Masters Theses

Date of Award

12-1986

Degree Type

Thesis

Degree Name

Doctor of Philosophy

Major

Plant, Soil and Environmental Sciences

Major Professor

W.L. Parks

Committee Members

E.R. Buckner, G.L. Butnley, R.J. Lewis, G.W. Smalley, J.D. Wolt

Abstract

Relationships among soil morphology, soil water, and forest tree growth were investigated on three forested Cumberland Plateau landtypes at two locations, and a forest land classification system was evaluated. Two soil pits on each plot were opened for morphological descriptions, characterization, and moisture cell installation. Moisture cells were read for two years. Thirty-two soil properties from three genetic soil horizons at 132 points located with a 10 meter grid were used in multivariate statistical analyses. Dominant soils were Typic Fragiudults on uplands; Humic Hapludults and Typic Fragiumbrepts on slopes; and Aquic Dystrochrepts and Typic Haplaquepts on first-order bottoms. Parent materials were Pleistocene loess over shale and sandstone residuum. Clay mineralogy of the upper sequum was relatively young. Chlorite was common in A horizons, but only acid upland soils contained hydroxy-interlayered vermiculite. Kaolinite and quartz dominated residual soils. Gibbsite was in the most leached soil horizons and within buried paleosols. Cation exchange capacities averaged 15 cmol(p+)kg-1 on uplands and bottoms and 20 cmol(p+)kg-1 on slopes. Base saturation ranged from less than 10 percent in bottoms to 45 percent in slope A horizons. Base saturation and cation exchange capacity increased as clay and organic matter increased. Soil moisture distribution in soil profiles and landscapes was related to soil morphology and landtypes respectively. Distribution of citrate-dithionite extractable Fe and Mn in profiles and landtypes was related to measured soil moisture distribution. Stem analysis of forest dominants revealed height growth related to the soil moisture gradient across the landscape. Site index of dominants on uplands and slopes increased down-slope, and yellow-poplar height on bottoms increased with increasing depth to the winter water table. Maximum likelihood factor analysis reduced 32 soil properties to four factors representing A horizon properties, soil texture, subsurface cations, and soil drainage and thickness. The 25 retained soil variables extracted 71 percent of the variance. Discriminant analysis classified all 132 grid observations into correct landtypes, revealing that measured soil properties were related to landtypes. The forest land classification system appears to be a viable method of grouping soils into units suitable for forest management on the Cumberland Plateau.

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

Share

COinS