Plant traits, range position, and flowering phenology in the Southern Rockies

Link To Share This Poster: https://cdmcd.co/PJ9xB8
Live Discussion Link: https://cdmcd.co/zY9JbJ

Michelle Kummel, Ecology and Evolutionary Biology, Princeton University, Princeton, NJ and Ian K. Breckheimer, Rocky Mountain Biological Laboratory, Gothic, CO

Background/Question/Methods
Flowering phenology influences plant distributions, but how phenological traits interact with other traits to do so is an important and poorly studied question. The Rocky Mountain Biological Laboratory, located in the Southern Rockies, has a good system to address this question, with a large community of plants monitored at numerous sites across an elevation gradient spanning over 2000 m. Data collected in 2019 and 2020 recorded flowering presence-absence along a 50m transect at 12 sites, monitoring over 100 species overall. Since discrete site visits were up to 10 days apart and did not completely cover the whole growing season, developing a Bayesian statistical model for floral phenology at the sites allowed us to understand floral behavior continuously across the season and relate phenology to other plant traits while propagating uncertainty in the data collection process. Data on the focal traits—plant height for herbaceous plants and RMBL’s position within each plant’s observed range—were extracted from the Botanical Information and Ecology Network and field estimates. Our questions were: do shorter plants flower earlier? Do plants where RMBL is located in the north of their range flower later at RMBL than plants where RMBL is in the south of their range?
Results/Conclusions
We estimated the linear relationship between peak flower timing (day of year) and plant height for each of 2000 posterior samples from the Bayesian model and detected a positive relationship between height and flowering time (posterior probability > 0.99), with an average rate of change across the samples of 15 days / 1m increase in height. Splitting the plants into short, medium, and tall categories showed that the positive relationship was primarily driven by the short plants, which had earlier mean peak flowering times than tall or medium plants (posterior probability > 0.99), while tall and medium plants had similar peak flowering times. An equivalent linear analysis of flowering optimum vs. range position (z-score) of RMBL detected a positive relationship as well (average rate: 2.2 days later / increase of 1 in latitude z-score, posterior probability > 0.99), indicating that plants flowered later if RMBL was in a more northerly part of their overall range. We did not detect a relationship between either of these traits and the duration of plant flowering periods. Overall, our results suggest that plant height and range position affect flowering phenology, and thus may play a role in how phenology shapes community.