Good estimates of storm motion are essential to improved air traffic control operations during times of inclement weather. Automating such a service is a challenge, however, because meteorological phenomena exist as complex distributed systems that exhibit motion across a wide spectrum of scales. Even when viewed from a fixed perspective, these evolving dynamic systems can test the extent of our definition of motion, as well as any attempt at automated tracking of this motion. Image-based motion detection and processing appear to provide the best route toward robust performance of an automated tracking system.