Pelagic fish abundance in relation to regional environmental variation in the Gulf of Finland, northern Baltic Sea

This study applies variation partitioning to analyse spatial patterns in hydroacoustic estimates of fish abundance in relation to regional variation in the hydrography, food resources, and geography of the Gulf of Finland, northern Baltic Sea. Using variation partitioning based on generalized additive models, daily fluctuations in hydroacoustic abundance estimates were first eliminated, and the remaining variation in fish abundance was decomposed into independent and joint effects of hydrography, food resources, and geography. The independent effect of geographic variables (spatial location and water depth) captured the largest fraction of the variation (9.3%) in the fish-abundance patterns, whereas the independent effects of hydrography (5.8%) and food resources (5.6%) captured slightly less. However, a considerable portion of the variation in fish-abundance patterns was accounted for by the joint effects of explanatory variables and may therefore be causally related to two or all three groups of variables. The model applied efficiently eliminated the spatial autocorrelation in the fish abundance between sampling units, especially at distances >2000 m. At smaller scales, the residual autocorrelation may have been due to fish behavioural patterns independent of the explanatory variables in this analysis.