Evaluation of the impact of nitrate-nitrogen levels in recirculating aquaculture systems on concentrations of the off-flavor compounds geosmin and 2-methylisoborneol in water and rainbow trout (Oncorhynchus mykiss)

Aquatic animals raised in recirculating aquaculture systems (RAS) can develop preharvest “off-flavors” such as “earthy” or “musty” which are caused by the bioaccumulation of the odorous compounds geosmin or 2-methylisoborneol (MIB), respectively, in their flesh. Tainted aquatic products cause large economic losses to producers due to the inability to market them. Certain species of actinomycetes, a group of filamentous bacteria, have been attributed as the main sources of geosmin and MIB in RAS. Previous studies have demonstrated that certain nutritional factors can stimulate or inhibit bacterial biomass and geosmin production by certain actinomycetes. In the current study, the effects of two nitrate-nitrogen (NO₃^--N) levels (20–40 mg/L and 80–100 mg/L) on geosmin and MIB levels in culture water and the flesh of rainbow trout (Oncorhynchus mykiss) raised in RAS were monitored. Water and fish tissue samples were collected over an approximately nine-week period from six RAS, three replicates each of low and high NO₃^--N, and analyzed for geosmin concentrations using solid phase microextraction–gas chromatography–mass spectrometry. Results indicated no significant difference in geosmin concentrations in water or fish flesh between the low and high NO₃^--N RAS. Therefore, higher NO₃^--N levels that may occur in RAS will not adversely or beneficially impact geosmin-related off-flavor problems.     

End-of-pipe single-sludge denitrification in pilot-scale recirculating aquaculture systems

A step toward environmental sustainability of recirculat aquaculture systems (RAS) is implementation of single-sludge denitrification, a process eliminating nitrate from the aqueous environment while reducing the organic matter discharge simultaneously. Two 1700 L pilot-scale RAS systems each with a 85 L denitrification (DN) reactor treating discharged water and hydrolyzed solid waste were setup to test the kinetics of nitrate and COD removal. Nitrate removal and COD reduction efficiency was measured at two different DN-reactor sludge ages (high θ_X: 33–42 days and low θ_X: 17–23 days). Nitrate and total N (NO₃⁻ + NO₂⁻ + NH₄⁺) removal of the treated effluent water ranged from 73–99% and 60–95% during the periods, respectively, corresponding to an overall maximum RAS nitrate removal of approximately 75%. The specific nitrate removal rate increased from 17 to 23 mg NO₃⁻-N (g TVS d)⁻¹ and the maximal potential DN rate (measured at laboratory ideal conditions) increased correspondingly from 64–68 mg NO₃⁻-N (g TVS d)⁻¹ to 247–294 mg NO₃⁻-N (g TVS d)⁻¹ at high and low θ_X, respectively. Quantification of denitrifiers in the DN-reactors by qPCR showed only minor differences upon the altered sludge removal practice. The hydrolysis unit improved the biodegradability of the solid waste by increasing volatile fatty acid COD content 74–76%. COD reductions in the DN-reactors were 64–70%. In conclusion, this study showed that single-sludge denitrification was a feasible way to reduce nitrate discharge from RAS, and higher DN rates were induced at lower sludge age/increased sludge removal regime. Improved control and optimization of reactor DN-activity may be achieved by further modifying reactor design and management scheme as indicated by the variation in and between the two DN-reactors.     

Effects of alkalinity on ammonia removal,carbon dioxide stripping,and system pH in semi-commercial scale water recirculating aquaculture systems operated with moving bed bioreactors

When operating water recirculating systems (RAS) with high make-up water flushing rates in locations that have low alkalinity in the raw water, such as Norway, knowledge about the required RAS alkalinity concentration is important. Flushing RAS with make-up water containing low alkalinity washes out valuable base added to the RAS (as bicarbonate, hydroxide, or carbonate), which increases farm operating costs when high alkalinity concentrations are maintained; however, alkalinity must not be so low that it interferes with nitrification or pH stability. For these reasons, a study was designed to evaluate the effects of alkalinity on biofilter performance, and CO₂ stripping during cascade aeration, within two replicate semi-commercial scale Atlantic salmon smolt RAS operated with moving bed biological filters. Alkalinity treatments of nominal 10, 70, and 200 mg/L as CaCO₃ were maintained using a pH controller and chemical dosing pumps supplying sodium bicarbonate (NaHCO₃). Each of the three treatments was replicated three times in each RAS. Both RAS were operated at each treatment level for 2 weeks; water quality sampling was conducted at the end of the second week. A constant feeding of 23 kg/day/RAS was provided every 1–2 h, and continuous lighting, which minimized diurnal fluctuations in water quality. RAS hydraulic retention time and water temperature were 4.3 days and 12.5 ± 0.5 °C, respectively, typical of smolt production RAS in Norway.It was found that a low nominal alkalinity (10 mg/L as CaCO₃) led to a significantly higher steady-state TAN concentration, compared to when 70 or 200 mg/L alkalinity was used. The mean areal nitrification rate was higher at the lowest alkalinity; however, the mean TAN removal efficiency across the MBBR was not significantly affected by alkalinity treatment. The CO₂ stripping efficiency showed only a tendency towards higher efficiency at the lowest alkalinity. In contrast, the relative fraction of total inorganic carbon that was removed from the RAS during CO₂ stripping was much higher at a low alkalinity (10 mg/L) compared to the higher alkalinities (70 and 200 mg/L as CaCO₃). Despite this, when calculating the total loss of inorganic carbon from RAS, it was found that the daily loss was about equal at 10, and 70 mg/L, whereas it was highest at 200 mg/L alkalinity. pH recordings demonstrated that the 10 mg/L alkalinity treatment resulted in the lowest system pH, the largest increase in [H⁺] across the fish culture tanks, as well as giving little response time in case of alkalinity dosing malfunction. Rapid pH changes under the relatively acidic conditions at 10 mg/L alkalinity may ultimately create fish health issues due to e.g. CO₂ or if aluminium or other metals are present. In conclusion, Atlantic salmon smolt producers using soft water make-up sources should aim for 70 mg/L alkalinity considering the relatively low loss of inorganic carbon compared to 200 mg/L alkalinity, and the increased pH stability as well as reduced TAN concentration, compared to lower alkalinity concentrations.     

