Modelling topography with SAR interferometry: illustrations of a favourable and less favourable environment

Abstract

Radar remote sensing has faced an increasing interest in a wide variety of Earth observation studies during this last decade. This interest is notably related to the necessity to map areas where the cloud cover does not allow the use of classical optical sensors, or the increasing interest for ocean studies. Amongst the large number of radar applications different techniques exist for mapping the topography. Digital elevation models (DEMs) are required in a growing field of applications and especially in earth sciences where their use in structural geology, environmental geology or engineer geology has become essential. In this paper, we discuss the results obtained with SAR interferometry (InSAR) from two different environments and for two specific applications. The first is the study of the recent and present day tectonic evolution of the Rukwa rift (SW Tanzania) and the other, the study of the topography from the Bulusan volcano (Philippines). In the first situation, interferometry gave encouraging results. The computed InSAR DEM helped to identify unsuspected lineaments closely related to tectonic. In the Philippines, SAR interferometry failed to provide DEM because of the temporal decorrelation due to the dense vegetation and the humid climate. Radargrammetry or radar stereoscopy was tested with few different geometries, but the obtained results were evidencing the need in accurate ground control points. Discrepancies with regard to the reference computed from existing 1/50,000 topographic maps are reported to be larger where no or too few GCPs were identified.