Multiscale Storm Identification and Forecast

We have recently developed a hierarchical K-Means clustering method for weather images. Using this technique, it is possible to identify storms at different scales. In this paper, we will describe an error-minimization technique to identify movement between successive frames of a sequence and show that we can use the K-Means clusters as the minimization kernel. Using this technique in combination with the K-Means clusters, we can identify storm motion at different scales and choose different scales to forecast based on the time scale of interest. We will show examples of this storm identification and forecast in action on weather radar images.