Background:
Salmon aquaculture involves freshwater and seawater phases. Recently there has been an increase in multifactorial gill health challenges during the seawater phase which has led to an urgent need to understand the gill microbiome. There is a lack of understanding on what drives the composition of the gill microbiome, and the influence the freshwater stage has on its long-term composition. We characterise the gill microbiome from seven cohorts of Atlantic salmon raised in six different freshwater operational systems—recirculating aquaculture system (RAS), flowthrough (FT) and loch-based system, prior to and after transfer to seven seawater farms, over two different input seasons, S0 (2018) and S1 (2019).

Results:
Using the V1-V2 region of the 16S rRNA gene, we produced amplicon libraries absent of host contamination. We showed that hatchery system influenced the gill microbiome (PERMAOVA R2 = 0.226, p < 0.001). Loch and FT systems were more similar to each other than the three RAS systems, which clustered together. On transfer to sea, the gill microbiomes of all fish changed and became more similar irrespective of the initial hatchery system, seawater farm location or season of input. Even though the gill microbiome among seawater farm locations were different between locations (PERMAOVA R2 = 0.528, p < 0.001), a clustering of the gill microbiomes by hatchery system of origin was still observed 7–25 days after transfer (PERMAOVA R = 0.164, p < 0.001). Core microbiomes at genera level were observed among all fish in addition to freshwater only, and seawater only. At ASV level core microbiomes were observed among FT and loch freshwater systems only and among all seawater salmon. The gill microbiome and surrounding water at each hatchery had more shared ASVs than seawater farms.

Conclusion:
We showed hatchery system, loch, FT or RAS, significantly impacted the gill microbiome. On transfer to sea, the microbiomes changed and became more similar. After transfer, the individual sites to which the fish were transferred has a significant influence on microbiome composition, but interesting some clustering by hatchery system remained. Future gill disease mitigation methods that target enhancing the gill microbiome may be most effective in the freshwater stage, as there were more shared ASVs between water and gill at hatchery, compared to at sea.