Using Satellite Based Techniques to Combine Volcanic Ash Detection Methods

Volcanic ash poses a serious threat to aircraft avionics due to the corrosive nature of the silicate particles. Aircraft encounters with ash have resulted in millions of dollars in damage and loss of power to aircraft engines. Accurate detection of volcanic ash for the purpose of avoiding these hazardous areas is of the utmost importance to ensure aviation safety as well as to minimize economic loss. Satellite-based detection of volcanic ash has been used extensively to warn the aviation community of its presence through the use of multi-band detection algorithms. However, these algorithms are generally used individually rather than in combination and require the intervention of a human analyst. Automation of the detection and warning of the presence of volcanic ash for the aviation community is a long term goal of the Federal Aviation Administration Oceanic Weather Product Development Team. We are exploring the use of data fusion techniques within a fuzzy logic framework to perform a weighted combination of several multi-band detection algorithms. Our purpose is to improve the overall performance of volcanic ash detection and to test whether automation is feasible. Our initial focus is on deep, stratospheric eruptions.