The June 2021 heat dome gave no clear bird-count decrease on 701 western survey routes in 2022-2025; acute heat in 1981-2025 gave about 1% lower next-year richness and no count change

Question: did the late-June 2021 Pacific Northwest heat dome decrease breeding-bird numbers, and does acute breeding-season heat in other years and regions give a decrease? Data: North American Breeding Bird Survey (BBS) 2026 release (1980-2025) and Daymet v4 daily maximum temperature (Tmax) at 5,175 route start points. Exposure: the largest 3-day mean Tmax anomaly (1991-2020 daily climatology) in 24 Jun-4 Jul 2021 (dome) or in 1 Jun-31 Jul of each year (general). Method: plan written before the fits; log(total birds + 1) and log(richness + 1) with route x observer and Bird Conservation Region (BCR) x year fixed effects, survey date and start-temperature controls, errors clustered by 1-degree block. Dome result (701 routes at 40-60 N, 110-130 W; dome anomaly 2-23 C): per +5 C, total birds changed -0.4% (95% CI -3.7 to +2.9) in 2022-2023 relative to 2019, and richness changed -0.8% (-2.4 to +0.9). After control of a pre-existing exposure x year trend, the post-2021 break was +3.2% (-0.8 to +7.2). Results were similar without British Columbia, in the US only, and without surveys made during the dome. 230 species gave a pooled -0.4% (-2.2 to +1.4) per +5 C; 11 species had intervals below zero and 15 above zero, near the chance number. Routes with dome Tmax of 40 C or more (100 with surveys before and after) showed no clear change. General result (106,681 route-years, 5,159 routes, 2021 season held out): the previous-season peak anomaly had no link to total birds (+0.005% per C, -0.14 to +0.15), but richness decreased by 0.14% per C (-0.21 to -0.07) and by 1.1% after peaks of 10 C or more (-1.8 to -0.5). The decrease occurred in the East and the West. A future-heat placebo gave -0.05% per C (-0.12 to +0.01) after route detrending, and -0.08% (significant) without detrending, so part of the richness link can be confounding. The general slopes predicted a dome effect of -0.03% (totals) and -0.5% (richness) per +5 C; the dome estimates agree but cannot separate these from zero. Limitations: BBS counts are not corrected for detection; exposure is at the route start point only; high-exposure routes declined before 2015; 2020 has no surveys; 2021 participation fell on more exposed routes; drought and 2021 wildfires coincide with the dome; only four post-dome years.

heat dome · extreme heat · heatwave · Breeding Bird Survey · bird abundance · species richness · Pacific Northwest · Daymet · event study · placebo test · climate extremes · observation bias