Advancements In The RUC Land-surface Model For Use In The Next-generation HIGH-RESOLUTION Weather Prediction Models

Abstract

The RUC land-surface model (LSM) has been used as a land component in several generations of operational regional weather prediction models at National Centers for Environmental Prediction (NCEP). It was first implemented in the operational Rapid Update Cycle (RUC) weather model and next in the operational convection-permitting High-Resolution Rapid Refresh (HRRR) model based on the Weather Research and Forecasting Model (WRF). Since then, land- and snow-process treatments and static land information in the RUC LSM were improved in the HRRR upgrades at NCEP and now with the next-generation high-resolution models. These improvements found to be important in this study include modifications to direct evaporation from soil surface, canopy evaporation, transpiration, irrigation parameterization and, for cold season, to estimation of snow thermal conductivity and snow cover fraction important for better accuracy in treatment of grid cells with patchy snow. Advances in RUC model physics, as well as more accurate information about soil and vegetation, were tested in both off-line and coupled frameworks. They were implemented in the coupled regional weather application of the Unified Forecast System (UFS) and resulted in improvements in predictions of near-surface air temperature and moisture. These advances are described in this paper.