Foam fractionation and ozonation in freshwater recirculation aquaculture systems

Foam fractionation is often considered an ineffective way of removing organic matter from freshwater due to the low surface tension of the water. There is, however, a lack of studies testing foam fractionation efficiency in replicated freshwater recirculating aquaculture systems (RAS). Foam fractionation can be applied with or without ozone. Ozone is a strong oxidiser previously shown to improve water quality and protein skimmer efficiency. To test the efficiency of foam fractionation and ozonation (20 g O₃ kg^-1 feed) separately and in combination in freshwater RAS, a two-by-two factorial trial was conducted with each main factor at two levels (applied or not applied). Each treatment combination was carried out in triplicates using 12 replicated pilot scale RAS stocked with juvenile rainbow trout (Oncorhynchus mykiss) and operated at a feed loading of 1.66 kg feed m^-3 make-up water. The trial lasted 8 weeks and samples were obtained once a week. Ozone applied by itself significantly reduced the number of particles (83%), bacterial activity (48%) and particulate BOD₅ (5-days biochemical oxygen demand; 54%), and increased ultra violet transmittance (UVT; 43%) compared to the untreated control group. Foam fractionation by itself lead to significant reductions in particle numbers and volume (58% and 62%, respectively), turbidity (62%), bacterial activity (54%) and total BOD₅ (51%). A combination of both treatments resulted in a significant additional improvement of important water quality variables, including a 75% reduction in total BOD₅, 79% reduction in turbidity, 89% reduction in particle numbers and 90% reduction in bacterial activity compared to the control. The removal efficiencies were within the same range as those observed in previous studies conducted with foam fractionators in saltwater systems (with or without ozone), corroborating that foam fractionation may become a useful tool for controlling organic matter build-up and bacterial loads in freshwater RAS.