Crowdsourced DEM from GPS and smartphone sensors


I´m interested in creating a digital elevation model from available open source elevation data collected from peoples GPSs and smartphones (preferably the ones with a built in barometric altimeter).
I posted tris tread explaining the idea in the OpenStreetMap Help forum :


, and someone suggested that I contact this working group as this may interest you guys.

As I explain in the post, my gut feeling is that this has the potential of being much more precise than the current free data available.

What do people think about it?

Data sources: any database of crowsourced public data on XYZ coordinates. The two sources I could find were:

any other suggestions on data sources?

I’m an economist, without much programming skills and I´m new to OSM, so I need help to get the data and understand the peculiarities of the gps elevation data.
Specifically the following items:

  1. download the “point cloud” with XYZ coordinates, just for a first look at the data, the points coverage, etc.
  2. scrap all GPX files from OSM, and some descriptive information (like the user who submitted the data), etc.
  3. Read the GPX files and transform them into a “square”, dataset, like csv, etc.
  4. Se if it is possible to estimate the precision of the elevation measurement from the GPX data. Perhaps if the metadata includes the brand and model of the device it will be possible to know if the GPS has a built in barometer or not.

Once I get the data will try construct a digital elevation model, or more precisely a digital terrain model, from it, using some statistical procedures.

Lucas Mation