| Five contributions are made in this dissertation. Firstly, an improved protocol is introduced to quantify 3-D macropore networks. Important macropore characteristics were quantified, including macroporosity along soil depth, macropore network density, macropore size distribution, surface area, macropore length density, length distribution, mean hydraulic radius, tortuosity, inclination or angle, and connectivity (i.e., pathway and node density). The approach developed in this study quantified the distinct morphological features of macropore networks in two soil types and two land uses.;The second contribution presented in this dissertation is the understanding of the variation of macropore characteristics in different types of soils and land uses and with soil depth. Within the same soil type, the soils with pasture land use had greater macroporosity, pore length density, and node density, especially in the subsoil. Within the same land use, the Morrison sand displayed fewer macropores and weaker structure than the Hagerstown silt loam. The macropores in the subsurface (i.e., Ap2 and Bt horizons) were less tortuous, more vertically oriented, and less interconnected than the macropores in the Ap1 horizon. The analysis of variance indicated that a combination of soil type and land use needs to be considered to evaluate the variation of the macropore characteristics. Such results contribute to the enhanced evaluation of preferential flow and transport potential in soils under different land uses.;The third contribution of this dissertation is the development of the quantitative relationships between macropore characteristics and soil hydraulic parameters (i.e., saturated hydraulic conductivity and hydrodynamic dispersion). For the soil columns taken from the Hagerstown silt loam with two types of land uses (i.e., crop and pasture), saturated hydraulic conductivity (Ksat) of each soil horizon and of the whole column were measured and the breakthrough curve (BTC) for CaBr2 was determined. For all the soil columns with two land uses, macroporosity and path (i.e., the number of independent paths between two ends of the soil volume) explained 75% of the variation in Ksat. Within each land use, macroporosity, path, and tortuosity were the most important characteristics to estimate Ksat of the horizon but with different rank of importance. The path, hydraulic radius, and the angle explained 97% of the variation in the dispersion coefficient of all the soil columns. Additionally, the good correlation between the hydrodynamic dispersion and Ksat of the Bt horizon implied that the hydrodynamic dispersion was mainly controlled by the horizon with the lowest conductivity.;The fourth contribution of this dissertation is the examination of the lacunarity to test if macropore networks and flow patterns are fractal and if lacunarity has a diagnostic value in characterizing soil macropore structure and preferential flow pattern where fractal dimension alone could not. Relative lacunarity functions (RLFs) and pore fractal dimensions, both in 2-D and 3-D, were calculated for the macropore networks and tracer distributions at the five positions in the soil column scanned with a micro-CT. The lacunarity function reflected the size distribution of macropores and the spatial pattern of flow and transport. The RLFs indicated that the tracer distributions exhibited more self-similarity than the macropore networks. Lacunarity is a potentially powerful parameter that may be coupled with the fractal dimension to better describe and model soil structural properties.;The fifth contribution of this dissertation is a simplified method to quantify soil structure (i.e., porosity) and preferential water flow and solute transport (i.e., the velocity, solute mass, and concentration) in the soil macropore domain, matrix domain, and their interface. The voxel-based soil porosity and solute tracer concentration, mass, and velocity were quantified using micro X-ray CT and digital radiograph in an intact soil column (10 cm in diameter and 30 cm in length) of a Hagerstown silt loam. Only part of the macropores, especially the highly continuous biogenetic macropores at the subsurface, were active in transporting the solute. This study illustrates that preferential flow pathways in the intact structured soil consist of a complex network of earthworm burrows, root channels, inter-aggregate macropores, and mesopores or even micropores in the soil matrix. Modeling of this flow network and its dynamics would require a new approach different from the classical continuous-domain approach. (Abstract shortened by UMI.)... |
