A tool for downscaling weather data from large-grid reanalysis products to finer spatial scales for distributed hydrological applications

A downscaling tool was developed to provide sub-daily high spatial resolution surfaces of weather variables for distributed hydrologic modeling from NASA Modern Era Retrospective-Analysis for Research and Applications reanalysis products. The tool uses spatial interpolation and physically based relationships between the weather variables and elevation to provide inputs at the scale of a gridded hydrologic model, typically smaller (∼100 m) than the scale of weather reanalysis data (∼20-200 km). Nash-Sutcliffe efficiency (NSE) measures greater than 0.70 were obtained for direct tests of downscaled daily temperature and monthly precipitation at 173 SNOTEL sites. In an integrated test driving the Utah Energy Balance (UEB) snowmelt model, 80% of these sites gave NSE > 0.6 for snow water equivalent. These findings motivate use of this tool in data sparse regions where ground based observations are not available and downscaled global reanalysis products may be the only option for model inputs.