Phenology, the timing of life-history events within a season, is an important topic that has gained new relevance due to climate change. Seasonal patterns of temperature and precipitation are shifting; when species respond to climate shifts at different rates this can lead to new patterns of temporal overlap among species, causing changes in the strength of species interactions. A common metric used to quantify changes in phenology is the record of when an event is first observed within a population, such as the first flowering, spawning, or arrival of a migrant. These ‘first events’ have been recorded for many species, but in cases where complete distributions of phenology are available it is common for trends relating first event dates to climate (or changes over time) differ from trends for other metrics such as peak activity. Instead of basing analyses of phenology on a single date, multivariate descriptions of phenological processes may give us a better understanding of how communities will change in response to shifts in climate.
Results/Conclusions
We used data from a long-term study of flower phenology in the Rocky Mountains to investigate how multivariate phenological responsiveness to climate variables is related to plant community structure. Multiple aspects of phenological responses are correlated with each other, but for many plant species there were also covariance patterns indicating that different aspects of phenological distributions have different sensitivities. Metrics based on individuals (i.e. first open flower in a population) were not good indicators of population-level phenological process, and simulations and data from a few species suggest this is because individuals within a population have different climate reaction norms, and because of spatial heterogeneity in microclimates. Plant communities in different habitats have, on average, different responsiveness to climate, suggesting that plant communities across a landscape will initial growth and flowering in novel combinations as climate change proceeds. Using 2x2 m plots as replicates, plant communities with greater responsiveness in flower phenology distributions also showed greater year-to-year changes in species percent cover, suggesting plant phenology traits may be useful for analyses of species performance and community composition trajectories.