Abstract: A weather radar system (21) utilizing enhanced fuzzy logic processing (300) to detect atmospheric microburst events. In addition to the wind shear likelihood (250), low level surface reflectivity likelihood (252), and clutter likelihood (254) used by existing fuzzy logic based microburst detection systems to generate a combined microburst likelihood, one enhancement includes using reflectivity measurements of high level storm activity to generate an independent high level aloft reflectivity likelihood image (310) that corresponds to and confirms the findings of the low level surface reflectivity likelihood (252). Another enhancement includes using a series of combined likelihoods saved over time (635) for N successive scans of the weather radar system (21) to generate a history likelihood (620) that eliminates spurious detections from one scan to the next.

Abstract: A radar station utilizes fuzzy logic processing to detect atmospheric microburst events. The detection process accesses radar velocity measurements to calculate wind shear and produce a wind shear likelihood image. Reflectivity measurements are used to create a reflectivity likelihood image as an indicator of storm activity. A clutter image of stationary objects close to the radar is used to create a clutter likelihood image as an indicator of false alarms. A weighting factor is applied to each set of wind shear, reflectivity, and clutter likelihood data, and the values are combined to produce a combined likelihood image. The combined likelihood image is subjected to averaging over a fixed area to eliminate spurious readings, and positions having a sufficient size in combination with a likelihood value exceeding a minimum threshold amount are selected as microburst events.