The Shuttle Radar Topography Mission (SRTM) (Wikipedia article) is a NASA mission conducted in 2000 to obtain elevation data for most of the world. It is the current dataset of choice for digital elevation model data (DEM) since it has a fairly high resolution (about 90 meters at the equator, and <30 meters in the U.S.), has near-global coverage (from 56° S to 60° N), and is released into the public domain.
Many OpenStreetMap-based projects use SRTM data to provide topography information, relief shading, and elevation profiles for trails and routes. An example is the OpenCycleMap rendering which shows Contours and relief shading derived from SRTM data.
The data is freely available for download from the NASA website http://www2.jpl.nasa.gov/srtm/ . It is available in two formats: SRTM3 has a resolution of three arc-seconds, or roughly 90 meters, and is available worldwide. SRTM1 has a resolution of one arc-second, or roughly 30 meters, but is only available for the United States. Two versions of the data were released: version 1 was close to the raw data collected by the shuttle and is considered research quality. Version 2 has undergone some postprocessing to smooth portions of the data and clip it to coastlines.
The data still has numerous small voids and a few large holes, but is mostly complete for its areas of coverage.
Several projects have postprocessed data that fills in data voids and improves accuracy in places:
- http://srtm.csi.cgiar.org - non-commercial only, therefore cannot be used with OSM data (since the cc-by-sa license can't have extra non-commercial restrictions added).
- http://www.viewfinderpanoramas.org/dem3.html (can be used in open source projects according to a note here)
- Well, according to the commenter (http://igorbrejc.net/openstreetmap/viewfinders-dem-data-comparison-with-srtm#comment-36413), the original data is licensed by a Russian company, so I'm not sure it can be used --Breki 20:08, 20 January 2010 (UTC)
Since 25 May 2011 the SRTM data by the German Aerospace Center ("Deutsches Zentrum für Luft- und Raumfahrt") is available for free for scientific use only. It's more accurate than the data collected by NASA but covers only 40% of the planet.
SRTM data has been used in many maps and projects and can also be used in conjunction with OSM data.
The data is divided into one degree square tiles. The outermost rows and columns of each tile overlap with the corresponding rows and columns of adjacent tiles.
The tiles are distributed as zip files containing .hgt files. The .hgt files have a very simple format: a series of 16-bit integers giving the height of each cell in meters. SRTM3 tiles have 144201 integers representing a 1201x1201 grid, while SRTM1 tiles have 12967201 integers representing a 3601x3601 grid.
Recent versions of GDAL support the .hgt files natively (as long as you don't rename the files; the names they come with are the source of their georeferencing), but the srtm_generate_hdr.sh script can also be used to create a GeoTIFF from the .hgt.zip files. (Note that the script has SRTM3 values hardcoded; if you're using SRTM1, you'll have to change the number of rows and columns to 3601, the number of row bytes to 7202, and the pixel dimensions to 0.000277777777778.)
Data in OSM format (XML)
The Hochschule für Technik Stuttgart has converted contours calculated from SRTM data to OSM format (XML) and has it ready for download at 
The lines are having the tags contour=elevation + contour_ext=elevation_minor every 40m + contour_ext=elevation_major every 1000m + ele=nn (elevation_medium is not used)
You can also do a manual conversion to OSM format using:
- Srtm2Osm to write contours as OSM ways
- Phyghtmap to write contours as OSM ways
- Groundtruth (see below) to write contours as OSM ways
other Tools that are using SRTM data:
- Srtm2wayinfo to calculate the altitude differences along each way
- Srtm to Nodes as a plugin for Osmosis to add a height tag to each node
- OSM-3D - uses preprocessed OSM SRTM
- OpenDEM - collects reliefs and started with SRTM
- OSM2World - reads raw SRTM to improve OSM-based 3D scenes