Location: Cool and Cold Water Aquaculture Research
2024 Annual Report
Objectives
Objective 1. Improve fish health, performance, and welfare in recirculating aquaculture systems.
Sub-objective 1.1 Evaluate salmonids grown to market size in a semi-commercial scale freshwater RAS.
a): Collaborate with NCWMAC to evaluate multiple strains of Atlantic salmon and their performance in a RAS environment.
b): Assess genetic strain of steelhead (including USDA-strain rainbow trout) raised to 4kg in a RAS environment.
Sub-objective 1.2 Assess environmental manipulation to reduce maturation in mixed-sex diploid Atlantic salmon.
Sub-objective 1.3 Improve biological monitoring and management of salmonids in RAS through technological integration of next-generation biomonitors.
Objective 2. Support land-based salmonid recirculating aquaculture systems production through increased technological and operational efficiencies and novel, supplemental revenue streams.
Sub-objective 2.1 Evaluate methodologies to convert RAS waste to value-added products.
1a: Assess feasibility of new composting technologies of RAS waste solids and their capacity to generate sellable products.
1b: Assess feasibility of anaerobic digestion of RAS waste solids to generate biogas/energy.
Sub-objective 2.2 Assess novel methods to improve RAS water quality, including optimized integration of membrane biological reactors.
Sub-objective 2.3 Pilot and evaluate new computing technologies for RAS integration to optimize system operational efficiencies.
Approach
The domestic salmonid aquaculture industry is currently experiencing a significant departure from traditional farming practices, as evidenced by recent, substantial capital investment in large-scale land-based, closed-containment facilities utilizing water recirculation aquaculture system (RAS) technologies. While this is an encouraging evolution for U.S. aquaculture overall, this relatively new approach to raising market-sized Atlantic salmon, steelhead trout, and other economically important species is still a frontier in agriculture, remains largely untested at commercial scale, and requires significant refinement and optimization in technological, biological, and economic methods and strategies. The Conservation Fund Freshwater Institute (TCFFI), an extramural program of the USDA-ARS, has been at the forefront of RAS technology research and development for over two decades, and at present we are uniquely suited to continue serving this growing agricultural sector through focused, industry-relevant research and innovation. Our next 5-year project plan seeks to address critical areas that are necessary to support the sustainable growth of the U.S. land-based, closed-containment aquaculture industry; specifically, our objectives fall under two broad categories aimed at improving i) the biological performance of salmonids in RAS, and ii) the technical and economic efficiencies of land-based closed-containment operations. Research activities will include identifying genetic strains of Atlantic salmon and steelhead for optimal performance in RAS, assessing methods to reduce early sexual maturation and improve water quality, developing next-generation biomonitors and computing technologies to improve fish health management and RAS environmental control, and developing means for RAS producers to monetize waste streams for enhanced economic viability.
Progress Report
In support of Sub-objective 1.1, salmon strains growout trial data were compiled and compared among the three participating research institutions – National Cold Water Marine Aquaculture Center (Franklin, Maine), Northern Aquaculture Demonstration Facility (Bayfield, Wisconsin), and The Freshwater Institute (Shepherdstown, West Virginia). At the time of report submission, study data are being analyzed for peer-reviewed journal manuscript submission in late 2024. Preliminary results have been presented to industry stakeholders at two scientific conferences: Aquaculture American 2024 (San Antonio, Texas – February 2024) and Recirculating Aquaculture Systems RASTech 2024 (Charlotte, North Carolina – June 2024). Results clearly indicate that the St. John River Atlantic salmon strain performs significantly better across the different RAS rearing environments, in terms of growth, survival, fillet yield, reduced early maturation, etc., versus the Gaspe strain. These results have informed research plans for the next 5-year USDA-ARS cycle to investigate within-strain St. John River salmon families and whether superior RAS performance is a heritable trait.
Sub-objective 1.2 has been completed, with all milestones met in FY2022.
