# How do I know the vertical/horizontal units of my DEM? I'm creating a slope using gdaldem

I'm using `gdaldem` to create a slope map.

I can apply a scale. The default is 1.

The notes say:

``````Ratio of vertical units to horizontal.
If horizontal unit of source DEM is degrees (e.g WGS84 projection), use scale=111120
if the vertical units are meters (or scale=370400 if they are in feet)
``````

My DEM is in NAD27 UTM 18N and in the projection properties it says linear units Meters and angular units degrees.

How do I know what the vertical units are?

Your horizontal units are metres (= linear units). You can not derive your z-value units (vertical units) from the coordinate system. It's like checking the coordinate system properties to try and work out if a temperature raster z-values are in centigrade, kelvin or fahrenheit. You need to go back to the metadata, if there is any.

You can probably assume your DEM z-values are metres, but I've been caught with that before when I didn't check the range of the values/metadata and they turned out to be in centimetres...

• Is there any other way to find out without metadata? Any tools that can read this information? Sep 9, 2013 at 4:15
• @GISer Where did you get the DEM from? Sep 9, 2013 at 5:21
• I don't really remember. It's just sample data that I found somewhere. Sep 9, 2013 at 12:23
• @GISer Then no. Not unless there is metadata contained within the file, in which case `gdalinfo` might be able to read it. However, if your horizontal (X, Y) units are metres, then in all likelihood, your z values are in metres. Assuming metres, does the range of values make sense? i.e. max < 8850 (height of Mt Everest)... Sep 9, 2013 at 23:36
• Sometimes the metadata don't exist. With a DEM, though, life is easy: find an alternative source of information about the elevations and cross-check them. The alternative needn't be detailed; it only has to give you a rough sense of what the elevations are. E.g., if your DEM shows a peak in the Canadian Rockies and its values range up to 4000, then obviously it's in meters and not feet, whereas if it has values above 5000 up to 13000 or so then it must be in feet. Units in centimeters, decimeters, and tenths or hundredths of feet will be equally obvious in most cases. Dec 11, 2013 at 14:59

I've been trying to figure out why the NED DEMs don't generate a good slope histogram by themselves. Thing is, those DEMs are in NAD83 (I think?) but basically are in a geographic coord system (spherical-ish surface, degrees lat and lon), not a projected coord system (2D surface, meters north and east). That means that a meter at the north end of the DEM is not the same as a meter at the south end of the map - basically you need to project it / save it with a projected coordinate system before you can use a single number like 111120 to convert xy units to z units. But, if you've projected, you're not going to need 111120, you're going to need to use z-factor of feet-per-meter or meter-to-feet as needed or ideally 1.0 if horizontal units and vertical units are the same.

Anyway sounds like your DEM is already projected so that whole sentence about 111120 or 370400 (both of which are inaccurate fudge factors anyway since you need to project it first) aren't relevant.

Note, I'm new at this and have been trying to figure it out for a few days, and got a good answer at this thread and tracked down some more info from there - BUT - please take this with a grain of salt as I still need to verify the slope measurements vs a good ol' USGS topo map.

UPDATE: see the much more clear answers here