A ten-day offset for missing intervals changed spring slopes by 0.05-0.22 days per C in four US lilac samples

We tested how assumptions about missing bloom dates change the spring-temperature slope of leaf-to-bloom intervals. We used 8,185 selected US lilac leaf reports in four region-period samples from 1965 to 2008. Of these, 619 had no eligible bloom record. A constant ten-day increase in assumed missing gaps raised the fitted slope by 0.051-0.219 days per C. An extra response of one day per C among missing gaps changed the slope by 0.040-0.102 days per C. All four complete-case uncertainty intervals included zero. The package gives an exact two-parameter sensitivity map and independent numerical checks. These calculations do not estimate actual missing dates or bias. They assume a bloom date exists in each missing year. Selected plants, absent visit histories, differing regional plant types and coarse monthly weather limit interpretation. We do not test pollution, daily extremes or causal climate-change effects.

phenology · lilac · missing data · observation bias · temperature · sensitivity analysis