Microbiome response to foam fractionation and ozonation in RAS

Efficient water treatment is required to maintain high water quality and control microbial growth in recirculating aquaculture systems (RAS). Here, we examined the effects of two treatment methods, ozonation and foam fractionation, separately and combined, on the microbiology in twelve identical experimental RAS with rainbow trout (Oncorhyncus mykiss) during 8 weeks. Microbes suspended in water and growing in biofilter biofilms were examined using flow cytometry analysis and high throughput sequencing of the 16S rRNA gene. The results showed that foam fractionation did not cause large changes in abundance or overall community composition of free-living microbes. Instead, through decreasing the organic matter availability in water, it targeted specific microbial taxa, leading to e.g. decreased potential for off-flavor production. In contrast, ozonation was found to have a profound impact on the system microbiology, by reducing the overall cell abundance, increasing microbial dead/live ratio, and changing the community composition of both free-living and biofilm microbes. Ozonation increased the abundance of certain key microbial taxa adapted to low carbon conditions, which might form a stable and more abundant community under a prolonged ozone dosing. Combining the two treatment methods did not provide any additional benefits as compared to ozonation solely, corroborating the high disinfection potential of ozone. However, ozone had only a minor impact on biofilter microbial communities, which were, in general, more resistant to water treatment than water communities. Water treatment had no effect on the overall genetic nitrification potential in the biofilter biofilms. However, foam fractionation led to changes in the nitrifying microbial community in biofilter, increasing the abundance of Nitrospira conducting complete ammonia oxidation to nitrate (comammox). Altogether, the results obtained indicate that although these two water treatment methods have similar outcomes on physico-chemical water quality and microbial activity, their underlying mechanisms are different, potentially leading to different outcomes under the long-term application.