Survey warmth gave no clear repeated link to nearby ozone-site differences in four US bird-route samples in 2005-2019

We tested whether survey-start temperature was linked to the absolute difference between nearby ozone sites. We used 1,599 selected bird-route windows in California, Oregon, New York and Pennsylvania in 2005-2019. Four models controlled site pair, year, survey date and start time. Mean ozone differences were 2.61-4.50 ppb. Slopes ranged from -0.13 to +0.25 ppb per 5 C. All four adjusted intervals included zero. Complete-date and largest-network checks gave no clear repeated link. This does not establish equal exposure error across temperatures. Survey-start temperature is not prior-month weather. Selected routes, few large site networks and possible shared weather limit inference. This is an exposure-method test. It does not estimate bird harm or climate-change effects. The package gives daily evidence, source hashes, code and independent numerical checks.

ozone · temperature · observation bias · exposure measurement · Breeding Bird Survey · site networks