Capabilities/limitations of landsat-tm data for bathymetric analysis of shallow marine environment: A case study

Abstract

The intensity of the light received by Landsat satellite sensors is inversely proportional to the path length though the sea water column. Since the light passes down to the sea floor and back again, the intensity should be related to the depth by an inverse square law. In practice, this simple relation is unlikely to be true due to the effect of several factors, each of which may distort the bathymetric analysis. Among these factors : sensor noise, atmospheric interference; vegetation effect; variations in sea-surface conditions; water absorbency and seabed reflectivity. This work aims at extracting accurate depth information from Landsat-TM data, which is confused by these variations causing some limitations. The approach adopted in this work comprises correction of source error within the Landsat-TM data sets, and a trial to distinguish between areas of variable depth against intensity relationship. To demonstrate this approach, various techniques of digital processing of Landsat-TM imagery data have been applied for bathymetric mapping of shallow water area in the Arabian Gulf, to the north of Qatar Peninsula. It has been corroborated with ground truth information form various data sets including echo sounder date; British Admiralty charts; topographic maps of the coastal area; meteorological and tidal state reports; and other ancillary data. This study shows that atmospheric and vegetation correction could improve the usefulness of Land-sat TM data bethmetric

analysis. Also, processing of Landsat TM band 1 and 2 and their ratio technique is found suitable for analysis of water depth particularly in shallow marine environment.