Uncertainties in Digital Elevation Models from Global Advanced Land Observing Satellite (ALOS) digital topography data in the Indian subcontinent

摘要

The study of the Earth's topographic surface using digital topography is an important component of many problems in the Earth Sciences. The 30-m Shuttle Radar Topography Mission (SRTM) and the Advanced Land Observing Satellite (ALOS) global digital topography datasets have been the most recently available global digital topography datasets. Most users directly download the data/DEMs in their native form assuming a vertical error close to their mission goals. We demonstrate, through the use of 221 dual-frequency static Global Positioning System (GPS) independent ground control points, that the datasets typically have outliers and voids that increase the error by an order of magnitude. Filtering the voids, outliers, and correcting systematic errors significantly improved the vertical accuracy of the datasets. We concluded from our study that the ALOS data generated the most accurate DEM in the Indian subcontinent. This finding needs to be tested and confirmed worldwide for generating the best DEMs.