Volume 83 - Issue 1 (May 2018)

ABSTRACT Regional (climate/soils) and local (aspect) physiography determine plant community composition. However, changes in initial floristic composition after a disturbance may be severe enough to alter the successional trajectory predicted by physiography. We addressed the question of which is more important, disturbance or physiography, in determining vegetation composition and the consequent successional trajectory. We evaluated understory vegetation of forest communities exposed to four disturbance types (control, single burn, diameter-limit cut, and first-removal shelterwood) 2–5 yr postdisturbance. Study sites were located within each disturbance type on northeast and southwest aspects within the Appalachian Plateau and the Ridge and Valley provinces. Vegetation composition was analyzed with nonmetric-multidimensional scaling, two-way nonparametric multivariate ANOVA, and indicator species analysis. The relationship between disturbance and key environmental variables, including canopy opening and soil fertility, was analyzed with generalized linear mixed models. There were 363 species in our study area. Composition differed by province and aspect. Composition also differed by disturbance but with a significant province interaction. Although physiography was more important, some species served as disturbance indicators that differed by disturbance type with two possible outcomes. First, expected successional trajectories (as defined by the regional and local environmental filters) may deviate toward recovery of native species (e.g., Epigaea repens) that benefit from low-level disturbance (as defined by fire as a filter). Second, successional trajectories after a relatively severe disturbance (as defined by shelterwood harvest as a filter) may deviate toward systems that are vulnerable to invasion by exotics or dominant native species.

Supplemental Material

ABSTRACT Soil diaspore banks are important temporal refuges for forest plants and, although fern spore banks are prevalent, they are understudied in relation to forest management practices. As urbanization increases, understanding the dynamics of spore bank resources in urban forests becomes increasingly important. Urban forests tend to have a greater propensity for species invasions and present challenges to plant species management not encountered in other managed forests. We studied the impacts of Lonicera maackii removal on fern assemblages aboveground and in diaspore banks in an urban forest. The diaspore bank assemblage had greater richness and abundance of ferns compared to the aboveground assemblage. Six years after management implementation, plots from which L. maackii was removed had fewer fern species and fewer ferns in the spore bank than paired control plots in which the shrub was not removed. Two nonnative ferns typical of residential ornamental plantings and with characteristics of potentially invasive species were also found in control plot spore bank assemblages. Environmental variables including pH and amount of leaf litter contributed to differences between spore bank fern assemblage structure in removal and control areas. These results indicate the need for further studies of fern diaspore banks in managed, urban eastern forests.

Supplemental Material

Persea borbonia and Persea palustris are two species in the Lauraceae family, native to the southeastern USA. Because of their similar morphologies, many authorities treat these two species as varieties of a more broadly defined group, Persea borbonia. Consequently, the species and communities associated with P. borbonia and P. palustris are not well defined. Extensive mortality has occurred in both species because of laurel wilt disease. Our objectives were to (a) determine whether P. borbonia and P. palustris are associated with different communities, (b) describe the communities and associated species in which P. borbonia and P. palustris occur, and (c) determine which communities support larger individuals of Persea spp. We analyzed data collected from 1988 through 2012 by the Carolina Vegetation Survey. To test differences among plant communities with natural growth of P. borbonia and P. palustris, we used ordination and analysis of similarities. We used cluster analyses and species-indicator analyses to group 388 plots into distinct communities. Results were verified by random forest analyses. We compared basal area of Persea spp. in each community with Kruskal-Wallis multiple comparisons. Persea borbonia and P. palustris communities were significantly different in species composition. Persea borbonia was limited almost exclusively to maritime coastal forests, whereas P. palustris had a larger geographical range, extending from near-coastal settings inland through the fall-line of the sandhills. The cluster and indicator species analyses identified eight unique groups, one for P. borbonia, and seven for P. palustris.