Zero discharge aquaculture by farming in integrated recirculating systems in Asia

In different experiments the nutrient balances for tilapia, Japanese flounder and sea bass were determined for different diets, comprising different feed ingredients. These experiments were focussing on the effects of fishmeal replacement by other feed stuffs on the nutrient balance, The altered feed ingredients were derived from other resources or from processes that can be integrated in the ZAFIRA concept. Next to standard laboratory techniques, advanced statistical means were used to evaluate the obtained results. In one of the experiments, SCP (single cell protein) was used as a possible fish meal replacement in tilapia diets. Furthermore the attraction of SCP for shrimps (Litopenaeus vannamei) was tested in a behaviour study. The results of the experiments were published as a conference contribution and peer-reviewed papers (Schneider et al., 2004 and Schneider et al., 2004. The used methodologies were transferred to partner SFU, to evaluate their experiments. This resulted in joined publications. The shrimp behaviour study demonstrated that the shrimp were attracted by SCP, thereby suggesting indirectly that it has potential as feed resource (Schneider et al., 2004; Schneider et al., 2006). However, the present data do not allow yet to conclude that SCP can be used successfully as a feed ingredient. That needs more specific experimentation.

Suspended solids are a valuable recycling material that can be processed in secondary detrivorous reactors. For the investigations with small, medium and larger juvenile sea bass (Dicentrarchus labrax) an experimental recirculating aquaculture system (total water volume U 3.4 m³) was set up in the research facilities at IFM-GEOMAR. In doing so the investigations into the waste removal, the influence of ozone and the operation of a fish waste based detrivorous reactor could be carried out under realistic conditions with medium sized and larger fish. Parallel to the experiments in Kiel, smaller fish (Dicentrarchus labrax) were investigated in a second prototype recirculation system at Erwin Sander Elektroapparatebau GmbH in Eltze, a cooperating SME. The results were included in the data set that was made available for the model development in the frame of the ZAFIRA project. During the project the qualitative and quantitative changes of total suspended solids (TSS) in a RAS were investigated, specifically addressing the particle size distribution and the removal efficiency for various size fractions of particles that were retained by the installed filtration units. The RAS configuration under study was designed to evaluate the performance of a system configuration that involved a two-step solid separation procedure (swirl separator and foam fractionation), each of them independent, however, partly affecting each others performance. The system was also designed to operate at low energy consumption and at a low water replacement. The study period lasted 437 days. The performance of fish served as criteria to assess the functionality of the system in terms of fish growth and feed conversion efficiency. The system was operated at a temperature of about 23°C. Water quality was continuously monitored and was always maintained within safe limits. Water replacement in the system accounted to 1% per day of the total system volume on average. The two step solid separation techniques allowed to obtain clear water conditions. The swirl separator worked effectively capturing larger solids, while fine particles were removed by the foam fractionator. The nutrient composition of the solids varied over time, depending on the nutrient content in the feed. The use of ozone was beneficial resulting in reduced bacteria counts in the water. The results allowed to estimate nutrient inputs (feed), nutrient retention (fish growth and conversion) as well as waste output (soluble and particulate wastes) as base for the design of secondary detrivorous reactors. Part of the results were compiled in a PhD dissertation (Orellana 2007). The detrivorous reactor was stocked with a marine polychaete (Nereis diversicolor). The solid waste was collected from the primary recirculating aquaculture system for European Sea bass. Based on the results of (1), the amount of waste removed in the two different types of solid separators (mechanical filters) could be estimated and was made available to design the secondary detrivorous reactor. Protein and energy content of applied solid waste ranged from 10 to 36 % and 11.6 to 18.7 kJ per gram dry weight, respectively. Reduction of Chemical Oxygen Demand (COD) was in the range of 6.7 to 24 %.

Waste produced from a pilot-scale fish farming unit was evaluated in batch experiments for its suitability to serve as substrate for heterotrophic bacteria production. The results of this study led to a PhD dissertation on heterotrophic bacteria conversion of carbon supplemented fish waste using lab scale bacteria growth reactors, analytical, statistical and modelling methodologies (Schneider 2006). Only by supplementing the substrate with organic carbon (first sodium acetate as reference and later molasses as a more natural product were tested) bacteria production was significantly enhanced and reached levels of 100-120gVSS/kg feed. Together with this conversion of the solid waste, also 90-95% of the dissolved nitrogen and phosphorus was converted. Furthermore other parameters, such as carbon supplementations levels, different nitrogen sources or variations of hydraulic retention times were studied. Finally the bacterial product was evaluated by biochemical and molecular methods and the re-use potential as aquatic feed determined. The last part of the work focused on the hypothetical reactor design integrated in a 100MT African catfish farm. From the experiments in the ZAFIRA project we concluded that production of bacteria biomass on the solid waste of fish is technically feasible, but at the current status of knowledge, economically only possible in small volumes. The upscaling to commercial size and the upgrading of the technology into an economically viable system still needs to be done. The results are summarized in different peer-reviewed papers, conference communications and a book (Schneider et al., 2004; Schneider et al., 2005; Schneider et al., 2005; Schneider, 2006; Schneider et al., 2006; and several papers in press: Schneider et al., in press).

