Impact of a Dense Surface Network on High-Resolution Dynamical Downscaling via Observation Nudging

Abstract

AbstractHigh-density surface networks have become available in recent years in a number of regions throughout the world, but their utility in high-resolution dynamic downscaling has not been examined. As an attempt to fill such a gap, a suite of high-resolution (4 km) dynamical downscaling simulations is developed in this study with the Weather Research and Forecasting (WRF) Model and observation nudging over Liaoning in northeastern China. Three experiments, including no nudging (CTL), analysis nudging (AN), and combined analysis nudging and observation nudging with surface observations (AON), are conducted to downscale the CFSv2 reanalysis with the WRF Model for the year 2015. The three 1-yr regional climate simulations were compared with the independent surface observations. The results show that observational nudging can improve the simulation of surface variables, including temperature, wind speed, humidity, and pressure, more than nudging large-scale driving data with AN alone. The two nudging simulations can improve the cold bias for the temperature of the WRF Model. For precipitation, both the simulations with AN and observation nudging can capture the pattern of precipitation; however, with the introduction of small-scale information at the surface, AON cannot further improve the simulation of precipitation.