A Statistical Nowcasting Method for an Urban Wind Field at Rooftop Level

An accurate description of the current urban wind field, given a sparse network of automatic weather stations, requires a statistical approach to the description of the spatial wind field. This approach is needed due to frequent diurnal periods in which the spatial wind field is heterogeneous. Such periods may be the result of various phenomena, such as changes in flow regimes or passage of weather fronts.

In the presentation a new method will be shown for statistical nowcasting of the spatial wind. The method is based on an analysis of a collection of wind measurements in different urban rooftop locations. This collection defines the spatial wind distribution. The method will be demonstrated on the metropolitan area of Tel-Aviv. It is shown that the spatial wind distribution can be very well represented by a small sample of only four weather stations.

Given online measurements of these representative weather stations, the method provides a real-time estimation of an elliptic tolerance region around the mean wind vector. This tolerance region bounds a given proportion of the wind realizations in the urban area of interest.