OpenElevationService gives different elevation than other services

E.g. Mt. Everest @ 27.9881° N, 86.9250° E

https://www.freemaptools.com/elevation-finder.htm gives 8840m
My own implementation of SRTM lookup gives 8840m

https://openrouteservice.org/dev/#/api-docs/elevation/point/get gives 8794

curl --include \
     --header "Content-Type: application/json; charset=utf-8" \
     --header "Accept: application/json, application/geo+json, application/gpx+xml, img/png; charset=utf-8" \
  'https://api.openrouteservice.org/elevation/point?api_key=...geometry=86.9250,27.9881'

Similarly for other points I have tried. Surely there is something wrong with my usage but I can’t seem to find the problem.

Any pointers appreciated!

Hi @ThomasDaheim,

I can’t determine the data resolution in that app, as there is no scale.
But as far as i know the openelevationservice is using CGIAR processed SRTM data with 90m resolution.

I think i remember SRTM providing higher resolution data in the meantime.
Your issue might be related to the resolution of the dataset.

There won’t be an update of the elevation service anytime soon.
But it’s open source if you want to take a look https://github.com/GIScience/openelevationservice

Best regards

Hi amandus,

in my own app I’m using srtm3 data - that is also the 90m resolution. And still I get the value of 8840m. So I don’t think its related to the resolution.

What are the coordinates that you’re using for testing? Maybe I could crosscheck on my end. I’m using the following “known” coordinates to cover the four hemispheres

Mt. Everest: 86.9250, 27.9881 / 27° 59’ 9.8340"N 86° 55’ 21.4428"W - 8848m
Mt. Rainier: -121.7603, 46.8523 / 46° 52’ 47.8812"N 121° 43’ 36.8616"W - 4392m
Mt. Cook: 170.1418, -43.5950 / 43°35’26.81"S, 170°08’16.65" E - 3724m
Acongagua: -70.0109, -32.6532 / 32° 39’ 11.4444"S, 70° 0’ 39.1104"W - 6908m

My SRTM3 dataset yields

Mt. Everest: 8840m
Mt. Rainier: 4369m
Mt. Cook: 3712m
Acongagua: 6929m

Using the OpenElevation I get:

Mt. Everest: 8794m
Mt. Rainier: 4370m
Mt. Cook: 3711m
Acongagua: 6868m

So the difference is probably due to a different SRTM3 dataset. Is yours the one https://cgiarcsi.community/data/srtm-90m-digital-elevation-database-v4-1/ ?

Hi @ThomasDaheim,

According to the OES code it should be the the dataset from your link (SRTM CGIAR v4.1).

I was basically clicking random points around mt. everest in your app :slight_smile: to see when the height changes.

If it is the same resolution the difference is probably due to the CGIAR post processing of the SRTM data. I think the problem with the original SRTM data were a lot of data “holes”.
We deemed continous data more important than exact height.
Of course the difference is quite significant sometimes, but as the resolution is 90m you won’t get exact data for every point anyways.

Hi @amandus,

I’ll give that dataset a try to compare it against the data I’m currently using (mainly from http://viewfinderpanoramas.org/dem3.html).