North Sea fish latitude did not follow winter temperature in 2002-2025 as Perry et al. (2005) found for 1977-2001, in two IBTS survey seasons

Perry et al. (2005) found that the mean latitude of 15 of 36 North Sea demersal fish taxa moved with winter warming in 1977-2001. We tested this result out of sample with the ICES NS-IBTS survey in 2002-2025 in quarter 3 (the season of the original survey) and quarter 1. We used fixed ICES-rectangle frames to control changes in haul allocation, and HadISST December-March sea temperature as a 5-year mean. In quarter 1 for 1980-2001, 11 of 14 shifting taxa had a latitude slope in the Perry direction (p = 0.029), with a mean of +29 km per degree C. In 2002-2025, only 7 of 15 (quarter 3) and 8 of 14 (quarter 1) taxa had the Perry direction. The mean direction-signed slopes were -15 km per degree C (95% interval -51 to +14) and -17 (-87 to +42). Shifting taxa did not respond more than non-shifting taxa. Temperature models trained on 1980-2001 had a larger 2002-2025 error than a recent mean for 19 of 30 taxa. The assemblage deepening of Dulvy et al. (2008) stopped after 2004 with fixed rectangle depths but partly continued with haul depths. An exploratory era comparison gave small northward displacements (median 10 km), which were not robust to removal of one country. Limitations: the GOV survey is not the exact English survey of the original study, SST is a proxy of bottom temperature, the series are short, and species identification and fishing pressure are not controlled.

climate change · range shift · North Sea · demersal fish · IBTS · DATRAS · out-of-sample test · replication · observation bias · bottom temperature · null result