ABSTRACT
This chapter discusses the newly developed technique to correct lidar- Digital Elevation Model (DEM) in more land cover types including sawgrass, tall mangroves, short mangroves, and coastal prairies in the coastal Everglades. The high point density and low vertical accuracy of lidar data reported by commercial vendors makes it especially attractive for low-relief coastal wetland habitats. Southwest of the tall mangroves site hurricanes resulted in a patchwork of coastal prairies dominated by three species: saltwort, glasswort, and saltgrass. In terms of the Mean Absolute Error and Root Mean Squared Error, there is no significant difference between the Empirical Bayesian Kriging-bias correction and Minimum Object-Based Bin (MOBB) lidar-DEMs for coastal prairies and short mangroves. The MOBB and object-based machine learning approaches increase efficiency in accurate and precise DEM generation and should be considered in future applications that require reliable elevation data in the Everglades.