Heterotrophic denitrification of aquaculture effluent using fluidized sand biofilters

The ability to consistently and cost-effectively reduce nitrate-nitrogen loads in effluent from recirculating aquaculture systems would enhance the industry's environmental stewardship and allow improved facility proximity to large markets in sensitive watersheds. Heterotrophic denitrification technologies specifically employing organic carbon found in aquaculture system waste offer a unique synergy for treatment of land-based, closed-containment production outflows. For space-efficient fluidized sand biofilters to be used as such denitrification reactors, system parameters (e.g., influent dissolved oxygen and carbon to nitrogen ratios, C:N) must be evaluated to most effectively use an endogenous carbon source. The objectives of this work were to quantify nitrate removal under a range of C:Ns and to explore the biofilter bacterial community using three replicated fluidized sand biofilters (height 3.9 m, diameter 0.31 m; fluidized sand volume plus biofilm volume of 0.206 m³) operated at a hydraulic retention time of 15 min and a hydraulic loading rate of 188 L/min m² at The Conservation Fund Freshwater Institute in Shepherdstown, West Virginia, USA. Nitrate reduction was consistently observed during the biofilter study period (26.9 ± 0.9% removal efficiency; 402 ± 14 g NO₃-N/(m³ biofilter d)) although nitrite-N and total ammonium nitrogen concentrations slightly increased (11 and 13% increases, respectively). Nitrate removal efficiency was correlated with carbonaceous oxygen demand to nitrate ratios (R² > 0.70). Nitrate removal rates during the study period were moderately negatively correlated with influent dissolved oxygen concentration indicating it may be possible the biofilter hydraulic retention time was too short to provide optimized nitrate removal. It is reasonable to assume that the efficiency of nitrate removal across the fluidized sand biofilters could be substantially increased, as long as organic carbon was not limiting, by increasing biofilter bed depths (to 6–10 m), and thus hydraulic retention time. These findings provide a low-cost yet effective technology to remove nitrate-nitrogen from effluent waters of land-based closed-containment aquaculture systems.     

End-of-pipe denitrification using RAS effluent waste streams: Effect of C/N-ratio and hydraulic retention time

Environmentally sustainable aquaculture development requires increased nitrogen removal from recirculating aquaculture systems (RAS). In this study, removed solids from a large commercial outdoor recirculated trout farm (1000 MT year⁻¹) were explored as an endogenous carbon source for denitrification. This was done by (1) a controlled laboratory experiment on anaerobic hydrolysis of the organic matter (from sludge cones, drumfilter, and biofilter back-wash) and (2) an on-site denitrification factorial experiment varying the soluble COD (COD_S)/NO₃-N ratio from 4 to 12 at hydraulic retention times (HRT) from 50 to 170 min in simple 5.5 m³ denitrification reactors installed at the trout farm.The lab-experiments showed that the major part of the readily biodegradable organic matter was hydrolyzed within 14 days, and the hydrolysis rate was fastest the first 24 h. Organic matter from the sludge cones generated 0.21 ± 0.01 g volatile fatty acids (VFA) g⁻¹ total volatile solids (TVS), and the VFAs constituted 75% of COD_S. Analogously, 1 g TVS from the drum filter generated 0.15 ± 0.01 g VFA, constituting 68% of the COD_S. Comparison of the laboratory hydrolysis experiments and results from the on-farm study revealed as a rough estimate that potentially 17–24% of the generated VFA was lost due to the current sludge management.Inlet water to the denitrification reactors ranged in NO₃-N concentration from 8.3 to 11.7 g m⁻³ and COD_S from 52.9 to 113.4 g m⁻³ (10.0 ± 1.2 °C). The highest NO₃-N removal rate obtained was at the intermediate treatments; 91.5–124.8 g N m⁻³_reactor d⁻¹. The effect of the C/N ratio depended on the HRT. At low HRT, the variation in C/N ratio had no significant effect on NO₃-N removal rate, contrary to the effect at the high HRT. The stoichiometric ratio of COD_S/NO₃-N was 6.0 ± 2.4, ranging from 4.4 (at the high HRT) to 9.3 (at the low HRT). A simple model of the denitrification reactor developed in AQUASIM showed congruence between modeled and measured data with minor exceptions. Furthermore, this study pointed to the versatility of the NO₃-N removal pathways expressed by the bacterial population in response to changes in the environmental conditions; from autotrophic anammox activity presumably present at low C/N to dissimilatory nitrate reduction to ammonia (DNRA) at high C/N, besides the predominate “normal” heterotrophic dissimilatory nitrate reduction (denitrification).     

