Oak savanna management strategies and their differential effects on vegetative structure, understory light, and flowering forbs

High quality oak savanna communities were once abundant in the North American Midwest, but have become exceedingly rare. Where remnant savannas remain, fire suppression and resulting woody encroachment have dramatically altered vegetative structure, resulting in reduced understory light levels and precipitating declines in herbaceous understory diversity. Restoration of fire suppressed oak savannas generally involves the reintroduction of fire, but questions remain regarding the necessity and impact of mechanical woody vegetation reduction in addition to fire. We report here on initial short-term results of a long-term experiment in a remnant fire-suppressed oak savanna in Southern Michigan to compare a gradient of oak savanna management intensities including; (1) unmanaged reference plots, (2) burning alone (low management intensity), and (3) progressive mechanical thinning combined with burning (high management intensity). We measured several metrics of restoration success: understory, shrub, and canopy cover, understory light levels, understory floral resources, and flowering forb diversity, in the first two growing seasons after initiation of restoration treatments. We found that increasing management intensity largely corresponded with increased understory light availability, abundance of floral resources (i.e. forb bloom abundance and diversity), and with decreased shrub and canopy cover. Low management intensity did not increase light availability, or decrease vegetative or canopy cover relative to unmanaged references, while high management intensity achieved both management goals. Both burning alone and thinning + burning generally increased diversity of flowering forbs, where the effect was greatest with high intensity management. These increases were most pronounced in the second growing season after burning, when we saw sharp increases in richness and abundance of flowering forbs. In restored plots, the flowering forb community consisted of pre-existing shade tolerant species, native and exotic ruderals, as well as savanna indicator species. In sum, low-intensity management can achieve some restoration objectives over the short-term; however, we show a clear initial advantage of coupling thinning with burning, relative to burning alone. We suggest that these differing intensities of oak savanna restoration may be appropriate under different temporal, financial, and ecological scenarios. Our work highlights the potential for restoration of understory forb communities by low or high intensity approaches, where relict populations and/or viable seedbanks exist.