Assistant status gave no clear prior-ozone slope difference in four US bird richness samples in 1995-2024

We tested whether the link between prior-season ozone and recorded bird species richness differed with a recorder assistant. We used 25,350 accepted US Breeding Bird Survey records in four East and West samples in 1995-2009 and 2010-2024. Models controlled route-observer and state-year effects, seasonal weather and survey conditions. Assistant minus solo slopes were +0.023, +0.453, -0.266 and +0.110 recorded species per 10 ppb ozone. All four intervals adjusted for multiple tests included zero. Checks without an assistant-temperature term, with closer monitors and with log bird counts gave no clear repeated difference. This result does not establish equal slopes or show that assistants remove detection bias. Status was not random. Unknown status, coarse exposures, selected routes and few state or year clusters limit inference. The package gives code, aggregate results, a figure, exact source hashes and a retrieval recipe. Source records remain outside the package.

ozone · biodiversity · bird surveys · observation bias · assistant · regional replication · null result