During the project period (2002-2006) several aquatic conversion processes were investigated in two environments: fresh and marine water. Four different fish species were used as waste producers: African catfish and tilapia (freshwater) and Sea bass & Japanese flounder (marine water). Because the separation of fish waste into solid and dissolved waste was recognised as an important treatment step in providing nutrients to the subsequent modules, the partitioning of the fish waste in these 2 waste streams was studied as a separate process. Studies on waste separation efficiencies are still scarce. Because separation processes in freshwater systems have been studied intensively already (Chen et al., 1997; Timmons et al., 2001), the ZAFIRA work focused on seawater systems, using European seabass as a model of study. The original purpose was to validate the relations found in the detailed Seabass studies, done in Europe, with data obtained in China on Japanese Flounder. Unfortunately, the latter data were insufficient to make this valdidation.. Furthermore, four different primary waste conversion processes have been studied: the conversion of fish waste into heterotrophic bacteria in freshwater systems, two phototrophic processes applying macro- and microalgae, both in seawater, and conversion of solid waste by worms. Again, the latter was only studied in marine water systems. The produced bacteria were intended to be used as feed for shrimp or fish. The phototrophic conversion products were used as feed for herbivores (macroalgae for sea urchins, microalgae for Artemia). The worms were treated as a valuable and harvestable end product. As a consequence the following conversion chains were investigated experimentally and the found nutrient balances used for the model: -"Fish Biomass Converter-Bacteria Waste Processor - Bacterivorous Conversion (freshwater system) -"Fish Biomass Converter - Fish Waste Processor -Phototrophic Converter - Herbiborous Converter (marine system) -"Fish Biomass Converter - Fish Waste Processor - Fish Waste processor- Detrivorous Converter (marine system) The developed model integrates all the experimental results on the mentioned nutrient chains into one model approach. According to the model output, the integration of multi-trophic conversion processes increases nutrient retention in intensive aquaculture systems. Table 1: increase of nutrient retention in intensive aquaculture systems -N-retention Fish 20 - 42 % Fish + Plants 60 - 85 % Fish + Plants + Herbivore 29 - 45 % Fish + Bacteria1+ 7 % Fish + Bacteria + Fish2 50 - 55 % Fish + Worms+ 0,06 % 1 based on Knoesche et al 1974; 2 based on Schneider et al, submitted The project results showed that the combination of fish culture with subsequent phototropic and herbivorous conversion increases nutrient retention in the culture system (e.g. 20-42% feed N to 29-45% feed N). Fish culture with phototrophic conversion increased nutrient retention of feed nitrogen (N) by 15-50% and feed phototrophic (P) by up to 53%. If in addition herbivore consumption was included, nutrient retention decreased again by 60-85% feed N and 50-90% feed P. This was according to the general observation of nutrient losses from one trophic level to the next. The conversion of nutrients into bacteria and detrivorous worm biomass contributed only in smaller levels (e.g. 7% feed N and 6% feed P and 0,06% feed N 0,03 x10-3 % feed P, respectively) and their incorporation in a multi-trophic production system can be questioned when the purpose is to develop a Zero Nutrient Discharge Aquaculture System. Further, the model allowed to evaluate system-planning and design for integrated systems. It was generally concluded that integrated intensive fish production systems offer a possibility to reduce nutrient discharge and to increase nutrient retention in harvestable products.

Three experiments, with different stocking densities of N. diversicolor were carried out at low (188 - 235 Individuals per m²) and high stocking densities (376 - 470 Individuals per m²) in order to monitor the effect of sludge compostion on survival and growth. Attempts were made to rear Nereis, but worm survival and growth under laboratory condition were lower compared to the tideland. Also the nutrient retention proved to be low. Average mortality ranged between 9 to 21 % during experiments. No relation between stocking rate and mortality and food composition and mortality was observed. A relationship between the specific growth rate and the amount of daily supplied energy per individual worm was elaborated. The average protein content of N. diversicolor reared under the experimental conditions ranged from 57 to 64 %. Protein retention due to growth of the worms ranged from 0.3 to 8.8 %. Beside of that, spontaneous reproduction occurred in the reactor that would allow to continuously produce and restocks the reactor in the future. Under the given conditions, the worm life cycle can potentially be closed in about 120 days. However, the control over reproduction and growth is still not completely understood and needs further investigations.

