The National Wetlands Inventory (NWI) produces information on the characteristics, extent, and status of the Nation’s wetlands and deepwater habitats. Even though the NWI system of detecting and mapping wetlands has over 90% confident interval, it suffers of an omission error that varies from 30% to 85%, depending on the location. In addition, NWI still uses a stereoscopic view of aerial photographs for manually digitizing wetlands. The result is high costs in map production. A third problem related to NWI is the interval incurring between map updating. The 10 year average interval greatly affects the estimate of gains and losses of wetlands.The purpose of this study is to develop a classification tree model (CART) able to discriminate among hydrophytic vegetation. The target will be achieved using satellite images from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). A multi-temporal analysis of the images and the fine ground resolution will provide information about the vegetation types. In addition, GIS data layers, such as digital soil maps and elevation data will be added to the model in order to improve the accuracy of the classification. The model will provide a highly efficient low cost method to determine wetland vegetation, and it may be used to decrease the omission error of NWI. In addition, the continuous availability of satellite imagery will shorten the time range in updating maps, bringing it to one-year interval.