Volcanic ash clouds can drift for many hundreds to thousands of miles down wind from a volcanic eruption. They are difficult to detect by pilots either visually or on their weather radar. The Aviation Weather Center (AWC) is tasked with issuing International Sigmets for volcanic ash in New York and Oakland Flight Information Regions (FIR). These Sigmets are used by the airlines to route traffic away from the drifting ash regions. Once the AWC has been notified of a volcanic eruption, satellite images are used to determine the horizontal extent and trajectory of the volcanic ash cloud. However the detection of the ash cloud is frequently a problem. For relatively dry ash clouds, the GOES band 4-5 difference will show the ash, but for low ash clouds or ash clouds with moisture in them, this channel difference is not always reliable. The GOES band 2-4 difference will show the ash cloud because of the particle shape differences, but there are problems with varying solar angles and day/night differences for the band 2 which makes this product difficult to interpret for all times of the day or night.
This paper will discuss a new four channel volcanic ash satellite image processing algorithm which has been developed at the AWC. The visible channel is used to normalize the band 2 data during the day to remove effect of changing sun angles. The band 4-5 difference is used in conjunction with the normalized band 2-4 difference to generate a product which enhances the volcanic ash cloud as much as possible while minimizing the water clouds.