The current interest among resource managers in ecosystem management necessitates a better understanding of the response of plant species diversity to forest management practices. This study attempted to assess the effects of one forest management practice–clear‐cutting–on plant biodiversity in a mid‐Appalachian hardwood forest by comparing species composition and diversity between two young (≈ 20 yr following clear‐cutting) and two mature (> 70 yr following selective cutting) watersheds. Sampling was confined to the herbaceous layer (vascular plants ≤ 1‐m in height) and woody overstory (stems ≥ 2.5 cm diameter at 1.5‐m height). The important tree species occurred on all watersheds, but the relative importance of these species varied greatly with stand age. Less shade‐tolerant species, such as black cherry (Prunus serotina) and tulip poplar (Liriodendron tulipifera), are replaced by more‐tolerant species, such as northern red oak (Quercus rubra) and sugar maple (Acer saccharum), as the stand matures. Analysis of tree species richness per plot suggests that the competitive thinning process decreases species evenness as the stand develops. Important herb‐layer species included stinging nettle (Laportea canadensis), violets (Viola spp.), seedling of striped maple (A. pensylvanicum), and several fern species. In sharp contrast with the trees, these species appeared to vary little with stand age. Species diversity (H') did not vary with stand age for either the overstory or the herbaceous layer. Detrended correspondence analysis showed a significant correlation between species composition of the two strata for the mature watersheds, but not the young, clear‐cut watersheds. Thus, we suggest a temporal shift in processes influencing species composition following disturbance from allogenic factors (e.g., soil characteristics) to autogenic factors (e.g., stand characteristics), which lead to a linkage between forest strata later in succession. The degree to which forest management alters species composition in these central Appalachian forest ecosystems may be tied to the degree of alteration of the link between strata.

No abstract

The herbaceous layer (vascular plants _80-yr-old control), and (~ 20-yr-old control). The herb layer was sampled intensively in I0 Io m 2 subplots within each sample plot, including determination of species composition, cover, and random biomass harvests. Harvested plant material was separated by species and analyzed for macronutrients, micronutrients, and AI. Soft was sampled from harvest subplots and analyzed for texture, pH, organic matter, and macro-and micronutrients. Few differences among watersheds for virtually all measured soil variables indicated minimal response of soil fertility to the acidification treatment. The herbaceous layer was also quite similar among watersheds with respect to cover-biomass and species diversity; WST, however, had 70% higher herb layer cover that both Ws3 and WS4, a result of the predominance of a few high-cover fern species and attributable to the north-facing aspect of WS7 vs. south-facing aspects of WS3/WS4. There was a high degree of species similarity among watersheds, suggesting no shift in species composition in response to acidification. There was also minimal response of element concentrations to acidification, although Fe and AI exhibited evidence of increased uptake in WS3. We conclude that, contrary to our expectations, there has been little substantive response of the soil and herb layer to acidification, but hypothesize that herbaceous layer species may experience toxicity problems with increased mobility of AI and micronutrients in the future. R ECENT STUDIES have supported the conclusion that current levels of acid deposition have contributed to a decline in productivity of forest ecosystems of the eastern United States (Johnson and Taylor, 1989; Adams and Eagar, 1992; Eagar and Adams, 1992). There has been lack of general agreement, however, among these studies as to the extent of forest damage directly attributable to increased atmospheric acidity. Such lack of agreement results from several factors, including the problems associated with

No abstract