March-April warmth gave no clear winter interaction for cherry bloom intervals in four Japan samples in 1953-2025

We tested whether winter warmth changes the spring temperature slope of the first-to-full cherry bloom interval. We used 4,455 paired station-years at 67 Japan stations in 1953-2025. Four separate East and West, early and late models included station effects, year effects, and curved seasonal terms. Winter by March-April interaction estimates ranged from -0.317 to -0.158 days per C squared. All four adjusted intervals included zero. Five sensitivity checks also gave uncertain interactions. This extends a published March-only test with a different spring window, regional split, and record selection. It is not an independent biological replication. Wide intervals, unknown visits and tree changes, selected complete dates, and coarse weather limit inference. Monthly spring windows can include weather after bloom. This result does not establish no interaction or a causal climate effect. The package includes code, source hashes, the matched panel, all estimates, numerical checks, and a figure.

phenology · cherry · winter warmth · observation bias · seasonal interaction · Japan