Graduation Year


Document Type




Degree Granting Department

Civil and Environmental Engineering

Major Professor

Sarina Ergas

Co-Major Professor

Maya Trotz


Constructed Wetland, Geotube, Moving Bed Biofilm Reactor, Recirculating Aquaculture System, STELLA


Integrated aquaculture systems (IAS) are a type of recirculating aquaculture systems (RAS) where the wastewater is treated and returned to the fish tanks. The important difference between the two is that in an IAS, wastes from the aquaculture component are recovered as fertilizer to produce an agricultural product whereas in an RAS, waste organics, nutrients and solids are treated and discharged. A pilot marine IAS at Mote Aquaculture Research Park in Sarasota, FL was studied for this project. Water quality monitoring, measurements of fish health and growth rates of fish and plants were performed over a two-year period to determine the effectiveness of the system in producing fish and plant products and removing pollutants. The goal of this portion of the project was to develop, calibrate and evaluate a model of the system, to understand the nitrogen transformations within the Mote IAS and to investigate other potential configurations of the Mote IAS.

The model was divided into the various compartments to simulate each stage of the system, which included fish tanks, a drum filter for solids removal, and moving bed bioreactor (MBBR) for nitrification and disinfection. A solids tank after the drum filter was used to store the drum filter effluent slurry, which was then divided between three treatment processes: a geotube, a sand filter followed by a plant bed, and a plant bed alone. Nitrogen species modeled were particulate organic nitrogen (PON), dissolved organic nitrogen (DON), ammonium and nitrate. Of the physical components of the IAS, models of the MBBR and the two plant raceways included physical, chemical and biological nitrogen transformation processes. The sand filter, solids tank and geotube models were simple mass balances, incorporating factional removals of each species based on the observed data. Other variables modeled included temperature, dissolved oxygen, volatile suspended solids and chemical oxygen demand concentrations. The model was built in a computer program, STELLATM, to simulate the Mote IAS.

The model calibration involved experimental, literature and calibrated parameters. Parameters were adjusted until the model's output was a best fit to the observed data by minimizing the sum of the squared residuals. During the sensitivity analysis, two model parameters caused large variations in the model output. The denitrifier constant caused the most variation to the model's output followed by the denitrifier fraction of volatile suspended solids.

Of the removal processes, denitrification was the largest nitrogen removal mechanism from the model, accounting for 59% and 55% of the nitrogen removed from the south and north plant raceways respectively. Plant and soil uptake represented only 0.2% of the overall nitrogen removal processes followed by 0.1% by sedimentation.

Finally, the model was used to investigate other treatment designs if the Mote IAS was redesigned. The first option involved a geotube and one plant raceway in series to treat the solid waste while the second option did not have a geotube, but two plant raceways. The first option was the most effective at removing nitrogen while the second was as effective as the original system and would cost less.