Hyperspectral remote sensing of salt marsh vegetation, morphology and soil topography

The present paper deals with the relationship between vegetation patterns and salt marsh morphology in the Venice lagoon and with the use of remote sensing to infer salt marsh morphologic characteristics from vegetation mapping. Field measurements indicate that salt marsh vegetation species (halophytes) are reliable indicators of ground elevation and live within typical elevation ranges characterised by standard deviations of less than 5 cm. A model is then developed which uses vegetation as a morphological indicator of soil topography to estimate ground elevation from fractional cover values of each vegetation type. The use of data from an airborne remote hyperspectral sensor is presented as a means of discriminating between different salt marsh vegetation communities. Vegetation maps obtained from unmixing techniques have then been used to produce digital elevation maps (DEM) of salt marsh areas. The DEM based on halophytes cover estimates and extracted from high spatial and spectral resolution data allows a high estimation accuracy, with an error standard deviation of a few centimetres in the considered study area within the Venice lagoon. The accuracy and resolution attainable through this method are comparable and often superior to those obtained through state of the art laser altimetry. © 2003 Elsevier Science Ltd. All rights reserved.