Masters Theses

Date of Award

12-1981

Degree Type

Thesis

Degree Name

Master of Science

Major Professor

H. R. DeSelm

Committee Members

Fred H. Norris, David K. Smith

Abstract

Plant communities of sandstone outcrops in six research areas on the Cumberland Plateau are determined as follows. Sample plots were distributed within four subjectively delineated life form zones: (1) Lithophyte Zone (substrate is bare rock), (2) Cryptogam-Herb Zone (usually some soil; no woody plants), (3) Shrub-Herb Zone (some soil; woody plants); and (4) Tree Zone (trees predominate). Three strata were sampled within each zone as appropriate: (1) trees and saplings, (2) other vascular plants, and (3) non-vascular plants. Cover was estimated for each species within each sample plot. Samples within each stratum of each zone were then grouped into communities using Reciprocal Averaging Ordination. Groups were tested using Stepwise Discriminant Analysis. Sorenson's (1948) index of similarity was used to compare communities of the same stratum between zones, and contingency tables were employed to examine association of communities of different strata within each zone.

Communities were characterized environmentally by calculation of means or medians of recorded environmental variables-- aspect, slope, microtopographic shape, degree of shading, canopy closure and height, thickness of soil horizons, soil depth, and soil pH. Communities were further characterized through known habitat preferences of dominant taxa, and through positions of communities on ordination axes. Average environmental conditions of zones were also compared.

The Tree Zone was vegetated by Pinus virginiana Community in all samples with the exception of one northerly aspect, dominated by Tsuga canadensis. Samples from vegetation islands surrounded by rock often had no tree species other than P. virginiana. The Tree Zone understory communities included several dominated by deciduous subsaplings (various species), and a P. virginiana Subsaplings Community, Vaccinium aboreum Community, V. vacillans and Smilax rotundiafolia Community, and a Grass-Forb Community. Non-vascular plants had low frequency compared to those of other zones.

Shrub-Herb Zone vascular plant communities included V. aboreum, V. vacillans-S. rotundifolia, Kalmia latifolia, Gaylussacia baccata, P. virginiana Subsaplings, Grass-Forb, and Helianthus longifolius-Danthonia sericea Communitis. Non-vascular plants were most profuse beneath the last four communities. The most frequent non-vascular communities were dominated by various proportions of Polytrichum commune, P. juniperinum, reindeer lichens (Cladina spp.) and Cladonia caroliniana.

Non-vascular plants usually had greater cover than vascular plants in the Cryptogam-Herb Zone. Cryptogams prevalent in the Shrub-Herb Zone were again important, as well as Campylopus spp. and Sphagnum spp. Vascular plant communities included Aster surculosus-Liatris microcephala, Talinum teretifolium-Grass-Annual Forb, Bigelowia nuttallii, and Panicum dichotomum Communities. Sedum smallii and Selaginella rupestris Communities were also observed on several sites.

Lythophyte Zone communities were exclusively compose of non-vascualr plants: the Grimmia laevigata, Cladonia caroliniana, Squamulose Cladonias, Powder Crustose Lichen, Filamentous Algae-Inkspot Crustose Lichen, and Mixed Crustose Lichens-Xanthoparmelia conspersa Communities; the last was the most frequent.

The six research areas were also compared floristically. Percentages of life forms proved similar between sites. Number of taxa increased with sample size (as expected) and also with outcrop size. Small outcrops, therefore, were usually depauperate. Sorenson's (1948) index of similarity was calculated between research areas and indicated that distance between the areas was not an important enough factor in their floristic similarity to be discernible through this technique.


Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Botany Commons

Share

COinS