Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy



Major Professor

H. R. DeSelm

Committee Members

A. J. Sharp, Fred H. Norris, M. E. Springer, J. D. Caponetti


The complex physiographic and climatic history of the Great Valley of East Tennessee has resulted in development of diverse alternating ridge and valley land forms and soils which provide equally diverse forest habitats.

The objectives of this study were (1) to quantify certain topographic features, (2) to analyze statistical relationships of vegetation and tree taxa to selected soil properties and topographic features, (3) to assess magnitude and predictability of these relationships, (4) to document taxonomic diversity of tree taxa, (5) to document and delimit contemporary forest communities and describe interactions with soil-site properties, and (6) to consider relationships of these communities to other forests of the Southern Appalachians.

In all, 684 concentric 1/10- and 1/100-acre temporary, circular plots were established among 58 old-growth forest stands; plot number varied with stand size. On 1/10-acre plots, trees 5 in. dbh and greater were measured and recorded and stumps of Castanea dentata were counted. On 1/100-acre plots, trees less than 5 in. dbh were recorded by taxon. At each plot, slope characteristics were recorded. Within 34 stands, soil samples were collected and physical properties were described. Properties determined in the field were thickness of A and B horizons, stone percent, pedon and mottling depth. Laboratory analysis included pH and textural determinations for each horizon; available water was calculated from published data. Topographic properties were quantified by field measurement and determinations from topographic maps. These included slope angle and aspect, calculation of slope form by width/depth ratios of concavities and convexities, slope position by relative relief, size of ridge and valley units, external protection afforded by adjacent land forms, and local relief.

Statistical analyses included simple linear correlations among soil and site properties and tree taxa. Linear models were as informative as logarithmic models. Stepwise multiple regression was used to predict relationships of relative density, relative basal area and importance value of selected overstory tree taxa to selected soil and topographic properties; quadratic and interaction terms of independent variables were also employed.

Soil and site correlations showed several interrelationships. Soil separates were frequently intercorrelated. Soil depth and stone content are two important variables controlling water availability in this area. Local relief and topographic length and width were frequently correlated with site variables; ecological significance is indirect and other on-site characteristics must be considered concomitantly. Correlations between slope angle and soil properties were the most important soil-topography interactions.

Stepwise regression analysis indicated that vegetation and values of overstory tree taxa were not easily predicted. Total density, total basal area, and values of oak, hickory and pine taxa had R2 values of 0.10 or less. Three taxa had R2 values of 0.25. Equations for Fagus grandifolia had the highest predictive values. Topographic characteristics accounted for more variation among taxa than soil properties. The most important soil variable predicting importance of various taxa was water availability. Other commonly correlated soil variables made limited contributions. Interaction and quadratic terms of independent variables were of limited value. Genetic variability among taxa, inherent differences among soils, low precision and accuracy in measuring soil characteristics, omission of important environmental parameters, and unknown stand histories were considered factors contributing to low predictability.

Forest communities recognized were associated with soil-parent material units of ridges and valleys. Overstory density was 110 ± 30 stems per acre; basal area was approximately 100 ± 40 ft.2 per acre. Of the 57 tree taxa recognized, the majority was widely distributed in deciduous forests; oak and hickory taxa comprised 30 percent.

Four vegetation complexes were recognized as the central core of forest development. The White Oak complex, comprised of the largest number of communities, is largest in aerial extent. The dominant or co-dominant, Quercus alba, appears to have the greatest ecological amplitude of any taxon in this area. Chestnut Oak, Mixed Mesophytic and Tulip Poplar Complexes and pine, restricted upland oak, and bottomland hardwood communities were also recognized. Major taxa that apparently replace Castanea dentata are Quercus prinus and/or Quercus alba.

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

Included in

Botany Commons