My notes from reading 'The earth isn't flat'.

12 February 2021

Not really a paper but a blog post by Loxcel Geomatics which shows some great comparisons of global elevation datasets, including some of the more recent ones.

• ASTER has significant noise and artefacts.
• NASADEM is a small improvement on SRTM.
• The AW3D30 and Copernicus datasets are much higher quality.

My focus for gpxz will be on merging datasets of high resolutions, as that’s where my experience lies. But I’m still going to want a global base layer, and the more I read the more I suspect that SRTM won’t be up to the task.

I’ve also seen a few papers like High-quality seamless DEM generation blending SRTM-1, ASTER GDEM v2 and ICESat/GLAS observations using ICESat observations as ground truthing, could be helpful for coregestering datasets.