Spatio-temporal wind speed predictions for Germany

State of the art wind forecasting models, like Numerical Weather Predictions utilize huge amounts of computing time. Some of them have rather low spatio-temporal resolution. Time series prediction model accomplish good results in high temporal settings. Moreover, their consumption of computing capacities is relatively low and return accurate short-term to medium-term forecasts. The recent literature shows increasing interest in the topic of spatial interdependence. This article deals with a spatial and temporal model for wind speed. We describe the temporal model structure independently on spatial correlations. Therefore, seasonality and a huge correlation structure are included. Subsequently, the model is extended and a spatial structure is included. The data set includes ten minute observations of several measurement stations in Eastern Germany. The validation procedure shows that the model is reliable.