Spring weather change gave no clear repeated link to US shrub timing change in 1965-2008

We tested whether larger temperature changes between consecutive springs were linked to larger changes in the time from first leaf to first bloom. A published US lilac and honeysuckle panel gave 8,960 selected year pairs in four region and period samples. Plant and ending-year effects, signed temperature change, and mean temperature were included. All four slopes were positive, from 0.146 to 1.318 days per C of absolute February-March temperature change. One state-cluster interval excluded zero after correction for four tests: the early East slope was 0.766, with an interval of 0.109 to 1.423. Its two-way state and year interval included zero. March-only and large-change checks also weakened this result. The evidence does not establish a link that repeats clearly across regions and periods. Selected complete records, missing visit dates, coarse weather, and few clusters limit inference. The test does not identify a change in biological variance or a causal climate effect. The package preserves the exact source panel, code, provenance, estimates, a figure, and numerical checks.

phenology · weather change · leaf-to-bloom interval · observation bias · replication · lilac · honeysuckle