Winter rain gave no clear larva-pupa slope contrast in four butterfly samples in England and Wales in 1980-2024

We tested whether winter rainfall had a different link to annual butterfly site indices for species that overwinter as larvae and species that overwinter as pupae. We used 323,899 site-species-year records from England and Wales in 1980-2001 and 2002-2024. Models controlled site-species differences, shared site-year changes, seasonal weather and a stage trend. Larva minus pupa slopes ranged from -0.066 to +0.031 log(1 + index) per 100 mm of winter rain. All four adjusted intervals included zero. All twelve sensitivity intervals also included zero. The evidence gives no clear repeated stage contrast. It does not show equal responses or no rainfall effect. Selected valid indices, broad country weather, trait error, few seasons and approximate uncertainty limit inference. We did not test pollution or causal survival effects. The package includes inputs, code, source hashes and numerical checks.

butterflies · winter rainfall · overwintering stage · observation bias · regional comparison · null result