The “self cleaning inherent gas denitrification-reactor” for nitrate elimination in RAS for pike perch (Sander lucioperca) production

Denitrification reactors have proven their functionality in commercial recirculation aquaculture systems (RAS). Nevertheless, clogging occurs due to the low hydraulic loads necessary to accomplish anoxic conditions for a successful denitrification process in RAS, which hampers the adjustment of stable working conditions within fixed bed denitrification reactors. Reactors working on the basis of activated sludge demand careful hydraulic control and/or complex configurations for sludge retention.To develop a low-maintenance denitrification reactor, an enclosed moving bed filter, driven by recirculation of the inherent, oxygen poor gas was designed. A Self cleaning Inherent gas Denitrification reactor (SID-reactor) of 0.65 m³, which offered a moving bed volume of 0.39 m³ was connected with a RAS of semi-industrial scale for pike perch (Sander lucioperca) production. This species indicates suboptimal environmental conditions (as e.g. NO₃-N concentrations above approximately 68 mg l⁻¹) by prompt reduction of the feed intake. In different experimental series, the SID-reactor was operated with denatured ethanol, methanol, acetic acid or glycerin as carbon sources and changing operational modes.Clogging was prevented by a 40 second inherent gas recirculation twice an hour, which provided continuous, maintenance free operation with marginal energy demand. With inlet (RAS) and outlet NO₃-N concentrations in the range of 49 mg l⁻¹ and 12 mg l⁻¹, mean denitrification rates of 199 g to 235 g NO₃-N per m³ moving bed volume and day were determined for all tested carbon sources. Negative effects on the feed intake of the reared pike perch were detected with all carbon sources except methanol. Changing the mode of operation to continuous circulation of the filter bed at inlet NO₃-N concentrations of 26 mg l⁻¹, the denitrification performance reached 451 g NO₃-N per m³ moving bed volume and day. The SID-reactor allowed for the reduction of freshwater exchange in the pike perch RAS from 600 l to 70 l (−88%) and the sodium bicarbonate buffer from 182 g to 31 g (−83%) per kg of administered food. The easy and reliable operation of the SID-reactor could help to establish controlled denitrification as a routine purification step in RAS.     

The effects of ozonation on select waterborne steroid hormones in recirculation aquaculture systems containing sexually mature Atlantic salmon Salmo salar

Steroid hormones have been shown to accumulate in recirculation aquaculture system (RAS) water over time; however, their influence on the reproductive physiology of fish within RAS remains unknown. Whether ozonation impacts waterborne hormone levels in RAS has likewise not been fully evaluated. To this end, a controlled 3-month study was conducted in 6 replicated RAS containing a mixture of sexually mature and immature Atlantic salmon Salmo salar to determine whether ozone, as typically applied in RAS to improve water quality, is associated with a reduction in waterborne hormones. Post-smolt Atlantic salmon (1253 ± 15 g) were stocked into each RAS; 109 of 264 fish placed in each system were sexually mature males, and 5 were mature females. Water ozonation, controlled using an ORP set-point of 290–300 mV, was applied with the pure oxygen feed gas within the low-head oxygenators of 3 randomly selected RAS, while the remaining 3 RAS did not receive ozone. The RAS hydraulic retention time was 6.9 ± 0.3 days. Study fish were raised under these conditions for 12 weeks; during weeks 10 and 12, triplicate water samples were collected from the following locations in each RAS: i) culture tank, ii) makeup water, iii) pre-biofilter, iv) post-biofilter, and v) post-gas conditioning. Concentrations of 3 waterborne hormones – testosterone, 11-ketotestosterone (11-KT), and estradiol (17β-estradiol) – were quantified using enzyme immunoassays (EIA). Estradiol was significantly reduced by ozonation; testosterone and 11-KT were also reduced by ozonation, although these reductions were not observed across all sampling locations and events. Testosterone and 11-KT concentrations, however, were significantly reduced following water passage through the biofilters of both ozonated and non-ozonated RAS. The results of this study demonstrate the potential for ozone to be used in RAS as a means of preventing the accumulation of steroid hormones. Further research is required to assess whether reducing hormones in this manner impacts precocious sexual maturation in RAS-produced Atlantic salmon.     

