In this paper, we describe the use of statistically derived hierarchical clusters of weather data to derive movement estimates from pairs of frames in a time sequence. We show that the use of hierarchical clusters enables small cells to be tracked over short periods of time while using the movement of the larger scale features they are embedded in for longer periods.

The motion estimator has been applied both to reflectivity data obtained from the National Weather Service Radar (WSR-88D) and to cloud-top infrared temperatures obtained from the GOES-11 satellite. We demonstrate the results on both these sensors.