Understory plant communities in boreal forests influence nutrient cycling, provide fuel for ground fires, stabilize the permafrost, and account for the majority of plant diversity. Following forest fires, plant community development can proceed along different trajectories, influenced by residual plants, competition, and local edaphic factors. Cajander larch (Larix cajanderi) forests of northeastern Siberia are experiencing increased wildfire activity in conjunction with climate warming. Mounting evidence suggests that increased fire frequency and severity can limit larch regeneration and lead to successional trajectories that favor low density larch forests dominated by shrubs and grasses. Yet, little is known about how and why understory plant communities develop along different trajectories following fire in this region. We surveyed understory plant communities along a larch tree density gradient within a 75-year-old fire scar near Cherskiy, Russia. Our primary goal was to determine how plant community diversity and taxonomic composition changed with tree density. We used linear regression to estimate how species richness and evenness change in response to forest density. We used non-metric multidimensional scaling (NMDS) ordination to visualize taxonomic shifts across the density gradient, and used multilevel models to investigate individual species’ response to forest density, leaf litter cover, and active layer depth.
Results/Conclusions
We found that plant communities differed considerably along the tree density gradient in this 75-year-old fire scar. Species evenness declined as tree density increased for both the exponent of Shannon’s Entropy (R2 = 0.14, p = 0.04) and the inverse of the Gini-Simpson index (R2 = 0.24, p = 0.01). Both our NMDS ordination and multilevel models demonstrated that plant communities in low-density areas were distinct compared with plant communities in high-density areas. Grasses and shrubs, like dwarf birch (Betula nana) and cranberry (Vaccinium vitis-idaea), dominated in areas with low tree density, while mosses (e.g. Sanionia spp. and Dicranum spp.) and liverworts (e.g. Ptilidum ciliare) were more common in areas with high tree density. Differences in species composition were more closely related to canopy cover than to leaf litter depth. We also found that the low-density forest with high shrub cover had deeper active layers compared with the high-density forest, where shrub cover was low. These results indicate that shifts in forest successional patterns lead to changes in understory plant communities. These changes can then have important implications for permafrost stability, primary productivity, and nutrient cycling in boreal forests.