Accuracy Assessment of Google Earth Imagery and SRTM DEM for Topographic Mapping: A Case Study of Aba North, Nigeria

Abstract

Accurate geospatial data is fundamental to reliable decision making for sustainable physical development on earth surface. Improvements in satellite technology have facilitated capturing of geometrically precise images of the earth’s surface. In this study evaluation of the horizontal and vertical accuracies of google earth satellite imagery and SRTM of 90m resolution has been carried out in absolute and relative senses to investigate their reliability for topographic mapping requiring high level of accuracy. Detail and contour maps of the study site were produced using ground data captured with DGPS and digital level respectively.  Accuracy of the satellite datasets were investigated visually by comparing its derived maps with these ground-based mapping products and statistically using root mean square as evaluation metrics. Software used are ArcGIS10.5 and sufer11. The statistical evaluation was carried out using 36 well dispersed and randomly selected real world planimetric control points and 50 vertical ground control point (GCP) which has been harmonized to the same horizontal and vertical datum with the corresponding values extracted from the satellite datasets. The relative comparison revealed that the detail map from the satellite data represented the ground data very well however, some similarities were observed between the SRTM and ground-based contour maps, but the level of conformity was not good enough for engineering projects requiring high accuracy. Absolute accuracy assessment in Northing, Easting, 2D plane and height yielded root mean square error values of 0.039, 0.046, 0.060 and 1.787 respectively. Correlation coefficient revealed values of 999997857, 0.999999354 and 0.5471. In eastings, northings, and height. These values attest that google earth imagery is effective for detail mapping requiring high accuracy based on 4.0 RMSE mapping accuracy recommended by American Society for photogrammetry and remote sensing. SRTM also produced good results but requires further refinement to render it fit for topographic mapping that demands level of accuracy required for such engineering projects.