Start-up performance of a woodchip bioreactor operated end-of-pipe at a commercial fish farm—A case study

There is a need for simple, maintenance-free technologies for removing nitrogen (N) from aquaculture effluents. Denitrifying woodchip bioreactors have been used successfully to remove nitrate-N (NO₃-N) from ground and surface waters and may potentially be applied to dilute aquaculture effluents as well. Real-life applicability in commercial, outdoor fish farms including practical start-up issues such as e.g. time till stable performance and potential leaching are, however, unknown to the industry.This case study consequently investigated the temporal performance of a woodchip bioreactor (12.5 m³) during start-up. The bioreactor was operated end-of-pipe at a commercial, outdoor rainbow trout (Oncorhynchus mykiss) farm in Denmark operated at low recirculation intensity. Applying an empty bed contact time (EBCT) of 5 h, the specific objectives of the study were to resolve: i) how fast the bioreactor would start to remove NO₃-N; ii) how fast steady state was achieved; iii) which NO₃-N removal rates could be attained at the relatively low effluent temperature (∼8 °C) and iv) to which extent any concomitant leaching of phosphorous (P), ammonia or organic matter would occur.In- and outlet grab samples were obtained every 6 h until the bioreactor was in steady state (2 weeks) followed by weekly 24 h pooled samples for another 3 weeks (5 weeks in total). Additional grab samples were obtained from 9 sampling ports within the bioreactor on 3 consecutive days during steady state. Samples were analyzed for dissolved nutrients (total N, nitrate, nitrite, ammonium, total phosphorous, ortho-phosphorous, BOD₅ and COD). In addition, oxygen, temperature and pH were logged every 30 min while sampling and alkalinity were measured once a week.Removal of NO₃-N started immediately and remained stable at 7.06 ± 0.81 g NO₃-N/m³/d (n = 6) throughout the sampling period. Increased effluent NO₂-N concentrations (peaking at 1.14 mg NO₂-N/l after 4–5 days) were transiently observed during the initial 11 days. After that, the woodchip bioreactor was largely in steady state with respect to N-balances corroborated by a close match between filtered total-N (TN_diss) and NO₃-N removal rates. Measurements within the bed showed that the majority of the influent dissolved oxygen (DO) was consumed within the first part of the bioreactor and that NO₃-N removal thereafter proceeded gradually with distance within the bed. Leaching of non-structural, dissolved organic compounds were observed just after startup, causing a short-term (1 week) increase in effluent concentrations of COD, BOD₅, P and ammonium.Additional measurements carried out until 147 days after start-up showed that the woodchip bioreactor continued to remove TN_diss at an average removal rate of 7.81 ± 0.82 g N/m³/d, and that the initial leakage of P stopped altogether.In summary, the study demonstrated that woodchip bioreactors can effectively remove NO₃-N from dilute aquacultural effluents at low temperatures and commercial conditions and that stable performance is achieved within a few weeks.     

The use of acoustic acceleration transmitter tags for monitoring of Atlantic salmon swimming activity in recirculating aquaculture systems (RAS)

Successful operation of recirculating aquaculture systems is dependent on frequent monitoring of the optimal function of water treatment processes in order to maintain environmental conditions for optimal growth and welfare of the fish. Real time monitoring of fish status is however usually not an integrated part of automatized systems within RAS. The aim of this study was to evaluate the use of implanted acoustic acceleration transmitters to monitor Atlantic salmon swimming activity. Twelve salmon post-smolts were individually tagged and distributed in three tanks containing salmon at start density of 50 kg m⁻³. The tagging did not cause any mortality and all individuals increased their body weight during this study. Following initial recovery, acceleration data were continuously logged for one month, including treatment periods with exposure to hyperoxic (170% O₂ saturation) and hypoxic (60% O₂ saturation) conditions, and different tank hydraulic retention times (HRT; 23 and 58 min). Changes in-tank dissolved oxygen levels to hyperoxic and hypoxic conditions reduced the total activity of Atlantic salmon in this study. On the contrary, increased and reduced tank HRT increased the total activity levels. Feeding periods induced a sharp increase in the Atlantic salmon swimming activity, while irregular feeding caused larger oscillations in activity and also lead to increased swimming activity of the tagged fish. Atlantic salmon responded with a maximum recorded total activity to stress caused by technical problems within the system and consequent changes in the RAS environment. The results of this study indicate that Atlantic salmon respond quickly with changed swimming activity to changes in the water quality and acute stress caused by normal management routines within RAS. The use of acoustic acceleration transmitters for real time monitoring of swimming activity within aquaculture production systems may allow for rapid detection of changes in species-specific behavioural welfare indicators and assist in the refinement of best management practices. In addition, acceleration tag could potentially serve as a valuable research tool for behavioural studies, studies on stress and welfare and could allow for better understanding of interaction between fish and RAS environment.     

