Generation and validation of cartosat-1 DEM for northern Aravali range of hillocks, Rajasthan, India

Abstract

Digital Elevation Model (DEM) is a spatial representation of any area or surface in the form of three dimensions (x,y,z) represent the height of the individual points (pixels) of the surface along with x and y rectangular coordinates.DEM is widely used to extract the information from many areas like subsidence/deformation monitoring, updating and preparation of drainage pattern, hydrology and water resource management etc. The selection of DEM (on the basis of accuracy) plays an important role while using it for any particular application. In some application area accuracy of DEM is not a matter of great concern. However, in some application, accuracy of DEM plays a very vital role to derive the required result.. In this study, area of interest belongs to a mineral rich zone, coming under the Khetri copper belt, a part of Northern Aravali range of hillocks in India. This particular belt is predominant with mining activities since late 1960's. Keeping in view of deformation studies in such area using conventional DInSAR technique, accuracy of DEM (normal baseline > 200 m), should be in higher side to eliminate the local topographic phase completely. Keeping in view of above background, DEMs (relative and absolute) were generated from high resolution Cartosat-1 stereo pair dataset (IRS) using GCPs, without GCPs and GCPs collected with the help of Google Earth. RMS error obtained from different techniques were compared along with SRTM DEM (3 arc second).Vertical accuracy in terms of RMS error has been found in the order of 2.54m, 10.10m, 37.97m and 76.66m for Carosat-1 DEM (GCP collected from GNSS), SRTM, Cartosat-1DEM (without GCP) and Cartosat-1 DEM (GCP collected from Google Earth) respectively. Validation of high accuracy DEM has been demonstrated in DInSAR processing (using ALOSPALSARSLC dataset) to eliminate topographical phase of the study area.