CalCOFI larval fish community temperature index changed 0.20 °C per °C of year-to-year 10 m temperature in 1951–2021, 59–117% of the spatial relation; a 1951–2013 fit predicted the 2014–2016 heatwave

Question: does the thermal composition of larval fish in the southern California Current change with year-to-year ocean temperature, how much relative to the change between warm and cold places, and did the 2014–2016 marine heatwave agree with earlier years? Data: 12,692 CalCOFI oblique net tows (ring net 1951–1977, bongo net 1978–2021) on the 66 core stations in 62 years, each with an in-situ 10 m temperature from a cast at the same station within 3 days, and 93 larval fish genera. Method: the species temperature index (STI) of a genus is the long-term station-season 10 m temperature where its larvae occur. It does not use year-to-year temperature. The community temperature index (CTI) of a tow is the log-density-weighted mean STI. We removed station-season means and calculated CTI-temperature slopes at different scales, with an STI-shuffle null, year-block and station bootstraps, and an out-of-sample test. Results: the year-to-year slope was 0.20 °C per °C (95% interval 0.16–0.23; shuffle p = 0.002). The spatial slope across stations was 0.34, 0.23 with distance offshore controlled, and 0.17 with distance offshore and line controlled. Thus the year-to-year slope is 59% (44–78%) to 117% of the spatial relation. The previous year's SST added 0.07 °C per °C (SE 0.02). Slopes of 2- to 5-year means were 0.20–0.23. A fit to 1951–2013 predicted a mean 2014–2016 CTI anomaly of +0.241 °C; the observed value was +0.248 °C. In 2015, temperature was 1.91 °C above normal, outside the earlier range, and the CTI anomaly was +0.391 °C (predicted +0.378 °C). Increases of warm-niche mesopelagic genera (Vinciguerria, Triphoturus, Cyclothone) gave 92% of the response; decreases of cold-niche genera gave 18%. The results were stable in nine variants (gridded ERSSTv5 SST, 0–100 m temperature, species units, other weights, cross-fitted STI, fewer genera, no fished genera). Limitations: the spatial benchmark depends on how much of the onshore-offshore habitat gradient one removes; the 1951–2021 trends (CTI +0.029 °C and temperature +0.10 °C per decade) do not determine the decadal tracking rate (ratio 0.84, interval 0.03–2.6); the 1969 change of tow depth could not be calibrated, but a paired 1977–1978 test shows no offset for the change of net (+0.02 °C, −0.04 to +0.08); STI is a position inside the CalCOFI grid only.

CalCOFI · ichthyoplankton · larval fish · community temperature index · thermal tracking · marine heatwave · California Current · space-for-time · climate change · mesopelagic fish