Southern Ocean zooplankton thermal composition followed daily SST anomalies at 39% of the spatial slope in SO-CPR samples (60–180°E, 2003–2023), but season and decade tracking was not resolved

Question: does the thermal composition of Southern Ocean surface zooplankton change with sea surface temperature (SST) across places, from day to day, from season to season, and over 2003–2023? Tarling et al. (2017) reported that copepod communities did not move south with warming. Data: the SO-CPR Survey (Darwin Core Archive v1.14, AADC, CC BY 4.0) gave 41,759 tow segments of 5 nautical miles at 60–180°E and 40–70°S, October–March, in 21 seasons (2003–2023). Each segment has the same-day NOAA OISST v2.1 SST. Identification practice changed during the survey, so we pooled 294 taxon codes into 63 stable genus or group units. Method: the species temperature index (STI) is the mean SST where a unit occurs. The community temperature index (CTI) is the log-count-weighted mean STI. One model with cell, month and night terms gives the slope on SST climatology (spatial) and on SST anomaly (temporal). We used season bootstraps, two STI-shuffle nulls, 9 variants and leave-one-unit-out runs. Results: the spatial slope was 0.136 °C per °C (95% interval 0.107–0.165). The daily-anomaly slope was 0.053 (0.027–0.077), 39% (19–60%) of the spatial slope, and stayed positive in all non-sector variants (0.032–0.082). It was above a null with STI shuffled among units of similar frequency (p = 0.001) but not above a free shuffle (p = 0.10), because a few common units control the CTI. The subantarctic copepod Neocalanus tonsus gave 26% of the response. The slope was clear only at 110–150°E (0.098). The season-scale anomaly slope was 0.050 (−0.008 to 0.106), so the data do not separate no tracking from strong tracking at that scale. SST at the segments increased only 0.11 °C per decade, and the CTI trend was −0.013 °C per decade (−0.047 to +0.019). Cells with faster warming had faster CTI increase, but only when the 150–180°E sector was included (60–150°E: −0.21, −0.83 to 0.42). Conclusion: at the scale of days and tens of kilometres, the community moves partly with warm and cold water. These data cannot test season or decade tracking: a 21-year CTI trend is detectable only with more than about 0.35 °C per decade of warming at full tracking. Limitations: identification changes between periods and laboratories, OISST anomaly errors that decrease the daily slope, near-surface CPR sampling, and a frequency-block null that we added after the first results.

Southern Ocean · Continuous Plankton Recorder · SO-CPR · zooplankton · copepods · community temperature index · thermal tracking · space-for-time · sea surface temperature · OISST · climate change · taxonomic harmonisation · Neocalanus tonsus · null model