Effect of different C/N ratios and hydraulic retention times on denitrification in saline,recirculating aquaculture system effluents

Data on operation and performance of cost-effective solutions for end-of-pipe removal of nitrate from land-based saltwater recirculating aquaculture systems (RAS) are scarce but increasingly requested by the aquaculture industry. This study investigated the performance of a (semi)commercial-scale fixed-bed denitrification unit using single sludge for treating effluent from a commercial, saltwater RAS used for production of Atlantic salmon (Salmo salar). A fixed-bed denitrification reactor was fed continuously with 3-days hydrolyzed sludge from the commercial RAS, and was operated at different hydraulic retention times (HRTs; 1.82, 3.64, 5.46, or 7.28 h) or influent C/N ratios (3, 5, 7, or 10). Twenty-four h pooled samples were collected from the inflowing RAS water and the hydrolyzed sludge as well as from the denitrification reactor outlet, and samples were analyzed for nutrients and organic matter content.Nitrate removal rates increased consistently with decreasing HRT (from 64.3 ± 5.2–162.7 ± 22.0 g NO₃-N/m³/d within the HRTs tested) at non-limiting C/N ratios, while nitrate removal efficiencies decreased (from 99.6 ± 0.3–58.2 ± 8.9 %). With increasing influent C/N ratios at constant HRT (3.64 h), nitrate removal rates increased until the removal efficiency was close to 100 % and nitrate concentration in the denitrification reactor became rate-limiting. A maximum nitrate removal rate of 162.7 ± 2.0 g NO₃-N/m³/d was achieved at a HRT of 1.82 h and an influent C/N of 6.6 ± 0.5, while the most efficient use of hydrolyzed sludge (0.19 ± 0.02 g NO₃-N removed/g sCOD supplied) was obtained with a HRT of 3.64 h and a C/N ratio of 2.9. Removal rates of organic matter significantly and consistently increased with decreasing HRT and increasing C/N ratio. In addition, reducing HRT and increasing C/N ratios significantly improved removal of total phosphorus (TP) and PO₄-P.In conclusion, optimal management of the operating parameters (HRT and C/N ratio) in a single-sludge denitrification process can significantly reduce the discharge of nitrogen, organic matter, and phosphorous from land-based saltwater RAS and thus contribute to increased sustainability.     

The effects of long-term 20 mg/L carbon dioxide exposure on the health and performance of Atlantic salmon Salmo salar post-smolts in water recirculation aquaculture systems

Previous research and experience has linked elevated dissolved carbon dioxide (CO₂) to reduced growth performance, poor feed conversion, and a variety of health issues in farm-raised fish, including Atlantic salmon Salmo salar. Supplemental control measures in water recirculation aquaculture systems (RAS) to reduce CO₂ accumulation, however, such as increased water pumping to decrease tank hydraulic retention time, can represent significant costs for operators. We exposed post-smolt S₀ Atlantic salmon (197 ± 2 g, 423 days post-hatch) to either high (20 ± 1 mg/L) or low (8 ± <1 mg/L) dissolved CO₂ in six replicated freshwater RAS for 384 days to investigate differences in performance and health as the salmon were grown to harvest size. All RAS were operated at moderate water exchange rates (1.0% of the total recirculating flow), a 24-h photoperiod was provided, fish were fed to satiation, and densities were maintained between 40 and 80 kg/m³. Over the study period, dissolved oxygen was kept at saturation, mean water temperature was 14.1 ± 0.1 °C, and alkalinity averaged 237 mg/L as CaCO₃. At study’s end, no significant differences in fish weight (high CO₂ mean weight = 2879 ± 35 g; low CO₂ mean weight = 2896 ± 12 g), feed conversion ratio (1.14 ± 0.12 vs. 1.22 ± 0.13, respectively), or thermal growth coefficient (1.45 ± 0.01 vs. 1.46 ± 0.01, respectively), were observed. No significant differences in survival (high CO₂ mean survival = 99.1 ± 0.4%; low CO₂ mean survival = 98.9 ± 0.3%) or culls due to saprolegniasis (3.5 ± 1% vs. 3.0 ± 1%, respectively) were determined, and no nephrocalcinosis was observed through histopathological evaluation. Blood gas and chemistry evaluation revealed higher pCO₂, bicarbonate, and total CO₂, and lower chloride and glucose, in the high CO₂ cohort. Molecular analyses of gill enzyme regulation showed significantly higher expression of Na⁺/K⁺ ATPase α1a in high CO₂ fish at 3-weeks post-challenge, indicating physiological adaptation to the higher CO₂ environment without any noticeable long-term impacts on health or performance. Overall, the results of this study suggest that, at 237 mg/L as CaCO₃ mean alkalinity, post-smolt Atlantic salmon can be raised in freshwater RAS to harvest size with up to 20 mg/L CO₂ without significantly impacting fish health and performance.     

