Experimental WRF Model

The numerical ARW WRF model (Weather Forecasting Research-Advanced Research WRF) is used to forecast winds. This is a model for forecasting mesoscale weather features (horizontal smaller scale to hundreds of kilometers), with operational and research capabilities. The WRF model-ARW is initialized with data from the Global Forecast System model (GFS). Among the variables used to initialize the model are: pressure atmospheric temperature, size and direction of winds and humidity. It also uses the topography of the region. The resulting product is at a resolution of 4 kilometers, that is, there is a prognostic value of every 4 kilometers, and 27 vertical levels. The 27 vertical levels "follow" the topography of the region. In terms of temporal resolution, forecasts are produced out to 48 hours at intervals of 3 hours. The forecast figure consists of a contour map of the magnitude of the wind knots and wind barbs that indicate both the direction and magnitude of the wind. The regions of colors indicate the magnitude of the wind, while the "wind barbs" indicate the direction from which the wind is coming. ARW WRF wind forecast for Puerto Rico and Virgin Islands are presented here.