Assessing the Influence of DEM and DBM Accuracy on Geoid Modeling at the Land-Sea Interface: A Case Study in Eastern Taiwan

In regions with complex terrain, terrain correction is essential for accurately computing geoid undulations. The accuracy of Digital Terrain Models (DTMs), which include both terrestrial and marine datasets, has a significant impact on terrain correction and geoid modeling. One striking example is the coastal region of Hualien in eastern Taiwan, located at the land-sea boundary. This area is renowned for its dramatic topographical shifts, where elevations plummet from over 2000 meters above sea level to ocean depths exceeding 2000 meters within just a few kilometers. This study evaluates the effects of various global and regional Digital Elevation Models (DEMs) and Digital Bathymetric Models (DBMs) on geoid modeling in Hualien. The geoid modeling strategy utilizes a remove-restore approach, combining global geopotential models, local gravity observations, and high-resolution DEMs and DBMs. Particular attention is given to the accuracy of these DEMs and DBMs at the critical land-sea interface, where topographic variations are most pronounced. To validate the results, we compare them against high-resolution satellite imagery and the Global Self-consistent, Hierarchical, High-resolution Geography Database (GSHHG) coastline data. Additionally, geoid models derived from different DEM-DBM combinations are assessed using high-precision GNSS-leveling geometric geoid observations at multiple locations within the study area. The primary aim of this research is to improve terrain correction accuracy in areas with complex topography along land-sea boundaries, ultimately enhancing the precision and reliability of geoid modeling results.