Nitrate removal effectiveness of fluidized sulfur-based autotrophic denitrification biofilters for recirculating aquaculture systems

There is a need to develop practical methods to reduce nitrate–nitrogen loads from recirculating aquaculture systems to facilitate increased food protein production simultaneously with attainment of water quality goals. The most common wastewater denitrification treatment systems utilize methanol-fueled heterotrophs, but sulfur-based autotrophic denitrification may allow a shift away from potentially expensive carbon sources. The objective of this work was to assess the nitrate-reduction potential of fluidized sulfur-based biofilters for treatment of aquaculture wastewater. Three fluidized biofilters (height 3.9 m, diameter 0.31 m; operational volume 0.206 m³) were filled with sulfur particles (0.30 mm effective particle size; static bed depth approximately 0.9 m) and operated in triplicate mode (Phase I: 37–39% expansion; 3.2–3.3 min hydraulic retention time; 860–888 L/(m² min) hydraulic loading rate) and independently to achieve a range of hydraulic retention times (Phase II: 42–13% expansion; 3.2–4.8 min hydraulic retention time). During Phase I, despite only removing 1.57 ± 0.15 and 1.82 ± 0.32 mg NO₃–N/L each pass through the biofilter, removal rates were the highest reported for sulfur-based denitrification systems (0.71 ± 0.07 and 0.80 ± 0.15 g N removed/(L bioreactor-d)). Lower than expected sulfate production and alkalinity consumption indicated some of the nitrate removal was due to heterotrophic denitrification, and thus denitrification was mixotrophic. Microbial analysis indicated the presence of Thiobacillus denitrificans, a widely known autotrophic denitrifier, in addition to several heterotrophic denitrifiers. Phase II showed that longer retention times tended to result in more nitrate removal and sulfate production, but increasing the retention time through flow rate manipulation may create fluidization challenges for these sulfur particles.     

Process requirements for achieving full-flow disinfection of recirculating water using ozonation and UV irradiation

A continuous water disinfection process can be used to prevent the introduction and accumulation of obligate and opportunistic fish pathogens in recirculating aquaculture systems (RAS), especially during a disease outbreak when the causative agent would otherwise proliferate within the system. To proactively prevent the accumulation of fish pathogens, ozonation and ultraviolet (UV) irradiation processes have been used separately or in combination to treat water in RAS before it returns to the fish culture tanks. The objective of the present study was to determine the process requirements necessary to disinfect the full RAS flow, using ozonation followed by UV irradiation, just before the flow was returned to the fish culture tank(s). We found that a proportional-integral (PI) feed-back control loop was able to automatically adjust the concentration of ozone (O₃) generated in the oxygen feed gas (and thus added in the low head oxygenator) in order to maintain the dissolved O₃ residual or ORP at a pre-selected set-point. We determined that it was easier and effective to continuously monitor and automatically control O₃ dose using an oxidative reduction potential (ORP) probe (in comparison to a dissolved ozone probe) that was located at the outlet of the O₃ contact chamber and immediately before water entered the UV irradiation unit. PI control at an ORP set-point of 450 and 525 mv and a dissolved O₃ set-point of 20 ppb provided almost complete full-flow inactivation of heterotrophic bacteria plate counts (i.e., producing <1 cfu/mL) and improved water quality (especially color and %UVT) in a full-scale recirculating system. Achieving this level of treatment required adding a mean dose of approximately 29 ± 3 g O₃ per kg feed. However, because water is treated and reused repeatedly in a water reuse system, the mean daily O₃ demand required to maintain an ORP of 375–525 mV (or at 20 ppb dissolved O₃) was 0.34–0.39 mg/L, which is nearly 10 times lower than what is typically required to disinfect surface water in a single pass treatment. These findings can be used to improve biosecurity and product quality planning by providing a means for continuous water disinfection in controlled intensive RAS.     

Effects of sub-lethal CO2(aq) concentrations on the performance of intensively reared gilthead seabream (Sparus aurata) in brackish water: Flow-through experiments and full-scale RAS results

The effects of sub-lethal CO_2(aq) concentrations were tested for the first time on gilthead seabream (Sparus aurata) juveniles (4–25 g; 64 growth days) and adult (∼300–400 g; 71 d) fish, both in fully controlled pilot tests and the latter also as part of full-scale RAS (recirculating aquaculture system) operation. In the pilot experiments (concentration range 5.2–56.3 mg CO₂/L) the specific growth rate, mortality rate, and physical fish disorders were monitored. In the full scale experiment, two groups of fish, originally from the same batch, were exposed for 197 d to controlled (by NaOH dosage) and uncontrolled pH conditions, resulting in exposure of the fish to significantly different CO_2(aq) concentrations. The pilot results showed, as expected, that the seabream fish grew faster at the lower CO₂ concentrations and that the growth rate of both juveniles and adult fish was only minimally inhibited up to roughly 20 mg CO₂/L (compared to a previously published curve). Mortality rate was considerable only at the highest CO₂ concentration (∼56 mg CO₂/L). Physical irregularities were not observed, apart from abnormally high absence of swim bladder at the highest CO_2(aq) treatment. The (statistically significant) results from the full-scale RAS operation showed that growing gilthead seabream for 197 d at roughly constant and relatively low (∼16 mg/L) CO_2(aq) concentration resulted in fish with ∼10% larger mean weight relative to the fish grown in ponds in which CO_2(aq) was not controlled and its concentration fluctuated daily between 19 and 37 mg/L.     

