Timing the bloom season: a novel approach to evaluating reproductive phenology across distinct regional flora

Context: Just as the timing of the vegetative growing season affects a host of ecological processes, the seasonality of floral availability impacts ecological processes from nectar availability and allergen production to competition for pollinator attention. However, no existing methodology is capable of evaluating multi-species bloom phenology in a standardized fashion across multiple ecosystems or compositionally distinct local flora. Thus, the manner in which the onset of the bloom season (during which the majority of species flower) differs along climate gradients and among distinct local flora remains largely unknown. Objectives: This study evaluates differences in the timing of the bloom season throughout the western United States, and the relationship of the bloom season to the vegetative growing season and to local climate conditions. Methods: This study estimated the season during which all but the earliest and latest 5 % of local species flower (the bloom season) using digital herbarium records. Bloom season timing was compared to land surface phenology, SI-x phenoclimate metrics, and PRISM climate normals. Results: Local differences in mean temperature of the coldest month explained 76 % of observed variation in bloom season onset. Variation in land surface phenology explained 50 % of observed variation, while SI-x Bloom estimates explained 64 % of observed variation in bloom season onset. Conclusions: These results confirm that bloom season phenology is distinct from the vegetative growing season, and that local temperature is a good predictor of bloom season onset. This work represents a new modality for studying multi-taxa flowering phenology at landscape and regional scales.