The effects of three regeneration harvest methods on plant diversity and soil characteristics in the southern Appalachians

We evaluated the effects of three regeneration harvest methods on plant diversity and soil resource availability in mixedhardwood ecosystems. The study area is in the Wine Spring Creek watershed on the Nantahala National Forest of the Southern Appalachian Mountains in western North Carolina. The regeneration treatments were: an irregular, two-aged shelterwood cut (2A), with 5.0 m/ha residual basal area; a shelterwood cut (SW), with 9.0 m/ha residual basal area; a group selection cut (GS), with 0.10–0.20 ha openings and 25% overstory removal on area basis at first entry; fourth, the control, consisted of two uncut sites (UC). Each harvest treatment was replicated three times across the landscape in similar plant community types. Within each treatment area, permanent plots were marked and inventoried for overstory, midstory, and herbaceous layer plants. In each permanent plot, we collected soil samples in winter (December–March) to reduce temporal variation due to vegetation phenological stage and rainfall events. We analyzed soil samples for extractable calcium (Ca), magnesium (Mg), potassium (K), cation exchange capacity (CEC), pH, bulk density, A-horizon depth, total carbon (C), and nitrogen (N). Species diversity of overstory, understory, and herbaceous layer species was evaluated using species richness (S), Shannon–Wiener’s index of diversity (H0), and Pielou’s evenness index (E). We used direct gradient analysis (non-metric multidimensional scaling, NMS) to explore the changes in vegetation–site relationships among herbaceous layer abundance, and soil characteristics and overstory basal area between pre-harvest (1994) and post-harvest (2000). Twelve minor overstory species were cut from the 2A treatments and nine species were cut from the SW treatments. Thus, it is not surprising that S and H0 were reduced in the overstory on the heavily cut sites. However, most of these species sprouted from cut stumps and were substantially more abundant in the midstory layer after harvest than before. For the midstory, we found higher S and H0 on the harvested treatments than the control; however, H0 did not differ significantly among the harvest treatments. We measured an increase in herbaceous layer H0 on the more heavily cut treatments (2A and SW) after harvest. We found an increase in average distance in the NMS ordination among sites in 2000 compared to 1994, which suggests greater herbaceous species diversity after harvest. However, we did not see a clear separation among harvest treatments in the NMS ordination. # 2005 Elsevier B.V. All rights reserved.