The effect of nearly closed RAS on the feed intake and growth of Nile tilapia (Oreochromis niloticus), African catfish (Clarias gariepinus) and European eel (Anguilla anguilla)

One of the challenges that Recirculating Aquaculture Systems (RAS) are still facing is the risk that in RAS fish grow less than in flow-through systems due to the accumulation of substances originating from feed, fish or bacteria associated with the water re-use. The present study investigated whether RAS with high and low accumulation levels of these substances affect feed intake and growth of Nile tilapia Oreochromis niloticus, African catfish Clarias gariepinus, and European eel Anguilla Anguilla. One-hundred and twenty individuals of each species were used (start body weights: Nile tilapia 264.8 ± 8.3 g; African catfish 253.2 ± 2.1 g and European eel 66.6 ± 1.3 g). For a period of 39 days, growth and feed intake were compared between high and low accumulation RAS. HIGH accumulation RAS was designed for maximal accumulation of substances in the water by operating the system at nearly-closed conditions (30 L/kg feed/d), using mature biofilters and high feed loads; and (2) LOW accumulation RAS was designed to be a proxy for flow-through systems by operating at high water exchange rates (1500 L/kg feed/d), new biofilters and low feed load. HIGH accumulation RAS induced a reduction in feed intake (42%) and growth (83%) of Nile tilapia, as compared to systems that are a proxy for flow-through conditions. This effect was not observed in European eel and African catfish. The cause of this reduced feed intake and growth rate of Nile tilapia is still unclear and should be addressed in further studies.     

A novel approach for ammonia removal from fresh-water recirculated aquaculture systems,comprising ion exchange and electrochemical regeneration

A new physico-chemical process for ammonia removal from fresh-water recirculated aquaculture systems (RASs) is introduced. The method is based on separating NH₄⁺ from RAS water through an ion-exchange resin, which is subsequently regenerated by simultaneous chemical desorption and indirect electrochemical ammonia oxidation. Approach advantages include (1) only slight temperature dependence and no dependence on bacterial predators and chemical toxins; (2) no startup period is required and the system can be switched on and off at will; and (3) the fish are grown in much lower bacterial concentration, making the potential for both disease and off-flavor, lower. A small pilot scale RAS was operated for 51 d for proving the concept. The system was stocked by 105 tilapia fish (initial weight 35.8 g). The fish, which were maintained at high TAN (total ammonia nitrogen) concentrations (10–23 mgN L⁻¹) and fish density of up to 20 kg m⁻³, grew at a rate identical to their established growth potential. NH_3(aq) concentrations in the fish tank were maintained lower than the assumed toxicity threshold (0.1 mgN L⁻¹) by operating the pond water at low pH (6.5–6.7). The low pH resulted in efficient CO₂ air stripping, and low resultant CO_2(aq) concentrations (<7 mg L⁻¹). Due to efficient solids removal, no nitrification was observed in the fish tank and measured nitrite and nitrate concentrations were very low. The system was operated successfully, first at 10% and then at 5% daily makeup water exchange rate. The normalized operational costs, calculated based on data derived from the pilot operation, amounted to 28.7 $ cent per kg fish feed. The volume of the proposed process was calculated to be ∼13 times smaller than that of a typical RAS biofilter. The results show the process to be highly feasible from both the operational and economical standpoints.     

Microscreen effects on water quality in replicated recirculating aquaculture systems

This study investigated the effects of three microscreen mesh sizes (100, 60 and 20 μm) on water quality and rainbow trout (Oncorhynchus mykiss) performance compared to a control group without microscreens, in triplicated recirculating aquaculture systems (RAS). Operational conditions were kept constant during a 6-week period where the microscreens were manually rinsed three times a day. The effects of microscreen cleaning frequency and nitrification performance were subsequently assessed.Compared to the control group, microscreens removed particles, reduced particulate organic matter, and increased β-values. Particulate parameters reached steady-state in all treatment groups having microscreens at the end of the trial. The time to reach equilibrium seemingly increased with increasing mesh size but the three treatment groups (100, 60 and 20 μm) did not significantly differ at the end of the trial. Increased backwashing frequency over a 24-h period had no further significant effects on the parameters measured. The results demonstrated the role and importance of a microscreen, and showed that mesh size, within the range tested, is less important at long operations under constant conditions.Fish performed similarly in all treatments. Preliminary screening of trout gills did not reveal any pathological changes related to microscreen filtration and the resulting water quality. Biofilter performance was also unaffected, with 0′-order nitrification rates (k_0a) being equivalent for all twelve systems (0.148 ± 0.022 g N m⁻² d⁻¹).Mechanisms for RAS equilibrium establishment, within and between systems with different mesh sizes, are discussed.     

