Remote Sensing of Coastal Ecosystems Using Spectral Indices

Abstract

Spectral indices are algorithms performed to improve the signal of certain features, such as vegetation, water and soil in satellite images. The objective of this work was to utilize the normalized difference vegetation index (NDVI), the modified normalized difference water index (MNDWI) and the ratio index for bright soil (RIBS) along with band compositing techniques in order to map and delineate the extent of the coastal ecosystems along the coasts of Oman, in terms of mangrove vegetation, wetlands, sabkhas and coral reefs, respectively. Satellite data were acquired from the Landsat-8 Operational Land Imager (OLI) during 2018. Some oceanographic characteristics: tidal range, sea surface temperatures (SST) and the depth of the sea floor of Oman offshore region were also utilized to interpret the spatial extent of these coastal ecosystems. Results showed that the applied indices were efficient to highlight 14 locations of mangroves, 19 locations of wetlands, 2 locations of sabkha and 15 locations of coral reefs. It is observed that mangroves and wetlands are much related to high tidal range coasts, whereas coral reefs are contingent to shallow off-shores with SST of 22-30°C. These corals occur either along the main coast or adjacent to the islands of the country. Sabkha and salt marshes occur along extended coastal flats of low-lying sandy coasts. The present study proved that the spectral indices are good surrogates to map coastal ecosystems.