ICESat-2 lidar bathymetry based on adaptive spatial density filtering

Objectives: The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2), equipped with the advanced topographic laser altimeter system (ATLAS), is a new measurement strategy with a photon-counting technique to obtain the high-accuracy elevation information in the surface of the Earth. ATLAS uses a 532 nm band laser to emit three pairs of scanning laser beams, each pair containing one strong and one weak sub-beam. The receiver collects millions of return photons from the echo, and each received photon is recorded. In the field of bathymetry, photon counting lidar has great advantages. For traditional lidar with linear detection regime, the only way to achieve high signal-to-noise ratio requirement is to increase the emitted laser energy, which results in a very bulky and huge system and high design power consumption. Photon counting lidar is no longer confined to the detection of clear high signal-to-noise ratio echo waveform. It uses statistical optics theory to achieve effective ranging, and maximizes the use of each photon energy in the laser echo, which can greatly improve the detection efficiency of bathymetry detection while reducing the system complexity, providing new opportunities for near-shore bathymetry detection. Based on ICESat-2 ATLAS data for bathymetry, the main problem is how to achieve automatic detection and extraction of signal photons for raw photons in different regions, different environments and different density distributions. Methods: To overcome these existing problems, an adaptive spatial filtering bathymetric method based on photon density distribution is proposed in this paper. The method first separate the photons into above the water, water-surface, and underwater photons. The filter parameters can be automatically determined by the photon density distribution generated based on the optical characteristics of different water bodies and water depth distribution, so that the signal photons of the water surface and bottom can be precisely extracted to achieve bathymetry. In order to verify the bathymetric accuracy of the method, experiments were conducted using ATLAS datasets obtained in different areas of the South China Sea, and finally the experimental results were verified by using high-precision bathymetric data acquired by airborne lidar. Results: The validation results show that the coefficient of determination between the ICESat-2 bathymetric results and the airborne lidar bathymetric results reaches 0.93, and the correspond root mean square error is 0.51 meters, which illustrate the bathymetric potential of the method proposed in this research. In addition, when using photonic data for bathymetry, it is influenced by the refraction of the water body, which mainly exists in the vertical direction and increases with the increase of depth. Conclusions: Due to the complex diversity of water body environment, the photon bathymetry accuracy will show a certain decreasing trend with the increase of water depth. In future research, bathymetric experiments will be conducted using the method of this paper and the ATLAS dataset to investigate a more accurate refraction correction model in order to explore the bathymetric potential of the satellite-based photon counting lidar in different water environments.