EASTERN HEMLOCK REGENERATION AND DEER BROWSING IN THE NORTHERN GREAT-LAKES REGION - A REEXAMINATION AND MODEL SIMULATION

White-tailed deer (Odocoileus virginianus) populations are currently extremely high in the upper Great Lakes region, and browsing may have severe negative effects on many forest species, including the eastern hemlock (Tsuga canadensis), a former forest dominant. We suggest that this problem must be examined in an expanded spatial and temporal context It may be incorrect to generalize from limited evidence on one scale of observation (stand level) to explain a regional, long-term failure of tree species regeneration We reviewed hemlock life-history characteristics, its long-term behavior on a regional scale, and evidence from recent studies in the context of long-term forest ecosystem processes. We then conducted 400-year computer simulations within a neutral model framework (see Caswell 1976), relative to browsing to test whether other factors might adequately explain the lack of hemlock regeneration in the northern lake states. The results suggest that browsing is not the critical step blocking hemlock forest re-establishment on a regional scale. We suggest that the interaction of climate, disturbance, hemlock life history, ecosystem processes, and historical land use produce positive feedbacks that prevent hemlock regeneration usually before deer browsing can occur. There is need to modify current single-species forest management and silviculture, and to manage forests at a landscape scale in a way that maintains regional biodiversity and sustainable forest ecosystems. These modifications should include reduced deer levels and maintenance of larger patches of mature ecosystems. But we caution against broadly invoking the negative effects of any single factor such as browsing as a basis for management changes without a thorough assessment of all factors.