Long-term changes in the spatio-temporal distribution of snailfish Liparis tanakae in the Yellow Sea under fishing and environmental changes

Tanaka's snailfish (Liparis tanakae) is a low-economic but ecologically important fish in the Yellow Sea, which is one of the most threatened marine ecosystems in the world due to environmental changes and human activities. Although it serves as both a dominant species and an apex predator, our knowledge about the long-term changes in the spatio-temporal distribution of this snailfish remains limited in the threatened ecosystem. In this study, by developing eight alternative vector autoregressive spatio-temporal (VAST) models including various combinations of sea surface temperature (SST), fishing pressure (FP) and the density of the prey (DP), we investigated the spatio-temporal distribution patterns of snailfish based on fishery-independent surveys conducted between 2003 and 2019 and examined the relative importance of different covariates. We found that SST was the most important factor in explaining variation in encounter probability and DP was the most important factor in explaining temporal variation in biomass density of snailfish. Surprisingly, incorporation of FP in the spatio-temporal models neither improved explanation of the variance in encounter probability nor biomass density. Based on Akaike's information criterion, we selected a spatio-temporal model with SST in preference to seven alternative models. The inter-annual distribution range of snailfish was relatively stable whereas the spatial patterns varied over time. In 2003-2006 and 2011, the hotspots of snailfish were widely distributed throughout almost the entire Yellow Sea area. In contrast, in other survey years, especially in 2007-2009, 2015-2016 and 2019, the distribution was more concentrated within the central Yellow Sea. No significant shift in centers of gravity (COGs) was detected for the population. The estimated effective area occupied correlated significantly with biomass density of snailfish (r = -0.71, P< 0.05). Outputs from this study enhanced our understanding of how and the extent to which multiple pressures influence the observed long-term changes in spatio-temporal distribution of snailfish in the Yellow Sea.