Date of Award


Degree Type


Degree Name

Master of Science


Ecology and Evolutionary Biology

Major Professor

Edward E. C. Clebsch

Committee Members

Ronald L. Hay, Edward R. Buckner, Peter S. White


Fire history, the initial effect of fire disturbance on community structure, the response of communities one and two years after fire disturbance, and the present-day vegetation patter were examined in the westernmost portion of the Great Smoky Mountains National Park in Tennessee and North Carolina.

Sectioning of fire scars, field observations, historical accounts, and fire control records indicated that the fire disturbance regime has changed dramatically during the last 200 years. Man-caused fires have probably been an important disturbance since Indians migrated into the area over 12 thousand years ago. Between 1790 and 1940, Euro-American man frequently burned the landscape. On south facing upper slopes the mean fire interval between 1860 and 1940 was 12.7 years. Field observations suggested that south facing upper slopes burned most frequently while north facing lower slopes and ravines burned least frequently. The interval between fires increased with increasing elevation. Under the present fire suppression policy, the fire disturbance regime probably resembled on other period in the last thousands of years. Estimates of lightning-caused fire frequency and potential size indicated that a policy which allows some natural fires to burn unhindered will not maintain the current or the historical vegetation pattern.

Fire disturbance severity was found to be a function of fuel, topography and weather. South facing upper slopes appeared to be the topographic positions most prone to severe fires. Disturbances to the canopy such as southern pine bark beetle attack also increased the chances of severe fire. Canopy closure, litter depth, species richness, basal area and stem density all decreased immediately following fire. Fire disturbance can be considered both a product of the community (endogenous) and a product of the environment (exogenous). The ability of trees to survive fires was a function of severity, species, diameter at breast height, bark thickness and the ability to initiate epicormic branches. Rapid growth rates were also of survival value when fire intervals were short.

Seedlings, seedling sprouts and root collar sprouts were important reproductive form one and two years after fire. The relative importance of each reproductive form was a complex function of position along the topographic-moisture gradient, litter depth, canopy closure, fire severity and community structure prior to disturbance. The ability of most species to generate sprouts indicated few species were actually eliminated from a site by fire. Yellow poplar, pine, sumac, sourwood, devil's walking stick, black locust and princess tree seedlings where most abundant in severely burned forests.

Forest canopy data from 100 permanent plots were examined using indirect ordination, direct ordination and community classification. Slope position, aspect, slope steepness, potential solar beam irradiation (SBI), concavity, elevation, disturbance by man and the time since last fire were all important in determining species distributions. Forests which were not recently cleared by man for agriculture or timber were divided into 7 cover types. From mesic to xeric sites these were included: cove hardwoods, hemlock, mixed hardwoods, mixed oak, white pine, chestnut oak-scarlet oak and yellow pine. Forests cleared by man had old-field pine on xeric sites and successional hardwood on mesic sites. Major successional change is expected in the old-field pine, successional hardwoods and yellow pine cover types. Continued fire suppression will lead to a substantial decrease in the yellow pine cover type in the next 50 years. Allowing some lightning-caused fires to burn unhindered will probably not maintain the present extent of the yellow pine cover type.

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