Clay-rich soils that crack on drying and swell on wetting cover a considerable fraction of the Earth?s land surface. The goal of this research is to improve the hydrological response of watershed models used in regions that have such soils. To accomplish this goal, the ability to spatially and temporally partition precipitation into infiltration and runoff must be improved. Partitioning of precipitation into infiltration and runoff is highly dependent on the spatial and temporal distribution and size of cracks in these soils. When a cracking clay soil is very dry, precipitation is partitioned nearly all to infiltration via the large physical capacity of the cracks. When a cracking clay soil is very wet and infiltrating water must travel through the fine pores of the clay matrix, significant amounts of precipitation are partitioned to runoff. What happens between these two soil moisture extremes is essential in predicting hydrological, agronomic, and environmental responses of watersheds with clayey soil to rainfall events, but has not been adequately characterized by hydrologists or soil scientists, and is the focus of this research. Recent developments in technologies to electromagnetically sense, collect, and manage spatial soil and environmental information support quantification of the state of a cracking soil at any given moment and location, and allow an improvement in the ability to track the spatial distribution and size of cracks. In our research effort, we will measure and characterize the spatial heterogeneity and temporal dynamics of soil cracking as a function of terrain attributes, soil water dynamics, and basic chemical, mechanical, and physical soil properties. Measurements will be made in a highly instrumented experimental watershed in Central Texas. Focusing our understanding of crack behavior on basic information and properties we will promote transfer of knowledge to other watersheds. In addition to being able to track the degree and distribution of cracking, the scale that this information needs to be addressed in hydrological models to improve their accuracy must be determined. To address this need, a hydrological model composed of a two-dimensional, diffusive wave runoff module and a one-dimensional, point-column biophysical module will be used to evaluate cracking as a function of landscape position, soil characteristics, and antecedent moisture conditions; allowing a dynamic partitioning of rainfall between infiltration and runoff during precipitation events. The model operates on a 3-dimensional grid and includes the ability to handle spatial variations in soil properties, moisture content, and other factors that may be used to predict the instantaneous size and water holding capacity of cracks. Equations for crack distribution, size, and capacity distilled from the observations in this study will be added to the model. The model will link the behavior of cracking within each grid cell to watershed hydrographs via runoff to surrounding grids cells in the field, the sub-basin, and the watershed. The contributions of each cell to infiltration and runoff are calculated, yielding spatial patterns of the fate of precipitation. By comparison of measured hydrographs from the experimental watershed with hydrographs generated from the model as the size of the grid cells are increased, the scale that information on soil cracking needs to be addressed to produce a degree of accuracy will be characterized. This research is collaborative, combining a soil hydropedologist, an environmental physicist, and a landscape hydrologic modeler; and bridges scales from millimeter to kilometer by linking soil microenvironments, field scale heterogeneity, and watershed hydrology. The research findings will be fully integrated into graduate and undergraduate courses in soil hydrology, soil morphology, environmental physics, and spatial statistics taught by the investigators. Collaboration with a local school district will provide outreach exposing rural students to the sciences and real-world, field-based science problems.
Certain types of soils have chemcal and physical properties that cause the soil to shrink and swell by a large amount, with large cracks opening and closing depending on the moisture content of the soil. The Blackland soils in Texas are one of these soils. Blacklands soils are very productive for agriculture, but are difficult to simulate in hydrologic and agricultural models, due to their high shrink-swell properties and rapidly changing crack volume. This grant funded part of a collaborative project between the University of Wisconsin-Madison and Texas A&M University to improve the state of the art in modeling such types of soils. In this project, a macropore (large pore) model has been adapted in the Precision Agricultural-Landscape Modeling System (PALMS) in order to better simulate the way in which opening and closing cracks allow water to either infiltrate deep into the soil when cracks are open, or prevent water from infiltrating when cracks are closed at the surface. The model has been shown to be an improvement over previous methods in simulating the moisture profile and crack volume of Blacklands soil. Field studies show, however, that water storage in the cracks is an important process to simulate. This process was subsequently included in PALMS. This improvement will be tested in the final year of the Texas A&M University collaborators' project.