Progress towards Sub-objective 1.3 has included the preparation and submission of a peer-reviewed publication manuscript to the journal Aquacultural Engineering, titled “Determining a safe nitrate-nitrogen threshold for post-smolt Atlantic salmon Salmo salar production in RAS with assistance from heart rate bio-loggers.” This manuscript is currently in journal review at the time of report submission. Further technology transfer is planned later this year with the presentation of study findings at the Smolt Production Workshop in Sunndalsøra, Norway in October 2024.
In support of Sub-objective 2.1, research activities have included i) a pilot-scale composting study that assessed the feasibility of composting aquaculture waste solids, ii) a follow-up study to investigate the potential of utilizing fish mortalities as a co-composting substrate with aquaculture waste solids, and iii) procurement of pumps, mixers, heat exchanger, and related accessories for the construction of a pilot-scale anaerobic digester. Both studies showed that aquaculture waste solids could be composted successfully. The final product was a compost mulch due to the addition of wood shavings as the bulking agent. Further dewatering of the aquaculture waste solids before composting could reduce the need to add a large volume of the bulking agent. Compost testing, according to US Compost Council seal of testing assuance (STA) certified standards, showed that the compost was mature, not phytotoxic, had low heavy metal concentrations, and low levels of pathogens (Salmonella and fecal coliforms), substantially below EPA specified limits, meeting the requirements for Class A biosolids. The final compost was also tested for per- and polyfluoroalkyl substances (PFAS) using EPA Method 1633 for biosolids, and all 40 analytes tested were below the limit of quantification, with a majority below the limit of detection. Results from the first composting study were presented at the RASTech 2024 (Charlotte, North Carolina). At the time of report submission, a manuscript with the results from both composting studies is being prepared for publication in a peer-reviewed journal. It is expected to be submitted in Fall 2024. A second manuscript on the effect of two different fish feeds on waste characteristics and their potential for biogas production from anaerobic digestion is also being prepared. Construction of the pilot-scale anaerobic digester is ongoing and expected to be operational in Fall 2024. The feasibility study is anticipated to be completed by the end of 2024. Project delays have occurred due to limited vendor availability for procuring a custom-designed automation and control system for the anaerobic digester.
Progress towards Sub-objective 2.2 has included successfully completing the research evaluating the feasibility of integrating multi-vessel membrane biological reactors (MBRs) with RAS. Study results indicated that i) MBRs are a viable approach to RAS water treatment, ii) that rainbow trout performance was unaffected by including MBRs in RAS, and iii) reusing MBR permeate reduced RAS water use by 94%. At the time of report submission, a manuscript is being prepared for peer-reviewed journal submission in late 2024. The findings from this study have also been presented to stakeholders at RASTech 2024 (Charlotte, North Carolina – June 2024).
In support of Sub-objective 2.3, a study was carried out to develop an Artificial Intelligence (AI)-enabled handheld device (‘FilletCam AI’) for precise, repeatable, rapid, real-time, and objective color evaluation of fish fillets. The AI model demonstrated robust performance in terms of detecting fillet and color palettes, achieving a mean average precision (mAP) of 99.5% and an F1 score of 0.99 at 100 epochs. The FilletCam and colorimeter color score predictions were compared with the expert scores to evaluate device performance. The developed FilletCam performed satisfactorily, and for most of the fillets, the device either accurately predicted the color score (22 out of 30 fillets) or predicted scores deviated by one to two points. The findings of this study are being compiled in a manuscript for a peer-reviewed publication. Furthermore, AI models are being refined to include fillet defect detection capability in the FilletCam. Further efforts to support Subobjective 2.3 included testing the feasibility of using computer-simulated virtual images to train a robust fish detection AI model for the RAS culture environment. The ‘virtual model’ performances were compared with models trained with real-world images and combinations of real and virtual images. The results of the study indicate that the virtual model trained solely with computer-simulated images could not perform satisfactorily (mAP = 62.8%, F1 score = 0.61) to detect fish in the real RAS environment; however, replacing a small number of the virtual images with real images in the training dataset significantly improved the model performance. The mixed model trained with 630 virtual and 70 real images (virtual to real image ratio: 90:10) achieved mAP and F1 scores of 91.8% and 0.87, respectively. Furthermore, the training time cost for the mixed model was seven times lower than the ‘real model’, respectively. This study is currently under review for peer-review publication.
