Evaluation of Weather Research and Forecasting Model Predictions Using Micrometeorological Tower Observations

Here we assess the predictive skill of short-range weather forecasts from the Weather Research and Forecasting (WRF) model with the help of micrometeorological tower observations. WRF model forecasts at a 3-h temporal resolution and 5000-m spatial resolution are compared with ground observations collected at micrometeorological towers during the year 2011 over the Indian landmass. Results show good agreement between the WRF model forecast and tower observed surface temperature and relative humidity, 10-m wind speed, and surface pressure. The WRF model simulations of surface energy fluxes, such as incoming shortwave, longwave radiation, and ground heat flux are also compared with micrometeorological tower measurements. Relatively high errors in incoming shortwave radiation flux may be attributed to the lack of accurate cloud prediction and the non-inclusion of aerosol load. The cyclic pattern of errors in surface relative humidity is found to be tightly and oppositely coupled with the incoming longwave radiation flux. Errors in soil heat fluxes during daytime hours are dominated by errors in the incoming shortwave radiation flux.