Oak savannas were once common in the Midwestern USA but are now one of the most endangered ecosystems in the world due to conversion to agricultural uses and absecnce of fire as a natural disturbance. Prescribed fire and thinning canopy and understory trees and shrubs are common restoration practices; however, assessment of ground layer response to these treatments has been local rather than regional. We examined the effectiveness of these management approaches for restoring oak savannas in the central Midwest from northwestern Ohio and Michigan to northwestern Indiana, northeastern Illinois and southeastern Wisconsin. We hypothesized that burning and thinning, mediated through changes to habitat structure (e.g., litter, canopy cover), as well as soil properties (e.g., chemistry, soil moisture) would increase ground layer richness and diversity. We sampled over 90 plots spanning 30 sites with vegetation characterized among unmanaged, burned alone, and thinned alone. Vegetation and soils were sampled using tenth hectare modified Whittaker plots with nested 1x1m abundance plots. Soil chemistry was represented by a standard panel of soil test results coupled with water holding calculations.
Results/Conclusions
Sites across the region ranged from sandy soils to deep soil silt loams. A combination of burning and thinning significantly reduced litter cover and increased light availability compared to plots that were burned or thinned alone. Ground layer diversity increases with lower litter cover and higher light availability. Together, these results suggest that management may positively affect understory plant diversity through influences on light availability and O horizon development. Cover of sassafras (Sassafras albidum), bracken fern (Pteridium aquilinum), and black cherry (Prunus serotina) showed negative relationships with litter cover, whereas Pennsylvania sedge (Carex pennsylvanica) showed a weak positive relationship with litter cover. Black oak (Quercus velutina) and red maple (Acer rubrum) cover were respectively negatively and positively correlated with total basal area. Black cherry and Virginia creeper showed respectively a positive and negative relationship with increasing soil texture. These results suggest that some species abundances respond to management, by being sensitive to liter cover and basal area that is mediated by burning and thinning. Other species abundances are related to factors that are constant and unmanaged at sites, such as soil texture. Once sampling is complete this year, we expect to better elucidate the direct and indirect causal factors that influence Midwestern oak savanna ground layer diversity in a context of land management approaches.