Submitted to: American Geophysical Union
Publication Type: Abstract Only
Publication Acceptance Date: 10/9/1998
Publication Date: N/A
Technical Abstract: Simulation of snow cover and melt is paramount to accurately predicting runoff from and soil frost penetration on agricultural lands in cold regions. We examined the performance of two soil frost models, SHAW and SOIL, in simulating snow cover on a prairie soil over two winters in Minnesota. Snow cover simulated by these finite difference models was compared with that measured in a harvested cornfield with 90% residue cover. The models predicted snow cover with reasonable accuracy in the fall, but tended to overestimate snow cover in the spring during snowmelt. The SHAW model simulated snow cover and melt, and therefore soil frost depth, better than the SOIL model. These models can be improved in estimating depth and occurrence of frozen soil by better simulating fall and spring snow cover and melt.