A prior warmth by thermal niche position term stayed negative with route-year effects in pooled North American bird counts, 1970-2024

We tested a prior season warmth by thermal niche position term in North American recorded bird counts. We added route-year effects to control changes shared by species in the same survey. The model also had species-route-observer and species-year effects. We used the BBS 2025 release and GHCN-CAMS 3.01. Thermal niche position used route climates, with mean counts as weights. The pooled sample had 6,870,825 records from 368 mapped species in 1970-2024. The term was -0.322, with a 95% interval from -0.543 to -0.102. Units are 100 x log(count+1) per C per niche standard deviation. Four region and period estimates were negative. All four individual 95% intervals included zero. Full first represented observer year removal and equal route niche weights retained negative prior warmth terms. The future warmth placebo interval included zero. Selected records, count scaling, detection changes, niche weights from all years, coarse weather, and approximate cluster intervals limit inference. This is an annual weather association. It does not establish a causal climate effect. The package includes code, source hashes, a compact offline panel, numerical checks, results, and a figure.

thermal niche position · bird counts · Breeding Bird Survey · weather variation · observation bias · fixed effects · regional checks · reproducible panel