| Climate change and anthropogenic alterations of the landscape like non-native species introduction, pollution, and habitat destruction substantially alter ecosystem functioning through shifts in key phenological events and changes in community composition as certain species and groups of species decrease in abundance and distribution. Alterations in the timing of flowering and leaf phenology have been observed worldwide but the dearth of long-term observational records limits assessment of phenological responsiveness in many regions and for many species. Similarly, there is insufficient observational data regarding the occurrences of the majority of described species to assess changes in population abundance or distribution over time. Thus, the conservation needs of these species cannot be assessed. To address these knowledge and data gaps, I used herbarium specimens as a non-traditional data source to develop a novel method for quantifying flowering shifts with temperature increase, to evaluate flowering responsiveness to temperature in 207 plant species in Ohio, and to determine changes in abundance and distribution of those species over the past 115 year. Further, I used one of the most comprehensive historic leaf phenology datasets in the world paired with modern observations to determine changes in total growing season length and to model spring and autumn leaf phenology using temperature and precipitation drivers.;Across the 116,000 km2 study area in north-central North America, the date of maximum flowering has advanced 2.4 d/°C, on average. However, species-specific variation in responses was high and ranged from -13.5 to +7.3 d/°C. Further, plant functional groups differed significantly in their phenological sensitivity to temperature with introduced, spring-flowering, and wind pollinated species showing greater flowering advancement with warming than native, summer-flowering, and biotically-pollinated species, respectively. Further, I found widespread evidence of threat to Ohio's biodiversity as 27% and 66% of the 207 species evaluated had decreased in abundance and distribution, respectively. Further, native species were seven-fold more likely to experience abundance decline and showed two-fold greater distribution contraction than introduced species. Traits that predispose species to decline such as reliance on a symbiont for growth or reproduction were also more common in native species and may partially account for performance differences among native and non-native species. Additionally, highly phenologically responsive introduced species seem to be favored with warming as these species significantly increased their distributions relative to less responsive species. In contrast, highly responsive native species received no performance benefits relative to those species that weakly shifted flowering with warming. Finally, a century of climate warming has resulted in growing season extension because of foliage coloration delays, and also, more variably, as a result of earlier budburst. Both contemporary observations and models of leaf coloration and fall suggest a compression of the duration of full canopy coloration as leaf fall is advanced in years with later leaf coloration. This relationship suggests that the timing of leaf abscission may be driven by endogenous factors associated with the timing of leaf coloration rather than direct effects of climatological variables. |
