Publication Date


Document Type


Committee Members

Christopher Barton (Committee Member), Mark Goltz (Committee Member), Allen Hunt (Advisor), W. Berry Lyons (Committee Member), Thomas Skinner (Committee Member)

Degree Name

Doctor of Philosophy (PhD)


The concept of soil formation has been studied since the beginning of 19th century. However, until recently, there has been little concrete progress on developing an analytical result to relate soil depth or quality to measurable variables that represent the five soil-forming factors including time, parent material, topography, climate, and organisms. It has become increasingly clear that soil formation rates are closely related to chemical weathering rates. In this dissertation, we propose a theoretical approach to model soil formation process within the theoretical framework of percolation theory, which has been shown to successfully predict solute transport in heterogeneous media. From percolation theory, solute transport rate does not equal to flow rate beyond the length scale of a typical pore size, as is the case in Gaussian solute transport. Rather, it diminishes in accord with heavy-tailed solute arrival time distributions as it travels. The basis of our model relies on the hypothesis that the chemical weathering of bedrock is simultaneously the limiting factor for soil formation and most strongly limited by solute transport in porous media. To test the hypothesis, we propose a revised method to calculate Damkohler number within the same theoretical framework to evaluate the relevant importance of solute transport in limiting chemical weathering, and results imply that chemical weathering is nearly always solute transport-limited in natural media. We then examine the proposed models by comparing predictions with field data across a wide range of climatic conditions, as well as at steep topography. Results show good agreement between predictions and field observations. We also present two applications of the proposed model to geomorphology to examine the local steady-state assumption of soil and to distinguish steady and stochastic erosion process in threshold landscapes. The applications demonstrate the potential to adopt our model into geomorphological models such as landscape evolution models, and landsliding models to predict shallow landslides.

Page Count


Department or Program

Department of Earth and Environmental Sciences

Year Degree Awarded