Date of Award:

5-1977

Document Type:

Thesis

Degree Name:

Master of Science (MS)

Department:

Watershed Sciences

Department name when degree awarded

Watershed Science

Committee Chair(s)

Gerald F. Gifford

Committee

Gerald F. Gifford

Committee

George E. Hart

Committee

Glen E. Stringham

Abstract

The objectives of this study were (a) to apply sediment and associated plot data from various infiltrometer studies to the parameters in the Universal Soil Loss Equation, a modified version of the original Musgrave Equation, and a modified version of the original Universal Soil Loss Equation, and compare the computed results with the measured soil loss, (b) to suggest reasons for any differences between computed and measured soil loss, and (c) to suggest improvements for each equation so that it will give results near the measured soil loss. The data used consisted of 2805 infiltrometer plots collected by previous researchers in a variety of rangeland conditions, both in the western United States and Australia, and included the necessary information needed to compute the factors in each of the above equations. Simple and multiple linear regression techniques were used to make the evaluations by computing the coefficient of determination (R2), correlation coefficients (r), and to optimize each factor in the equations by placing an exponent on it.

The results showed that the three soil loss prediction equations are not universal, but, for the most part, explain sediment yield with varying degrees of accuracy in different situations with no apparent trends or patterns. However, most individual mine sites and other sites with loosely consolidated soil resembling fallow conditions showed high R2 values when the computed sediment yield was regressed against measured sediment yield. Little improvement was made in reducing the variability of the equations by placing exponents on each factor indicating that the factors, as determined in each equation, do not explain sediment yield under western rangeland conditions. In summary, the prediction equations are not recommended as "universal" predictors of sheet erosion in western rangelands, but, may be applied in specific situations.

Checksum

e4444bc97f139e232ec7ff85393a915d

Share

COinS