Browsing by deer and a recalcitrant layer of pawpaw (Asimina triloba) cause increased sapling mortality and decreased sapling growth in an ancient urban forest1

When evaluating ecosystem services and biodiversity of secondary forests, ancient forest remnants provide valuable benchmarks for management goals. In the eastern United States, a principal challenge to forest health, both for ancient and secondary forests, involves browsing by white-tailed deer (Odocoileus virginianus). In addition to eliminating browse-sensitive tree species, browsing can create a recalcitrant layer, which limits light availability, thereby reducing sapling growth rates and extending sapling exposure to deer. To evaluate the influences of browsing by deer and shading by pawpaw (Asimina triloba (L.) Dunal), caged versus uncaged pairs of saplings were examined in plots where pawpaw trees remained or were removed. Uncaged saplings had significantly greater mortality than caged saplings both in removal plots and control plots ( = 14.0; P = 0.0002; = 11.5; P = 0.0007, respectively). In contrast, removal of pawpaw trees did not significantly affect mortality of caged saplings ( = 0.5; P = 0.5). However, sapling growth was significantly greater in removal plots than in control plots (F1,56 = 6.5; P = 0.013), with sapling growth in control plots less than 40% of that in removal plots. By creating a recalcitrant layer of pawpaw, browsing by deer indirectly exacerbated the negative effect that deer had on canopy regeneration. Although ancient forest remnants provide aspirational goals for managers of secondary forests, the urban ancient forest in this study, Wesselman Woods Nature Preserve, must itself be managed to prevent the negative consequences of browsing that will otherwise diminish its quality. The concern for Wesselman Woods Nature Preserve is discussed more broadly in the framework of ancient forests.