Thomas R Allen, Old Dominion University, 20170228, Marsh classification raster dataset for South Atlantic Landscape Conservation Cooperative: .
Summary
Salt marshes classification of the South Atlantic Landscape Conservation Cooperative geography covers the northern Outer Banks (and extreme southeastern Virginia, Back Bay area) south through NC, SC, and Georgia to approximately Sapelo Island. The marsh classification is derived from Landsat 8 OLI imagery acquired in May 14-19, 2014. This georeferenced imagery was atmospherically corrected, mosaicked, and water masked prior to deriving a set of three Normalize Difference Indices (NDX) bands: Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI) and Normalized Difference Soil Index (NDSI). Prospective salt marshes and associated tidal non-forested wetlands were classified using object-oriented image [...]
Summary
Salt marshes classification of the South Atlantic Landscape Conservation Cooperative geography covers the northern Outer Banks (and extreme southeastern Virginia, Back Bay area) south through NC, SC, and Georgia to approximately Sapelo Island. The marsh classification is derived from Landsat 8 OLI imagery acquired in May 14-19, 2014. This georeferenced imagery was atmospherically corrected, mosaicked, and water masked prior to deriving a set of three Normalize Difference Indices (NDX) bands: Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI) and Normalized Difference Soil Index (NDSI). Prospective salt marshes and associated tidal non-forested wetlands were classified using object-oriented image analysis (OBIA) techniques and Monteverdi Orfeo Toolbox software (v.1.22x). The region was subdivided into several zones within states and classified by image segmentation, mean shift clustering, and class labeling. Training sites for the classification focused on Intensive Study Areas (ISAs) spread across the region where higher-resolution imagery, ancillary wetland and land cover, and access were highest were used for training and subsequent accuracy assessment. Following classification and editing, the vector polygons were remerged into a an Esri vector layer and packaged for shipment. In addition to this vector polygon dataset, a raster grid and mosaic source Landsat NDX imagery are available.
The marsh classification was conducted to inventory the extent and characteristics of salt marshes across the SALCC landscape, with the goal of including improved marsh thematic classification zones to enhance ecological management and conservation and assess future vulnerability to sea level rise.