Abstract
The US National Elevation Dataset is used in order to assess the vertical accuracy of the SRTM‐1 DEM in the Humboldt Range (Nevada State of USA). Firstly the upland region, in which the error (elevation difference) is maximised, is delineated by density slicing of the slope image combined with size filtering of the segmented objects. Then major terrain sub‐classes are defined in the uplands region on the basis of both aspect and slope. Statistics indicate that the error is both geographic direction dependent (elevation is underestimated towards W, NW and N and over‐estimated towards E, SE and S), and linearly correlated to the terrain slope (the steeper the slope, the greater the /error/). Elementary terrain objects (aspect regions) are defined and parametrically represented on the basis of their geomorphometry and overall accuracy. A terrain object classification scheme and subsequent mapping prove that aspect is the key feature determining the SRTM accuracy.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call