Advances in Volcanic Cloud Satellite Remote Sensing

Operational volcanic ash detection techniques used at the various Volcanic Ash Advisory Centers (VAACs) are generally qualitative and require manual analysis. Reliable, and automated, satellite-based ash detection techniques are few and far between due to the difficult nature of separating volcanic clouds from meteorological clouds and other non-volcanic features on a global basis. As such, globally applicable automated volcanic ash detection requires a combination of advanced spectral and spatial techniques. The spectral sensitivity to volcanic ash is improved by utilizing effective absorption optical depth ratios in lieu of brightness temperature differences. Further, volcanic ash clouds are composed of pixels of varying spectral uniqueness. A cloud object based ash detection approach allows pixels, within an ash cloud, that have the strongest spectral signature (e.g. the most spectrally unique pixels) to be used to detect the entire ash cloud. It will be shown that cloud object based volcanic ash detection can be used to issue accurate automated ash cloud alerts to VAAC meteorologists. In addition, an objective estimate of the cloud height, effective particle size, and mass loading is needed to help forecast ash cloud dispersion. These parameters can be retrieved using infrared radiances and volcanic ash microphysical models. We will present results from a fully automated, and globally applicable, optimal estimation technique used to retrieve these important parameters. All of the techniques described in this talk are utilized in an automated ash detection, retrieval, and alert system currently being transitioned to NOAA operations.