Polaris: Towards an Improved Representation of Spatial Heterogeneity in Land Surface Models.

Current computation and data resources are allowing the hydrologic and climate modeling communities to move towards simulating the terrestrial water, energy, and biogeochemical cycles at higher spatial resolutions (100 – 1,000 m) over continental extents. However, the coarse resolution of many global environmental datasets (e.g., soil) limits the benefits of increased model resolution. Digital soil mapping presents a promising path forward by using legacy soil databases and available high-resolution information to create state-of-the-art global soil databases. In this presentation, I will introduce POLARIS, a new database of soil series probabilities maps and derived soil property maps that have been produced over the contiguous United States at a 30 m spatial resolution. This innovative database was constructed using available high-resolution geospatial environmental data and the DSMART-HPC algorithm to remap the Soil Survey Geographic (SSURGO) database. The soil series vertical profile information in SSURGO was then harmonized and mapped using the soil series probability maps and the PROPR algorithm to create a 30 m soil properties database over the contiguous United States. In this presentation, I will illustrate and discuss how the final data product can be used towards improving the representation of soil spatial and vertical heterogeneity in continental hydrologic and climate models.