Towards a New Global Digital Elevation Model

The SRTM dataset provides the first near-global elevation model of the Earth's continental land surface at 3 '' resolution. This unique dataset is currently used for a range of applications. However, there are a number of known quirks and omissions in these data, such as voids over inland water and erroneous heights from forest canopy; in addition a recently completed global analysis of these data utilising a dataset derived from multi-mission satellite radar altimetry has revealed spatially correlated error signatures possibly caused by incorrect heights in ground control points used to seed the solution (Berry et al., Remote Sensing of Environment, 106(1): 17-27). This paper presents the initial results from work currently underway towards an enhanced global elevation dataset, ACE2, a follow-on to the highly successful ACE GDEM. This development utilises over 67 million height measurements derived from multi-mission satellite radar altimetry using an expert system approach to validate and augment the SRTM dataset to derive a full global GDEM. One of the key results is that in many regions the SRTM dataset is more accurate than its stated vertical precision of +/ - 16 m.