Quality Assessment of TanDEM-X DEM 12m Using GNSS-RTK and Airborne IFSAR DEM: A Case Study of Tuba Island, Langkawi

Abstract

Digital elevation models (DEMs) have been recognized as a primary spatial dataset and essential for numerous scientific applications. The advent of TerraSAR-X for digital elevation measurement (TanDEM-X) has opened a new potential to obtain an accurate DEM. Nowadays, the demand/use of TanDEM-X DEM in scientific applications has become increasingly popular as it offers an alternative to the widely used DEMs: ASTER and SRTM DEM. Although many researches have been conducted to assess the performance of the TanDEM-X DEM at different locations in the world, however, only several multi-regional studies have been performed in Malaysian region. Currently, there are two types of DEMs published by DLR i.e., non-open access (12m and 30m resolution) and open access (90m resolution). In this article, the accuracy of TanDEM-X 12m has been comprehensively and systematically evaluated using 1284 GNSS-RTK control points over Tuba Island and airborne IFSAR-DEM as a reference height. Besides, four available global DEMs: TanDEM-X 90m, AW3D30 DEM, SRTM DEM, and ASTER DEM have also been evaluated to identify the accuracy of TanDEM-X DEM 12m. Based on the evaluation using GNSS-RTK points, TanDEM-X 12m exhibits the highest accuracy with an RMSE of ±1.553m. Unexpectedly, AW3D30 DEM shows a better performance compared to TanDEM-X 90m with RMSE of ±1.964m, followed by SRTM DEM with RMSE of ±3.296m Meanwhile, ASTER DEM exhibits the lowest accuracy with RMSE of ±4.100m The comparison of TanDEM-X 12m and the well-known DEM, SRTM DEM with airborne IFSAR-DEM shows the opposite results. Based on the topographic profile at flat and forest area, the SRTM-DEM exhibits better accuracy than TanDEM-X 12m