Evaluation of a Statistical Approach for Extracting Shallow Water Bathymetry Signals from ICESat-2 ATL03 Photon Data

In this study we present and validate a simple empirical method to obtain bathymetry profiles using the geolocated photon data from Ice, Cloud, Land Elevation Satellite-2 (ICESat-2) mission, which was launched by NASA in September 2018. The satellite carries Advanced Topographic Laser Altimeter System (ATLAS), is lidar that can detect single photons calculate their bounce point positions. ATLAS uses green laser, causing some of penetrate air–water interface. Under right conditions shallow waters (<40 m), these are reflected back after interaction with ocean bottom. Using ICESat-2 four different overflights above Heron Reef, Australia, comparison SDB showed median absolute deviation approximately 18 cm Root Mean Square Errors (RMSEs) down 28 cm. Crossovers between two Reef difference 13 For an area north-west Sisimiut, Greenland, done multibeam echo sounding data, RMSEs 35 cm, correspondingly deviations 33 49 proposed works well under good clear such as Great Barrier Reef; however, for more difficult areas advanced machine learning technique should be investigated order find automated distinguish other signals noise.