Wetlands play an important role in the global ecosystem as they provide many ecosystem services and habitat for wildlife. The United States developed the National Coastal Wetlands Inventory (NCWI) to provide information on the spatial extent of the coastal wetlands. However, the NCWI has become outdated and needs to be updated for the purpose of natural resource management. Monitoring invasive species is also essential because they threaten wetland health. The overall objective of this research is to develop accurate and replicable tools for monitoring coastal wetlands in a time- and cost-efficient manner. An integrated approach using the National Wetlands Inventory (NWI), multitemporal Landsat satellite imagery, Soil Survey Geographic Database (SSURGO), National Hydrography Dataset (NHD), Light Detection And Ranging (lidar), and Advanced Synthetic Aperture Radar (ASAR) data will be used to map wetlands. NWI, Landsat, SSURGO, and NHD data can effectively identify most wetlands, but they sometimes miss vegetated wetlands. Lidar and radar data will be incorporated to reduce these omission errors and improve the overall classification accuracy of wetland delineations. Maximum likelihood, a decision tree analysis, or object-based classification will be applied to the various combinations of data. The least expensive method of wetland delineation with the highest accuracy will be determined and proposed for national implementation. Mixture-tuned matched filtering on hyperspectral data will be used to map common reed (Phragmites australis, Cav.) an invasive species on the U.S. East Coast. Results of this research will provide remote sensing strategies for updating coastal wetlands inventories every 5 to 10 years with improved classification accuracy. These strategies are more efficient than the traditional approaches of aerial image interpretation and field surveys. The wetlands inventory data from this research will provide the information needed by natural resource managers to protect and conserve wetlands.
|Original language||English (US)|
|Number of pages||11|
|Journal||Photogrammetric Engineering and Remote Sensing|
|State||Published - Jun 1 2012|