Accomplishments
1. Utilizing artificial intelligence for rapid, real-time color evaluation of fish fillets. The color of salmon fillets is an important quality influencing consumer choice at the seafood counter. As such, salmon producers expend considerable effort to ensure that their fish products are visually appealing to potential retail and wholesale customers. However, traditional methods used to quantify the color of salmon fillets are time-consuming, subjective, and often unreliable. Extramural ARS scientists in Shepherdstown, West Virginia, have developed an artificial intelligence (AI)-enabled handheld device for rapid, accurate, objective, and real-time color evaluation of salmon fillets, using a state-of-the-art, one-stage convolutional neural network model. Known as ‘FilletCam AI’, this technology improves reliability of fillet color assessment with reduced time and effort, thereby enhancing production efficiency and product consistency and quality.
2. Characterization and evaluation of compost derived from aquaculture waste solids. The economic viability of land-based fish farms using recirculating aquaculture system (RAS) technologies can be supported by research that converts farm waste materials into sellable products. Specifically, the conversion of waste solids into compost products can (i) provide farmers with additional revenue, and (ii) prevent waste from potentially impacting the environment. Extramural ARS scientists in Shepherdstown, West Virginia, have developed protocols to convert aquaculture solids to compost and have demonstrated that compost derived from aquaculture solids is not toxic to plants, has very low concentrations of heavy metals and human pathogens, and contains little to no PFAS (i.e., below the limit of quantification, and for the most part below the level of detection). These results will assist RAS farmers with developing additional revenue through production of sellable compost and will assure RAS farmers and customers regarding the safety and quality of aquaculture solids-derived compost.
3. A low-ration feeding strategy to minimize fish weight loss during depuration. Recirculating aquaculture systems (RAS) accumulate microbial biofilms that can release “off-flavor” compounds, which are taken up by fish and impart objectionable flavors to fillets. Therefore, RAS-raised fish require depuration (i.e., transferring fish to separate, biofilm-free systems before harvest to rid the flesh of off-flavors). Previous research has indicated that Atlantic salmon lose weight while fasting during depuration, which reduces farmgate revenue. Extramural ARS scientists in Shepherdstown, West Virginia, evaluated a low-ration approach over the depuration period to assess both weight loss and depuration efficacy. These scientists determined that salmon lost only 0.3% body weight (compared to 1.1% in fasted fish) while being fed at a low-ration rate, with no impact on off-flavor elimination. RAS salmon producers can significantly benefit from these findings by altering their depuration protocols to minimize fish weight loss before harvest, while maintaining high product quality through effective off-flavor purging.
Review Publications
Davidson, Iii, J., Schrader, K., May, T., Knight, A., Harries, M.D. 2023. Evaluating the feasibility of feeding RAS-produced Atlantic salmon (Salmo salar) during the depuration process: Effects on fish weight loss and off-flavor remediation. Journal of Applied Aquaculture. 36(2):436-456. https://doi.org/10.1080/10454438.2023.2259892.
Lovy, J., Iwanowicz, L.R., Welch, T.J., Allam, B., Getchell, R., Geraci-Yee, S., Good, C., Snyder, J., Raines, C., Das, N. 2024. Seasonal mortality of Atlantic menhaden is associated with neurologic disease caused by a virulent clone of Vibrio anguillarum: Fish kills and implications for biosecurity. Transboundary and Emerging Diseases. 18:8816604. https://doi.org/10.1155/2024/8816604.
Liu, D., Straus, D.L., Pedersen, L., Good, C., Lazado, C.C., Meinelt, T. 2024. Towards sustainable water disinfection with peracetic acid in aquaculture: A review. Reviews in Aquaculture. Article 12915. https://doi.org/10.1111/raq.12915.
Crouse, C., Knight, A., May, T., Davidson, J., Good, C. 2023. Performance, processing yields, and fillet composition of specific United States diploid and triploid rainbow trout (Oncorhynchus mykiss) lines reared in a semi-commercial scale freshwater recirculating aquaculture system. Aquaculture Reports. 33. Article 101794. https://doi.org/10.1016/j.aqrep.2023.101794.