Morning temperature rise gave no clear taxon-count slope in four US bird survey samples in 1990-2024

We tested whether a larger morning temperature rise predicts recorded bird taxa at fixed mean survey temperature. We used 7,072 consecutive-year pairs from the North American Breeding Bird Survey. Each pair had the same route and observer. Separate models covered California and Oregon, and New York and Pennsylvania, in 1990-2006 and 2007-2024. Models controlled date, start time, duration, assistant status, observer experience, wind, sky and state-year changes. Estimated slopes ranged from -0.174 to +0.170 taxa per 5 C larger rise. All four-test intervals included zero and lay within -0.60 to +0.60 taxa. Four sensitivity checks also gave intervals that included zero. These data give no clear common direction for this conditional weather contrast. They do not prove that morning warming has no effect. Complete-record selection, weather error and dependence across routes limit the result. The study does not separate detection from ecological change or estimate effects of pollution, extreme heat or long-term climate change. The package includes code, exact model inputs, source hashes, design matrices, numerical checks and offline instructions.

Breeding Bird Survey · observation bias · temperature rise · biodiversity · weather · null result · first differences