Integrating Optimal Terrain Representations from Public DEMs using Spaceborne LiDAR

Digital Elevation Models (DEMs) are crucial for various scientific applications such as hydrology, geomorphology, and urban planning. However, the precision and reliability of DEMs vary widely across regions due to disparities in data sources and the processing methodologies. This study proposes a novel approach for extracting the optimal parts from public DEMs by integrating the high-accuracy elevation data provided by the spaceborne Light Detection And Ranging (LiDAR) satellite, Ice, Cloud, and land Elevation Satellite-2 (ICESat-2). The method involves a systematic process of identifying the nearest LiDAR data points, assigning appropriate weights, ranking DEMs based on error statistics relative to the LiDAR data, and selecting the best segments from each DEM. This strategy ensures integrating the highest quality data from each DEM while preserving the original data’s integrity. The method was experimentally applied to the Loess Plateau in northwestern China, a region characterized by complex and erosive terrain. Using over 11 million high-precision ICESat-2 ATL08 points, optimal terrain retrieval (OTR) was performed on seven publicly available DEMs.