5B.2 North American regional climate simulations with WRF/Noah-MP: Validation and the effect of groundwater interaction

Barlage, Michael, Mukul Tewari, Fei Chen, Kevin Manning, National Center for Atmospheric Research, and Gonzalo Miguez-Macho, University of Santiago de Compostela, Spain

Six-month regional climate simulations are conducted for two March – August periods 2002 (a dry year) and 2010 (a wet year) using the WRF model coupled to the Noah and Noah-MP land surface models. Within Noah-MP, three groundwater options are used to simulate the effects of aquifer interactions with WRF’s two-meter soil: free drainage (similar to Noah), groundwater scheme with default settings, and groundwater with equilibrium water table. Model results are verified against approximately 2000 surface stations (temperature, dewpoint, wind) and CPC gauge-based precipitation analysis. In general, using Noah-MP improves the warm and dry bias simulated when using Noah, especially in spring, though Noah-MP temperatures (dewpoint) tended to be too high(low) during summer daytime. All models have regional wind bias of less than 1 m/s. Precipitation results are mixed. All models perform similarly during the first three months, while the Noah-MP models had less precipitation in the summer months. This improved the Noah model wet bias in 2002, but exacerbated the dry bias in 2010. Regional analyses are conducted to determine the effect of groundwater interaction on the simulations. Five regions in the central part of the domain show significant upward transport of water from the aquifer in summer months. Simulations with interactive groundwater reduce the daytime summer warm/dry bias found in the Noah-MP simulations without groundwater access. Precipitation with Noah-MP over regions with groundwater interaction was generally lower, especially in the central US. Reduction in convective available potential energy in these regions contributes to a reduction in model-produced convective precipitation with Noah-MP.