Year-to-year biomass responses of demersal fish to bottom temperature were about 3% of their spatial temperature slopes in 13 trawl survey units (1963–2023)
Question: species distribution models use space-for-time substitution. Does the spatial bottom-temperature slope of a demersal fish species predict its biomass response to warm and cold years in the same survey area? Data: FishGlob_data v2.1 (CC BY 4.0). We used 13 survey units with in-situ bottom temperature (eastern Bering Sea, Gulf of Alaska, Gulf of Mexico, US Northeast and Southeast shelves, Scotian Shelf, English Channel, northern Spain, Porcupine Bank), 83,005 hauls in 948 core 0.5-degree cells, 1963-2023, and 646 species-by-unit populations. Method: the plan was written before species results. For each population, a quasi-Poisson GLM gave the spatial slope s of log biomass density on cell climatological temperature. A second GLM, with cell fixed effects and a spline of year, gave the interannual slope b on the survey-wide annual temperature anomaly. The transfer slope lambda is the regression slope of b on s across species (1 = full transfer, 0 = none). A random-effects mean combined the units, and a year-permutation test kept the correlation among species. Results: mean lambda was 0.034 (95% CI -0.002 to 0.070); the permutation test gave one-sided p = 0.013. Lambda was below 1 in all 13 units. Of 340 populations with a clear spatial slope, 54% had an interannual slope with the same sign. Thus the direction transfers weakly, but the interannual response was about 3% of the size that full transfer predicts. In 11 variants (linear trend, split years, depth-adjusted slope, footprint trimming, one-year lag, quadratic spatial model, local anomalies), the mean lambda was 0.016 to 0.115. The community temperature index changed by 0.14 (CI 0.03 to 0.25) of its spatial gradient per degree of annual anomaly. Limitations: the spatial slopes include habitat gradients that correlate with temperature; spatial slopes come from the survey footprint only; anomalies are mostly within 1 degree C; biomass CPUA does not separate catchability, movement and population change; only 3 European units had temperature data; we did not test multi-decade change.
space-for-time substitution · species distribution models · bottom temperature · demersal fish · bottom trawl survey · FishGlob · community temperature index · thermal niche · climate change · interannual variability · transferability
References
10- usesdoi:10.5281/zenodo.19056965
FishGlob_data v2.1 is the source of all haul, catch and bottom-temperature data.
- usesdoi:10.1038/s41597-023-02866-w
Data descriptor of FishGlob_data; describes survey standardization and flags that we used.
- Against static-distribution replicates, demersal fish in 15 FishGlob trawl survey units experienced 4–10% less year-to-year bottom-temperature change (16–35% on the US Northeast shelf) sci:sha256:64e9211e9d66f96b3c901393f2eddaa2e16985901d0046a2ef094b470aff3b9e
Related FishGlob study: fish experienced less year-to-year temperature change than static distributions predict; this is one possible cause of the weak transfer we report.
- UK butterfly transect communities did not shift to warm-niche species at the rate of 1976–2024 site warming, while thermal niche predicted year-to-year responses sci:sha256:a37a9927e8c25069416d317953a1c1943c4017f40a065b89c7de4187728e5393
UK butterfly study in which thermal niche predicted year-to-year responses more strongly than in these fish data.
- Early-half shift rates of 1,404 fish populations do not predict their late-half rates of latitude or depth change in 25 bottom-trawl survey units (1967–2021) sci:sha256:440f3c8b47c8a5ab03f5738ceeee021cc15fd66ba491c22caf206c3966b8dac8
Related FishGlob study on the low persistence of fish distribution shift rates.
- doi:10.1073/pnas.1220228110
Tests of space-for-time substitution in biodiversity models.
- doi:10.1098/rspb.2008.0878
Community temperature index and climatic lag method.
- doi:10.1038/s41558-019-0631-5
Marine community warming responses and thermal affinities.
- doi:10.1111/gcb.12917
Community composition responses to local year-to-year climate variation.
- doi:10.1126/science.1239352
Marine taxa track local climate velocities in trawl survey data.
Cited by
10 publicationsPublic catalog entries on Sciscape.
- Adult rodents at Portal, Arizona, were not lighter in warmer years and did not lose mass in 1977–2026 in 11 species; their hindfoot length decrease is mainly measurement drift
Related test of year-to-year temperature responses in long-term survey data (demersal fish).
- 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
Demersal fish year-to-year responses were about 3% of spatial slopes; contrast with the larval result here.
- usesFish mean body mass did not decrease in warmer years in 13 trawl survey units, spatial slopes were weak, and steps at 1985 and 2009 gear changes made most of the US Northeast shelf decrease
We reuse its unit selection rules and temperature anomaly model, and compare our body-mass transfer slope with its biomass transfer slope.
- supportsThe fish community temperature index did not increase at 2,006 English river sites in 2000–2025 while air temperature rose 0.29 °C per decade; it followed summer temperature only from year to year
It found year-to-year fish responses far smaller than spatial temperature slopes; our year-to-year CTI sensitivity is about 8% of the spatial slope.
- The annual linear temperature link to fish catch richness was uncertain at 118 eastern Bering Sea stations, 2000–2025
Compares spatial and annual temperature slopes for fish biomass. Our separate test uses caught species richness.
- Within-species mean fish mass in 31 trawl survey segments (1980–2026) changed +0.3% per °C of yearly bottom-temperature anomaly (95% CI −0.7 to +1.3), not the −2.5 to −5% of the temperature–size rule
Same FishGlob surveys, year-to-year biomass responses to bottom temperature; body mass not examined.
- North Atlantic CPR plankton temperature indices followed year-to-year SST in 1958–2019, but three copepod taxa caused all of the long-term copepod change
Trawl fish year-to-year responses were about 3% of spatial slopes.
- Pre-solstice warming advanced and post-solstice warming delayed leaf colouring of four German tree species (1951–2025), but these effects did not predict later colouring in 2016–2025
Related contrast between short-term and long-term temperature responses in fish biomass.
- usesIn 13 trawl survey units (1963–2023), fish mass was lower in warmer cells (−3.6%/°C, mostly a depth effect) but not in warmer years (+1.8%/°C); spatial slopes did not predict temporal slopes
We reuse its prepared haul tables (core cells, cell temperature climatology, annual bottom-temperature anomalies). Its biomass result (year-to-year responses about 3% of spatial slopes) has the same pattern as our body-mass result.
- Catch history reduced mean prediction error more than bottom temperature in tested ridge models for 35 eastern Bering Sea fish taxa, 2007–2025
This prior test compares spatial and annual temperature responses in trawl biomass. Our test uses occurrence, catch history, and later-year prediction scores.