In the experiments dealing with micro- and macroalgae integrated together with herbivorous organisms the potential of phototrohic conversion of dissolved fish waste was demonstrated. The potential of an outdoor culture of the diatom Navicula lenzii, as a candidate for nutrient removal from marine aquaculture effluents was demonstrated (task 1). A sequential-batch culture system which removed almost 100% of the nutrients in the intensive marine fish pond effluents was developed. It's worth mentioning that nitrate and phosphate removal rates using Navicula lenzii is limited to 0.5 gN/m2/d and 0.06 gP/m2/d), respectively. Techniques and management practices were developed for an outdoor microalgae culture system. A protocol was developed as to reach a stable growth pattern which maintained Navicula lenzii as a monoculture at its log phase independent of the season. This was achieved by using harvesting cycles of 3-7 days (summer or winter, respectively) of 50% of the culture tank volume, through continuous replacement and separation of the algae based on a solids filtration upflow bead filtration system. The use of silicate (Na2SiO3) and FeCl2 enrichment and daily use of Sodium Hypochlorite supported the monoculture of N. lenzii at concentrations of up to 0.5 106 cells ml-1. Modelling the nutrient uptake and growth performance of the algae shows that 55.8% of the excreted N and 69.2% of the excreted P from the fish tank could be accounted for in the micro algae system, but only 60.5% and 89.1%, respectively, was retained in the algae biomass. The loss of remaining N and P could not be explained by the algal growth but could potential be in a microbial sink which we did not measure. Artemia salina can serve as an efficient herbivore of N. lenzii produced in such bio filtration systems. Better Artemia growth performance was obtained in indoors system with higher SGR, nitrogen and phosphorus retention efficiency and lower FCR. Modelling the nutrient uptake and growth performance of Artemia shows that only 16.7% of algal N and 0.06% of algal P was retained in Artemia while the rest was released to the medium (as DIN and Fecal matter). In both algal models (N. Lenzii and Ulva lactuca) we found relatively large unexplained sinks of nutrients so that the apparent uptake values were higher than measured accumulation of N and P in the added growth by both the microalgae and Ulva. This could be potentially explained by an accumulation in a microbial biomass. This was not measured in this project. A major drawback of incorporating algae in a multi-trophic system such as the ZAFIRA concept was the high volume needed to grow algae. The experiments at NCM further demonstrated that in the tested configuration, P removal efficiency seemed not high enough. Next, when an algae step is inserted in the system using the effluents of a fish recirculation system, nitrogen is mainly supplied to the algae as nitrate which is less favourable for algal growth than ammonia. This can be bypassed when either an appropriate algal species is used or (possibly more easy) the effluent from the fish tank is used directly as input to the algae tanks without first passing trough a bacterial nitrification reactor. Techniques for culture of microalgae diatom Navicula lenzii in an outdoor culture system have to be improved in order to achieve algae performance like in indoor conditions. Further, in the present set of experiments, Artemia was used as herbivore. For future system development, other more valuable algal grazers such as oysters or abalone should be tested in different integrated system configurations.

Three experiments, with different stocking densities of N. diversicolor were carried out at low (188 - 235 Individuals per m²) and high stocking densities (376 - 470 Individuals per m²) in order to monitor the effect of sludge compostion on survival and growth. The experimental data, which were collected during the project, were used to draft a simulating model for a multi-trophic aquaculture system based on the ZAFIRA concept. This model may help and evaluate production strategies for intensive integrated aquaculture systems in the future. The model allowed to: 1) transfer experimental results into numeric equations that allowed judging the waste conversion efficiency and performance of the single modules, 2) join single modules with each other and to study the effect of module interaction on nutrient retention, 3) investigate the effect of a changed performance of one module on the subsequent module, 4) compare the nutrient retention of intensive integrated aquaculture production to conventional intensive fish production methods. During the project period (2002-2006) several aquatic conversion processes were investigated in two environments: fresh and marine water. Four different fish species were used as waste producers: African catfish and tilapia (freshwater) and Sea bass & Japanese flounder (marine water). Because the separation of fish waste into solid and dissolved waste was recognised as an important treatment step in providing nutrients to the subsequent modules, the partitioning of the fish waste in these 2 waste streams was studied as a separate process. Studies on waste separation efficiencies are still scarce. Because separation processes in freshwater systems have been studied intensively already (Chen et al., 1997; Timmons et al., 2001), the ZAFIRA work focused on seawater systems, using European seabass as a model of study. The original purpose was to validate the relations found in the detailed Seabass studies, done in Europe, with data obtained in China on Japanese Flounder. Unfortunately, the latter data were insufficient to make this valdidation.. Furthermore, four different primary waste conversion processes have been studied: the conversion of fish waste into heterotrophic bacteria in freshwater systems, two phototrophic processes applying macro- and microalgae, both in seawater, and conversion of solid waste by worms. Again, the latter was only studied in marine water systems. The produced bacteria were intended to be used as feed for shrimp or fish. The phototrophic conversion products were used as feed for herbivores (macroalgae for sea urchins, microalgae for Artemia). The worms were treated as a valuable and harvestable end product. As a consequence the following conversion chains were investigated experimentally and the found nutrient balances used for the model: -"Fish Biomass Converter-Bacteria Waste Processor - Bacterivorous Conversion (freshwater system) -"Fish Biomass Converter - Fish Waste Processor -Phototrophic Converter - Herbiborous Converter (marine system) -"Fish Biomass Converter - Fish Waste Processor - Fish Waste processor- Detrivorous Converter (marine system) The developed model integrates all the experimental results on the mentioned nutrient chains into one model approach. According to the model output, the integration of multi-trophic conversion processes increases nutrient retention in intensive aquaculture systems.

Deliverables

Share this page

Download