The impact of production intensity on water quality in oxygen enriched,floating enclosures for post-smolt salmon culture

The main aim of the study was to decide the effect of specific water consumption (L/kg/min) and feed load per water flow (g/m³) on the water quality parameters pH, CO₂, total ammonia nitrogen (TAN) and suspended solids (SS) in two large semi-closed containment systems (S-CCS). The reported production parameters (range) in the two S-CCS were specific water consumption (q): 0.04–0.47 L/kg/min and feed load per water flow: 9.0–64 g/m³. The study period was split in two sub-periods; January–May (4.4–7.5 °C), and June–September (7.5–13.2 °C) before a regression model was used to determine the relationship between production intensity (q, feed load) and water quality (pH, CO₂). With the acceptable level of CO₂ defined as ≤10 mg/L, the model predicted a minimum specific water consumption (L/kg/min) between 0.07 (winter) and 0.2 (summer). The predicted maximum feed load per water flow (g/m³) was between 35 (summer) and 45 g/m³ (winter). These calculated limits for production intensity were close to the values earlier reported for smolt or post-smolt production in large, onshore tanks.     

Effect of water recirculation on seawater quality and production of scallop (Pecten maximus) larvae

Scallop larval production systems in Norway have changed from the use of batch to continuous flow through systems (FTS) during the last decade. Energy use to heat water in both larval and spat nurseries is considerable. Two experiments (June 2010 and February 2011) using water recirculation technology (RAS) were performed in large scale systems (3500 L larval tanks) supplied with continuous addition of algal feed, and 20% renewal of seawater.In the RAS a gradual increase in CO₂, decrease in pH and dissolved oxygen was observed over time. This was most obvious during experiment two, when the total organic carbon content increased in both FTS and RAS. The total bacterial number was lower and more stable in FTS than in the RAS. The variations in seawater quality parameters were smaller during the first experiment compared to the second, when values of oxygen saturation were reduced to <70%, pH was 7.8 and NO₃⁻ reached 5 mg L⁻¹. Even though these changes would seem less beneficial for survival and growth of scallop larvae, results showed that the survival at the end of the larval stage was higher in the FTS, but the yield of competent larvae ready for settlement was not significant different (p > 0.05) due to large variations between tanks. The CV% was 28.9% in FTS, while it was 49.9% in RAS. In FTS the mean yield was 40.2%, while it was 26.5% of initial number of larvae in RAS. Large variations in survival and yield were found between the larval tanks as well as gradual reduction in pH and oxygen in RAS tanks. The results indicate that there is a large potential for 80% reduction in water use by utilizing recirculation technology.     

Activated sludge denitrification in marine recirculating aquaculture system effluent using external and internal carbon sources

Stringent environmental legislation in Europe, especially in the Baltic Sea area, limits the discharge of nutrients to natural water bodies, limiting the aquaculture production in the region. Therefore, cost-efficient end-of-pipe treatment technologies to reduce nitrogen (N) discharge are required for the sustainable growth of marine land-based RAS. The following study examined the potential of fed batch reactors (FBR) in treating saline RAS effluents, aiming to define optimal operational conditions and evaluate the activated sludge denitrification capacity using external (acetate, propionate and ethanol) and internal carbon sources (RAS fish organic waste (FOW) and RAS fermented fish organic waste (FFOW)). The results show that between the evaluated operation cycle times (2, 4, and 6 h), the highest nitrate/nitrite removal rate was achieved at an operation cycle time of 2 h (corresponding to a hydraulic retention time of 2.5 h) when acetate was used as a carbon source. The specific denitrification rates were 98.7 ± 3.4 mg NO₃⁻-N/(h g biomass) and 93.2 ± 13.6 mg NO_x⁻-N/(h g biomass), with a resulting volumetric denitrification capacity of 1.20 kg NO₃⁻-N/(m³ reactor d). The usage of external and internal carbon sources at an operation cycle time of 4 h demonstrated that acetate had the highest nitrate removal rate (57.6 ± 6.6 mg N/(h g biomass)), followed by propionate (37.5 ± 6.3 mg NO₃⁻-N/(h g biomass)), ethanol (25.5 ± 6.0 mg NO₃⁻-N/(h g biomass)) and internal carbon sources (7.7 ± 1.6–14.1 ± 2.2 mg NO₃⁻-N/(h g biomass)). No TAN (Total Ammonia Nitrogen) or PO₄^3- accumulation was observed in the effluent when using the external carbon sources, while 0.9 ± 0.5 mg TAN/L and 3.9 ± 1.5 mg PO₄^3--P/L was found in the effluent when using the FOW, and 8.1±0.7 mg TAN/L and 7.3 ± 0.9 mg PO₄^3--P/L when using FFOW. Average sulfide concentrations varied between 0.002 and 0.008 mg S^2-/L when using the acetate, propionate and FOW, while using ethanol resulted in the accumulation of sulfide (0.26 ± 0.17 mg S^2-/L). Altogether, it was demonstrated that FBR has a great potential for end-of-pipe denitrification in marine land-based RAS, with a reliable operation and a reduced reactor volume as compared to the other available technologies. Using acetate, the required reactor volume is less than half of what is needed for other evaluated carbon sources, due to the higher denitrification rate achieved. Additionally, combined use of both internal and external carbon sources would further reduce the operational carbon cost.