Assessment of US Frost Depth Maps Considering Climate Change Effects

The goal of this paper is to determine the accuracy of US frost depth maps used to determine the foundation depth in cold regions. Estimating the frost depth continues to represent a challenging task to engineers, specifically in cold regions. Frost depth and heave put significant risks on several infrastructure including highways, runways, pipelines, and buildings. It is, therefore, recommended to locate the foundation depth of this critical infrastructure below the expected frost depth at any given location. In the United States, predictions of the frost depth for foundation design have been performed using maps that were developed decades ago prior to the significant climate changes we experience nowadays. In this study, we will revisit these maps with the sole aim of assessing their accuracy considering recent weather trends. We employ existing semi-empirical and fully empirical predictive models to estimate the frost depth at different US locations using recent weather data. These models estimate the frost depth based on correlations between frost depth and the cumulative freezing degree day (CFDD); in this study, the latter is estimated using recent weather data. We then assess the accuracy of existing frost depth maps by comparing the predicted frost depth for the selected locations to those estimated from existing maps. The results of the analysis show that frost depth is a function of meteorological parameters and soil's dry density and water content. We also conclude that the existing contour maps overestimate the freezing depth when compared to the values predicted by the modified Berggren formula as a representative of semi-empirical equations widely employed to calculate frost depth.