Winter warmth gave no clear nitrogen contrast in four British butterfly samples in 1986-2012

We tested whether winter-temperature slopes of butterfly site indices differed with nitrogen deposition and host-plant nitrogen preference. Four regional and time-period models used 356,011 records in Britain in 1986-2012. Winter was the prior December plus January and February. All four intervals included zero after correction for four tests. Two record-selection checks gave the same conclusion. A positive unadjusted result in the later southern sample did not pass that correction. The test uses fixed effects for local species, site-year conditions and regional species-year patterns, with spring and summer controls. Static deposition, coarse weather, selected sites and species-specific detection limit inference. This result does not establish no interaction or a causal pollution effect. The package includes inputs, source hashes, code, a figure and numerical checks.

butterflies · winter temperature · nitrogen deposition · observation bias · regional replication · null result