Date of Award


Degree Name

Biological Sciences


College of Science

Type of Degree


Document Type


First Advisor

Frank S. Gilliam


The effects of fire on the mixed hardwood forests in West Virginia have long been of interest both biologically and economically. This study examined the effects of a Fall, 1991 wildfire in Wayne County, West Virginia on soil fertility and herbaceous layer response.

Fire intensity was determined by comparison of 1) remaining organic material, 2) understory survival and 3) height of charcoal on the stems of trees following the fire. Study plots were established in areas representing fires of high, moderate and low intensities with a control established in an adjacent unburned area of similar aspect and elevation.

Composite soil samples were taken from each site and analyzed at a soil testing laboratory for pH, cation exchange capacity and extractable ions including N03 , NH4+, PO43,-,K+, Ca2+ and Mg2+. Soil texture was determined by the hydrometer method for each plot. Soil fertility response was evaluated with a plant bioassay experiment, using Poa pratensis (Kentucky bluegrass) grown in soil from each burn type and amended with a variety of nutrient solutions containing N, P, K, Ca and Mg present in full treatment. A seedbank study was performed to further assess potential herb layer recovery. Replicate sections of forest floor and surface soil were taken from each burn type and allowed to develop in the greenhouse with only periodic watering. Herbaceous layer and canopy cover measurements were made at each site during the spring, summer and fall of 1992. Overstory species were measured and importance values determined.

The bioassay suggested that soil fertility was increased by the fire in the burned plots. Nitrogen may possibly be the most limiting nurtient on all plots. The least fertile soils were found in unburned areas. There was a negative correlation between herbaceous cover and canopy opening. Herbaceous cover was greatest on high intensity burned plots and lowest on unburned plots. Both richness and diversity were greatest in the burned plots. Canonical discriminent analysis was performed using SAS CANDISC procedures. Species composition over time varied most on low and high burned plots. The species composition of unburned and moderately-burned plots were most similar throughout the first growing season.

Species which developed in the seedbank study were generally well-represented in the herbaceous layer (80.6%). A few species which developed in the seedbank sample were not represented in the herbaceous layer. This suggests that the seedbank potential was limited to a degree in the herbaceous layer by some on-site factor(s).

It is suggested that herbaceous cover following burning is not determined by any single factor (fertility, canopy cover, etc.) but a combination of all on-site factors. The highest herbaceous cover was found in areas of moderate fertility increases following burning. This suggests that the increase in solar input as a result of canopy openings caused the fire may be the most important factor in increasing herbaceous growth. This was not the case for low and moderately burned plots. Canopy cover was the same at the end of the growing season yet herbaceous cover was much greater at low burned plots. This may indicate that the interaction of factors which allowed for the greatest herbaceous cover were not the same at each plot.


Hardwoods - West Virginia.

Forests and forestry - West Virginia.

Forest fires - West Virginia.