Tracking spring wildflower phenological sensitivity to microclimate in North American temperate deciduous forests

In the temperate deciduous forests of the northern hemisphere, spring-active wildflowers are vulnerable to climate change based on their strategy for seasonal light acquisition.  This diverse group of plants is characterized by their temporal niche, emerging after snowmelt but before canopy leaf-out to assimilate a significant portion of their yearly carbon budget. This shade-avoidance strategy is particularly important for the spring ephemerals, which rely entirely on the spring light window for their yearly growth. As canopy trees leaf out earlier with warmer spring temperatures, is wildflower phenology keeping pace? Recent studies conducted at a broad spatial scale have demonstrated that the answer to this question varies across continents and regions. The purpose of this study is to track spring wildflower phenological sensitivity to climate at a local scale. In the spring of 2025, we established 32 plots across two mountains in northeastern North America, determining plot location by stratifying across gradients of topography and elevation. Specifically, we 1) identified warm and cool aspect slopes, 2) separated each slope into increments spanning 60 m elevation gain, and 3) identified convex and concave landforms within each slope aspect and elevation band combination. We resampled all plots nine times between early spring and fall, following National Phenology Network (NPN) protocols to generate more than 3000 phenological observations of 60 wildflower taxa. For a subset of these species, we measured physiological parameters using a LI-COR LI-600. At each plot, we stationed a TOMST TMS-4 datalogger to generate measurements of air temperature, soil temperature, and soil moisture at sub-daily temporal scales. We used a Bayesian framework to construct generalized additive models of microclimatic variation over time and generalized linear models of the timing of key phenological events. Our results demonstrate the effect of slope aspect, elevation, and landform on microclimatic variation. Air temperature was lower at higher elevations and on the cool aspect slopes, and soil moisture was higher in the concave landforms. Wildflower phenological traits displayed variation along these microclimatic gradients, with the timing of peak vegetative abundance and peak flowering slightly later at higher elevations and on the cool aspect slopes. Proxies for photosynthetic rate were highest for the spring ephemerals, indicating an aggressive growth strategy to compensate for their short growing season. These initial findings encourage further study to assess the potential for localized habitats to function as microrefugia for spring wildflower biodiversity – microclimates where the extremes of macroclimatic warming are buffered, providing vulnerable taxa with more time for adaptation and migration.