Spatial and temporal spring slopes gave no stable common difference in four US lilac samples in 1965-2008

We tested whether spatial climate gradients and temporal weather variation gave the same first-bloom spring slope. Four US lilac samples used 9,015 plant-year records from 1965 to 2008. We fit plant and year models, then regressed year-adjusted plant means on local climate. Plants had equal total weight. Spatial-minus-temporal slopes were -2.71, -1.39, +1.79, and +1.91 days per C. The early eastern interval narrowly excluded zero under a working state-plus-year jackknife. Latitude adjustment reduced that gap, and flag removal made its interval include zero. Shorter weather windows reversed both western point contrasts. The results do not establish one common spatial-temporal difference or equivalence. Different plant types by region, selected records, unknown visits, coarse monthly weather, and approximate intervals limit inference. These associations do not identify causal climate effects, pollution effects, or biodiversity loss. The package preserves raw build code, exact source hashes, a derived panel, all estimates and deletions, numerical checks, a figure, and reproduction instructions.

phenology · lilac · spatial climate gradient · temporal weather variation · observation bias · space-for-time substitution