A ten hectare catchment located in south-central Minnesota has been instrumented to quantify the spatial variability of frost depth within the catchment area. The catchment forms a depressional area which inherently has a wide range of soil moisture levels in addition to a complete range of slope aspects. Vegetative cover on the catchment includes forest, shrub, grass, and cultivated crops. Instrumentation has been located on the catchment to provide a measure of the spatial variability of water content, soil temperature and frost depth. The instrumentation includes frost tubes, thermocouples, piezometers, neutron probe and capacitance probe access tubes, and time domain reflectometry probes. In addition, a temporary system for measuring the spatial variability of snow depth and snow water equivalent has been employed at the site. Measurement of some variables at the site began in December 1993, but the instrumentation system was not completed until the summer of 1994. A complete data set was collected in the winter of 1994 and is continuing through the winter of 1995. Data collected during the winter of 1994 are being analyzed to evaluate correlations of frost depth with topographic attributes such as slope aspect and upslope contributing area in addition to site specific characteristics such as bulk density, soil moisture, snow depth, and residue cover. The study will also focus on how these correlations change during the freezing and thawing periods. Topographic features have been quantified based on a detailed topographic survey and application of TAPES, a digital terrain model. In addition, we are applying models of frost penetration to test their abilities to simulate frost penetration into the soil profile at selected instrumented sites. The models considered include the SHAW model, a modified Stefan equation, and a model based on the principles of the Benoit model. Soil physical properties and surface cover measurements have been made at these selected sites to provide input to the models. A sensitivity analysis of the SHAW model will further point out characteristics most critical to simulating the freezing process in soil. Using the correlations of frost depth with topographic attributes and the results of pre-established models will lead to recommendations for spatially distributed soil frost modeling.

Dr. John L. Nieber
Department of Biosystems and Agricultural Engineering
University of Minnesota
St. Paul, MN 55108
Phone: (612)-625-6724
Fax: (612)-624-3005
E-mail: nieber@gaia.bae.umn.edu