Warming in the year prior to flowering alters flowering phenology in boreal understory plant species

Christa Mulder, Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, AK and Pamela K. Diggle, Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT

Background/Question/Methods
Warmer temperatures in spring time are correlated with earlier flowering times in some, but not all, plant species; in some plant communities mean date of flowering has not advanced during the past few decades despite rapid warming and advanced flowering in warm years. Flowering is the last stage in a developmental process that for many boreal species begins approximately 1 year earlier. We hypothesized that temperatures in the flower initiation year (the year prior to flowering) affect flowering phenology by altering rates of development of preformed buds, and that this can help explain long-term trends in flowering phenology (or lack thereof). We tested this hypothesis in a field experiment with a full-factorial design in which we raised temperatures (using open-topped chambers) in the initiation year, the flowering year, both, or neither for two cohorts (initiated in 2017 and 2018). We established 114 warmed and 114 control plots across 19 sites in Interior Alaska and compared flowering phenology of five species in warmed and control plots.
Results/Conclusions
Responses to warming differed by species. For Vaccinium vitis-idaea (lowbush cranberry or lingonberry), plants in warmed plots in the initiation year (IY) delayed flowering, while warming in the flowering year (FY) advanced flowering. In Shepherdia canadensis (buffaloberry) warmer temperatures in IY delayed flowering while warming in FY had no effect. In Vaccinium uliginosum (bog blueberry) warmer temperatures in both IY and FY advanced flowering; the pattern was similar for Viburnum edule (highbush cranberry), though not statistically significant. For Rosa acicularis (prickly rose) warming in IY had no effect while warming in FY advanced phenology. There was no evidence for an interaction between warming in the two years for any species. These results suggest that for wild plant species with bud preformation predictions for shifts in phenology should include temperatures during two growing seasons. For example, they help explain why V. vitis-idaea has not advanced in phenology over time despite rapid warming: effects of warming in IY counter those in FY. The differences between species in the strength and direction of their responses, especially in the initiation year, suggests that patterns of co-flowering within the community are changing and will continue to do so as warming continues.