No abstract

Climate refugia, where local populations of species can persist through periods of unfavorable regional climate, play a key role in the maintenance of regional biodiversity during times of environmental change. However, the ability of refugia to buffer biodiversity change may be mediated by the landscape context of refugial habitats. Here, we examined how plant communities restricted to refugial sky islands of alpine tundra in the Colorado Rockies are changing in response to rapid climate change in the region (increased temperature, declining snowpack, and earlier snow melt-out) and if these biodiversity changes are mediated by the area or geographic isolation of the sky island. We resampled plant communities in 153 plots at seven sky islands distributed across the Colorado Rockies at two time points separated by 12 years (2007/2008–2019/2020) and found changes in taxonomic, phylogenetic, and functional diversity over time. Specifically, we found an increase in species richness, a trend toward increased phylogenetic diversity, a shift toward leaf traits associated with the stress-tolerant end of leaf economics spectrum (e.g., lower specific leaf area, higher leaf dry matter content), and a decrease in the functional dispersion of specific leaf area. Importantly, these changes were partially mediated by refugial area but not by geographic isolation, suggesting that dispersal from nearby areas of tundra does not play a strong role in mediating these changes, while site characteristics associated with a larger area (e.g., environmental heterogeneity, larger community size) may be relatively more important. Taken together, these results suggest that considering the landscape context (area and geographic isolation) of refugia may be critical for prioritizing the conservation of specific refugial sites that provide the most conservation value.

Many studies of community assembly focus on a single ontogenetic stage (typically adults) when trying to infer assembly processes from patterns of biodiversity. This focus ignores the finding that assembly mechanisms may strongly differ between life‐stages, and the role of ontogenetic dependency: the mechanisms by which one life stage directly affects the composition of another life stage. Within a 4‐ha forest dynamics plot in California USA, we explored how the relative importance of multiple assembly processes shifts across life stages and assessed ontogenetic dependency of seedlings on adults in woody plant communities. To assess variation in assembly processes across life stages, we examined how β‐diversity of adult and seedling communities were each influenced by space and 13 environmental variables (soils, topography) using distance‐based redundancy analysis and variation partitioning. We then assessed the ontogenetic dependency of seedlings on adults by including adult composition as a predictor in the seedling community variation partitioning. We found differences between adult and seedling composition. For the adults, we found 18 species including pines, oaks and manzanitas characteristic of this mid‐elevation forest. For seedlings, we found 11 species, and that oaks made up 75% of all seedlings while only making up 45% of all adults. Adult β‐diversity was primarily explained by space (44.0%) with environment only explaining 18.6% and 37.4% unexplained. In contrast, most of the explained variation in seedling β‐diversity was due to ontogenetic dependency alone (13.6% explained by adult composition) with 1.6% explained by space and the environment jointly, and 62.8% unexplained. Synthesis: Here, we describe a conceptual framework for integrating ontogeny more explicitly into community assembly research and demonstrate how different assembly processes structured adult and seedling β‐diversity in a temperate dry forest. While adult β‐diversity was largely driven by spatial processes, seedling β‐diversity was largely unexplained, with ontogenetic dependency comprising most of the explained variation. These patterns suggest that future assembly research should consider how assembly processes and their underlying mechanisms may shift with ontogeny, and that interactions between ontogenetic stages (ontogenetic dependency) are critical to consider when assessing variation in assembly processes.