Validation of DEMs for two-pass SAR differential interferometry — A case study in Wangfeng coal mine, Jiaozuo city

Abstract

The two-pass method is widely used in the SAR differential interferometry (D-InSAR). It requires the external digital elevation model (DEM) as topographic phase signature to be removed from the interferogram. In the past decades, significant advance in global elevation modeling have been made with the release of the SRTM DEM and ASTER GDEM at no cost. These DEMs are often used for two-pass method, but their influence on the quality of the interferometry remains poorly understood. Taking the Wangfeng coal mine in Jiaozuo city as test site, this paper evaluate the validation of the latest release of SRTM DEM (v4.1), ASTER GDEM and 1∶50000 topographic map-derived DEM for two-pass D-InSAR measurement. The height accuracy, the precision of co-registration and the precision of D-InSAR measurement are qualified using ERS SAR images and leveling data from 2002 to 2005. The results demonstrated that the ASTER GDEM has lower height accuracy and lower precision of co-registration with SAR images, causing higher error in two-pass D-InSAR measurement. While the ASTER GDEM has an unprecedented horizontal resolution of ∼30 m, systematic errors present in this research-grade version of the ASTER GDEM will impede its use for two-pass D-InSAR.