Abstract: In response to the problems of large operational difficulties, high price and low accuracy of tidal flat topography monitoring, based on ICESat-2 satellite laser point cloud profile data, this paper proposes a digital elevation model inversion method combining multivariate optical images. The two major sandbar systems of the Tiaozini and Gaoni in the Yellow Sea of China are used as research objects. Using low-pass filtering techniques to extract topographic height values and combine them with tidal water level data to assign values, the conversion from 2D to 3D is achieved, and improves the inversion accuracy of the current mainstream waterline tide level assignment method. The model shows more detail of the topographic variation of the tidal flat surface, with significant differences in height, allowing the presence of clear tidal trench topography to be observed. Comparing the measured data, the correlation coefficient is 0.89 with a root mean square error of 0.34 m. The results of this study demonstrate the significance of ICESat-2 data for the inversion of beach topography and lay the data foundation for the construction of the theory of tidal flats evolution under the influence of human activities.

Key words: ICESat-2, tidal flat, waterlines, DEM