Summer warmth gave no clear nitrogen contrast in next-year butterfly site archive absence in Britain in 1986-2012

We tested whether summer warmth had a different link to next-year site archive absence at sites with more nitrogen deposition. We used 15,204 active site-years from 1,757 standard butterfly transects in Britain in 1986-2012. Four separate region and period models controlled persistent site differences, country-year conditions, spring warmth and a nitrogen-specific time trend. The warmth and nitrogen slopes ranged from -0.52 to +1.26 percentage points per C per 5 kg N/ha/year. All four adjusted intervals included zero. Four sensitivity checks gave the same broad result. This does not show that archive absence is random or that abundance estimates lack bias. No source row can reflect survey loss, source processing or no recorded species. Static nitrogen, coarse weather, few year clusters and unmeasured local change limit inference. The package gives the full site-year presence table, model inputs, source hashes, code, checks and a figure.

butterflies · observation bias · archive retention · nitrogen deposition · summer temperature · regional replication · null result