The dynamics of eruption clouds will be analyzed and modeled using satellite images, atmospheric wind fields, and the Puff tracking model. Satellite images will be used as the control element to which model predictions will be compared. The influence of wind fields, physical variables and eruption conditions will be evaluated and compared to improve the accuracy and the types of information that can be derived from tracking models. Experiments will be conducted using a variety of wind fields as input to the Puff tracking model to evaluate their effect on predicting the location and shape of the clouds. The sensitivity of variables internal to the tracking model, such as dispersion and grain size, and a variety of eruption conditions will also be evaluated. The analyses will use selected data of historical volcanic eruptions where satellite and gridded wind field data are available (1980 - present), that exemplify a variety of eruption and environmental conditions. In addition, the capability of the new hyperspectral satellite data will be evaluated and compared to presently available data to improve the characterization of eruption clouds and to further "tune" the models. We anticipate that this project will provide insight into the elements that control or influence the movement and prediction of eruption clouds and will improve existing tracking models.