-1.1) 0.9 (0.5-1.1) Nitrate-N 42.2 (26.7-54.7) 42.9 (30.9-51.8) Total dissolved solids 532 (490-560) 550 (490-560) Alk (mg/L, as CaCO ) 113.2 (65.6-206.7) 88.6 (65.6-115.6) 3 Water Quality - Basil Trial TAN and nitrite-N concentrations remained within safe limits for fish culture (Table 3). Nutrient concentrations were lower than the levels normally found in hydroponic systems, but they were generally acceptable for aquaponic systems because nutrients were produced daily, excreted directly by the fish or generated from the mineralization of organic matter. The slow removal of solids from the clarifiers and filter tanks provided sufficient time for substantial mineralization to occur. In addition to removing fine particulate matter, microorganisms in the filter tank also removed dissolved organic matter, preventing it from inducing excessive microbial growth on the plant roots. In this aquaponic system water develops a tea color due to the accumulation of refractory organic compounds (e.g., tannic acid), but suspended solids levels are generally low (<10 mg/L) and the water remains clear. These characteristics were exhibited in all previous trials with lettuce. With basil the water became turbid, and DO levels in the rearing tanks decreased to a range of 4.0 to 5.0 mg/L. In previous lettuce trials DO concentrations ranged from 5.0 to 6.0 mg/L in the rearing tanks. pH was maintained at an average value of 7.1 to 7.4 (Table 2). This value is considered to be high for a hydroponic system, but in an aquaponic system pH must be maintained above 7.0 to promote nitrification. Fish excrete large quantities of ammonia, which must be oxidized to nitrate to prevent toxicity. Nitrification produces acid and the pH decreased daily. To increase pH from the initiation of the first trial to the end of the second trial, including the inter-trial period, 22.5 kg of Ca(OH) was added in 23 doses and 21.5 kg of KOH was added 2 in 22 doses. Based on this data, the system would annually require 40.65 kg of Ca(OH) and 2 9 38.85 kg of KOH. In addition to regulating pH, base addition supplements Ca and K, which are not generated in sufficient quantities from fish feed for good plant growth in aquaponic systems. During the same period, 18.17 kg of iron chelate (13%) was added to the system in eight doses. The annual projected iron requirement for the system would be 32.83 kg. All other required nutrients came from fish waste. Daily make-up water averaged 2.61 m3 or 2.4% of system volume. Water loss was attributed to sludge removal, net filter washing, splashing, evaporation and transpiration. Okra Production During the 11.7-week okra trail, the plants grew luxuriously in the aquaponic system and began to produce fruit (pods) after 4.7 weeks (Figure 2). Although batch cultivation was used, there were no signs of nutrient deficiency during the production period. The ad libitum feeding method was monitored closely to avoid underfeeding. During the production trial, the ratio between daily feed input and plant growing area was 95.6 g/day/m2. Okra production was consistently higher at the high density (4 plants/m2), based on unit area (Table 4). However, production per plant was substantially greater at the low density (2.7 plants/m2). The highest production (3.04 kg/m2) was attained by the variety ‘North South’ at the high density. Projected annual production of ‘North South’ is 13.37 kg/m2 and 2,862 kg system (214 m2). Okra can be sold for $2.20/kg in the U.S. Virgin Islands. Therefore, gross income from the sale of okra would be $29.42/m2/year and $6,296/system/year. Rainfall was unusually heavy during the experimental period, and an outbreak of mildew may have lowered production. Table 4. Okra production data. Production Variety Low Density High Density Field (kg/m2) Clemson 2.67 2.87 0.15 Annie Oakley 2.37 2.76 - North South 2.58 3.04 - Mean 2.54 2.89 - (g/plant) Clemson 988 706 69 Annie Oakley 878 677 - North South 955 747 - Mean 940 710 - The field okra was slow to become established and started to produce fruit at 7.7 weeks. By the end of the experimental period, total production (0.15 kg/m2) of field okra was only 5% of the production obtained in the aquaponic system. The management of field okra was more intensive than aquaponic okra and required the following additional procedures: soil preparation (plowing, rototilling and application of gypsum and inorganic fertilizer); weeding and mulching; pesticide application for ant control; periodic drip irrigation; and several foliar applications of trace elements. 10