Prior October-January precipitation gave no clear change in spring warmth slopes of US lilac bloom in 1965-2008

We tested whether prior October-January precipitation changed the link between February-March warmth and first-bloom dates. Four separate plant and year models used 8,947 US lilac plant-years in two regions and two periods from 1965 to 2008. They controlled for winter temperature and curved weather responses. Interaction estimates were -0.036, -0.091, +0.461, and -0.629 days per C per precipitation standard deviation. All four adjusted intervals included zero. Planned sensitivity checks and later state omission checks did not establish a common interaction. The later western sample was small, and Montana supplied 42% of early western interaction information. This null result does not show that water supply has no effect. Plant type and region are confounded. Missing visits, selected records, irrigation, coarse weather, and approximate cluster intervals limit inference. We do not estimate soil moisture, damaging extremes, air pollution, species loss, or causal climate-change effects. The package includes a raw-data build, a matched weather panel, source hashes, all results, numerical checks, and figures.

phenology · lilac · precipitation · temperature interaction · observation bias · regional replication · null result · reproducible dataset