Evaluation of Modeled and Reanalysis Estimates of Frost Depth through Comparison to Ground Observations
Abstract
Frozen soils can withstand heavy vehicle loads where the soils would otherwise be too weak. This is important as vehicle mobility models require input of the ground conditions to assess seasonal traffickability. Modeling these frost depths uses the properties of the soil along with any snow cover and air temperature data. Though weather stations with air temperature data are becoming more widespread, station density varies worldwide. Gridded reanalysis products, on the other hand, provide weather data on a global scale. Our objectives are (a) to evaluate the usefulness of reanalysis data to provide estimates of frost depth and (b) to determine the accuracy of simulated frost depths using either measured or reanalysis air temperature data. We use two one-dimensional models: a coupled heat and moisture flow model that simulates frost and thaw depths throughout the winter and a decoupled heat and moisture model that calculates the maximum frost depth. We compare frost depths determined from measured soil temperatures at stations in North Dakota and Minnesota and reanalysis frost depths with modeled values using both station and reanalysis air temperature data. Modeled frost depths depend more strongly on the depth and density of the snow layer than on the soil properties.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2019
- Accession Number
- AD1073394
Entities
People
- Carrie Vuyovich
- Kathleen Jones
- Lynette Barna