Date of Award

Fall 2020

Project Type


Program or Major

Biological Sciences

Degree Name

Master of Science

First Advisor

David L. Berlinsky

Second Advisor

Peter S. Konjoian

Third Advisor

Jessica Bolker


While aquaculture production accounts for half of the world fish production, a growing problem emerges with the amount of effluent being produced. Waste treatment of aquaculture effluent is expensive and energy- intensive as conventional approaches to waste remediation have remained mostly unchanged. To improve the economic sustainability, the aquaculture industry needs to integrate with other production systems similarly as terrestrial animal agriculture has done with soil-based crop production. Integrating waste production in a wastes-to-resources approach as fertilizer for hydroponic cropping systems will allow aquaculture producers to monetize waste treatment. However, a full accounting of aquaculture nutrient production is necessary to develop a strategy to monetize costly effluent treatment. Capturing fish waste from aquaculture facilities provide an opportunity to offset operational costs by producing a naturally derived nutrient source as fertilizer.

Three replicate recirculating aquaculture systems (RAS) were designed and operated under pilot-scale production conditions to evaluate plant-available nutrient production from two commonly grown aquaculture species, tilapia (Oreochromis niloticus) and rainbow trout (Oncorhynchus mykiss). A nutrient mass balance was conducted while the research systems operated under “pseudo-steady state” conditions. Pseudo-steady state was defined as consistent feeding and waste production activity during periods of fish growth and increasing feed demands while still accounting for fish growth and increasing feed demands. The macro-nutrients Ca, K, Mg, N, and P and micro-nutrients B, Cl, Cu, Fe, Mn, Mo, S, and Zn were analyzed over an 81-day period. Both the tilapia and trout nutrient production experiments revealed that all nutrients required for hydroponic crop production were present and available in the system culture water and effluent streams.

Macro-nutrients Ca, K, Mg, P, and N, and micro-nutrients, Cl, Mo, and S were observed primarily in the liquid portion of the wastewater and micro-nutrients B, Cu, Fe, Mn, and Zn were primarily observed in the particulate waste. The results of the first experiment indicated that tilapia excreted 3.39 ± 0.55 g Cu, 10.78 ± 1.90 g Fe, 5.61 ± 1.78 g Mn, 0.23 ± 0.08 g Mo, and 7.26 ± 0.89 Zn, per 100kg feed daily. Many of the tilapia nutrient production rates were determined to be statistically different between systems due to dilution and limits of measurement, notably -4.36 ± 4.78 g B, -76.71 ± 350.20 g Cl, -19.97 ± 163.60 g S, 1172.44 ± 706.72 g Ca, 405.27 ± 740.68 g K, 181.72 ± 196.13 g Mg, 704.34 ± 582.05 g P, and 2896.13 ± 4133.70 g Total Nitrogen (TN), per 100kg feed. The difficulties surrounding the accurate characterization of nutrient production from tilapia RAS were resolved and strict sampling procedures applied to the second nutrient mass balance experiment measuring nutrient production from rainbow trout in RAS.

Rainbow trout excreted nutrient production was 706.29 ± 49.58 g Cl, 1.01 ± 0.04 g Cu, 13.41 ± 0.51 g Fe, 7.08 ± 0.71 g Mn, 3.11 ± 0.57 g Mo, 312.95 ± 45.59 g S, 11.95 ± 0.58 g Zn. 2043.37 ± 29.18 g Ca, 659.48 ± 51.15 g K, 445.58 ± 7.61 g Mg, 690.11 ± 42.57 g P, and 5729.49 ± 540.33 g TN, per 100kg feed. It is important to distinguish if the nutrients would be directly available as a fertilizer. This study found that Cl, Mo, S, Ca, K, Mg, P, and N are nutrients solubilized in liquid portion of rainbow trout waste rendering them immediately available for plant uptake. Alternately, B, Cu, Fe, Mn, and Zn, were retained in the solid particulate portion of the rainbow trout waste stream. Nutrients retained in the solid particulates require mineralization to make these nutrients plant available. Differences in nutrient production between the two species are due to variation in the feed composition and physiological distinctions such as gut length and muscle tissue composition.

The results from these experiments were inconsistent with the previous literature and differences are likely due to experimental design, system design, feed, fish species, and dilution effects. Experimental design is the key factor that limited the determination of nutrient production in this research because no tracer was used in the diet which would have allowed for a full accounting of nutrients assimilated and expelled by the fish.

This research supports the need to establish a predictive model for aquaculture-derived nutrient production for integration with other crop production systems. The results from this study demonstrate that nutrient reuse from RAS is possible for hydroponic crop production, but treatment of RAS effluent will be required to fully develop a valuable nutrient source as many of the nutrients are trapped in the solid particulate form.