Long-term series demonstrate small-scale differences in trends within fish assemblages explained by climate variability

Over 30 years (1985–2018) of environmental monitoring data in the Belgian part of the North Sea allowed us to study common and individual species trends within three spatially distinct demersal and benthopelagic fish assemblages in relation to climate-related variability. A combination of multivariate (dynamic factor analysis) and univariate modelling techniques (Trendspotter and Spearman rank correlations) showed that the observed trends in fish densities in two coastal (mud and fine sand) assemblages were best explained by sea surface temperature (SST)-related variables and the Atlantic multidecadal oscillation winter index (AMOwinter). Higher fish densities seem to be linked with extreme cold winters in the year before (e.g. 1985, 1996, 2010), while warmer years may be linked to lower fish densities (fine sand assemblage) a few years later. Trends for the more offshore coarse sand fish assemblage were rather related to the North-Atlantic Oscillation index (NAO) (next to AMOwinter), coupled to the circulation and inflow of warmer saline waters through the English Channel. The temporal decrease in densities in this assemblage may partially reflect climate and subsequent biological regime shifts, as described for the broader North Sea region.