Can ASTER-data be used for bathymetric mapping of coral reefs in the Red Sea using digital photogrammetry?

Coral reefs are main contributors to the national income of many developing countries. As they are experiencing more and more stress due to inappropriate human actions and global change, they need to be studied in order to monitor them and preserve them for the future. Bathymetric mapping of coral reefs has always experienced difficulties. Due to their rather shallow depth and/or often remote location in the middle of the ocean, they are not always suitable for standard bathymetric sounding campaigns. Here remote sensing can bring a solution. Different methods have been developed to derive bathymetric information out of remote sensing data. Some methods are based on information gathered by active sensors, e.g. LIDAR; others are based on the different attenuation coefficients of sequencing wavelengths detected by multispectral passive sensors. The aim of this research paper is to examine the possibilities to map bathy-metry of coral reefs using digital photogrammetric processing of stereoscopic ASTER-images. Through-water photogrammetry has already been used for subaqueous mapping, for example by Westaway et al. (2001) and Butler et al. (2002) in close-range river bed surveys using mounted photogrammetric cameras. Due to the refraction of the light on the water surface, calculated depths tend to be underestimated in comparison with the true depth. Westaway et al. (2001) have developed a refraction correction algorithm to deal with this problem. The ASTER stereoscopic data have already proven its usefulness in generating terrestrial DEM’s using the digital photogrammetrical software package VIRTUOZO 3.2. Due to the fact that stereoscopic ASTER - images are only made in the infrared, its application in bathymetric mapping is limited to the upper meters of the reefs. As study area the coral reefs surrounding Big Giftun Island off shore Hurghada (Egypt), are selected. First a DEM is generated in the conventional way. The strict use of absolute ground control points that lay under the water surface, has been proven to be beneficial for the accuracy of the depth estimations. Unfortunately, due to refraction at the air/water interface, an additional shift in X and Y then occurs. Some errors are also occurring after match editing which are distorting the calculated depths. Afterwards the correction algorithm for the two media problem developed by Westaway et al. (2001) has been applied. Although the first, preliminary result presented here, is still subject to some errors, general trends in reef topography can already be noticed. It can be concluded that it is possible to estimate depth of the coral reefs using digital photogrammetry and ASTER data. Further investigation is required for the need of additional adaptation of the refraction correction algorithm to use with ASTER data and the remaining problems with geometric accuracy and the match editing procedure