[gdal-dev] Proper algorithms for computing slopes for global/polar DEMs?

Michael Patrick geodesy99 at gmail.com
Sun Feb 22 19:43:00 PST 2015


(This may not be target subject matter for the list)

>
> We're trying to figure out a good algorithm for computing topographic
> slope for Alaska, which is both huge and near the pole.  Our DEM is
> currently in geographic coordinates, but we need slope over a fixed
> distance (change in elevation vs. horizontal meters).  The issue is, there
> is no projection to switch this to to get an accurate set of slope
> calculations.  Of course with degrees, the X and Y distances change over
> the globe -- so the horizontal distances will be variables from one window
> to the next.  I have seen various solutions published, but I wanted to see
> if someone knows of a decent coded-up algorithm for this.  Thoughts?
>

You don't mention specifically what dataset your DEM source is, but read
carefully the accuracy statements in the documentation, and the counter
answers on the Stack Exchange article.  For instance, if you are using
GTOPO30 , see Section 6.1 Data Sources of the Readme
<https://lta.cr.usgs.gov/sites/default/files/GTOPO30_README.doc>

You would probably better off using one of the original sources, rather
than calculate on top of the already considerable distortion. Also there
are DEM collections available for various areas of interest, like the
Stereo-Photogrammetric
Digital Elevation/Surface Model <http://www.pgc.umn.edu/elevation/stereo>
from the Polar Geospatial Center. You don't mention where the AOI or what
scale you are interested, but there are a considerable number of others out
there ( http://nsidc.org/data/dems/datasets.html )

Michael Patrick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20150222/1481c97d/attachment.html>


More information about the gdal-dev mailing list