Spatial bias analysis for the Weather Research and Forecasting model (WRF) over the Apulia region

The Weather Research and Forecasting model (WRF) based on CORINE land-cover database has been used to simulate 10m wind speed and 2m temperature over the Apulia region. A validation procedure against ground data from 48 monitoring stations for both a winter and a summer period inform on the spatial distribution of the model bias. A preliminary analysis based on three indices of model performance revealed that the summer period is better simulated than the winter one. Computation of Moran's index for the spatial distribution of the model bias shows that the summer 10m wind speed bias results to be weakly autocorrelated while this autocorrelation is absent for both winter and summer 2m temperature as well as for winter 10m wind speed. On the other hand, semivariogram analysis shows a slight correlation with a range value less than 2000m for 2m temperature's winter and summer biases.