Modelling the distribution of fish around an artificial reef

Artificial reefs are a widely used tool aimed at fishery enhancement, and measuring the scale at which fish assemblages associate with these artificial habitat patches can aid reef design and spatial arrangement. The present study used rapidly deployed underwater video (drop cameras) to determine the magnitude and spatial scale of associations between a fish assemblage and a coastal artificial reef. Count data from drop cameras were combined with distance and bathymetry information to create a suite of explanatory generalised linear mixed models (GLMMs). The GLMMs showed that artificial reefs can influence surrounding fish abundance, but that the magnitude and scale is species specific. Three of the eight taxonomic groups examined showed a positive association with the artificial reef (with model fit poor for the remaining groups); and depth and bottom cover were also influential variables. The spatial scales of these associations with the artificial reef were small, and it was generally the presence of reef (i.e. a reef bottom type) that explained more variation in fish abundance than did distance to reef. The schooling baitfish yellowtail scad was an exception, and had elevated abundance >50 m from the artificial reef. Further distribution modelling of artificial reefs will benefit species-specific design and management of artificial reefs.