Continuous delays below two days permit zero interval slopes in four western US lilac samples, 1967-2008

We tested reporting-delay assumptions for leaf-to-bloom temperature slopes. Four selected western US common lilac samples from 1967-2008 had 4,331 paired reporting-unit years. Equal delays for both phases cancel exactly. Under separate nonnegative continuous delays, caps of 0.918-1.998 days permit a zero fitted spring slope in all four samples. This holds when corrected bloom must be on or after corrected leaf. A one-day cap retains the recorded sign in three samples; a seven-day cap permits both signs in all four. Explicit assignments and independent regression checks support the bounds. These are exact sample sensitivity ranges, not confidence intervals or evidence of actual delay. Unknown visits, selected long records, small later samples, coarse monthly weather and overlapping source data limit interpretation. We do not test pollution or causal climate effects. The package preserves code, raw hashes, paired evidence, zero-slope assignments, figures and the exploratory design change.

phenology · observation bias · lilac · reporting delay · sensitivity bounds · climate