An integral part of the Land Use Projects work involves the use of high-resolution biophysical data about the land surface. Central to this is topographic data, which controls many environmental processes from climate to biology and to soils. In the past, our research has used global elevation datasets such as the GTOPO30 from the United States Geological Survey (USGS), or digitized cartographic maps and produced high-resolution digital elevation models (DEMs) for more specific regions. The release of the Shuttle Radar Topographic Mission (SRTM) from NASA heralds a new age in the way in which we do research. We can now work anywhere in the globe and have base topographic data without the need for lengthy digitalization. Predictive models which use terrain attributes can also be applied globally, and is no longer restricted to the coverage of internally produced DEMs.

The SRTM data will cover the entire globe with a 3 arc second (approx. 90m) digital elevation model. A 1 arc second data product also exists, but is not available on general release. The data is currently available for small regions for download from the National Map Seamless Data Distribution System, or is available for the entire globe from an ftp site. The vertical error is said to be less than 16m.

Pre-Processing of SRTM 90m DEM

The data currently being distributed by NASA contains no data holes where water or heavy shadow prevented the quantification of elevation. These are generally small holes, which nevertheless render the data less useful, especially in fields of hydrological modeling.

The Land Use project, in collaboration with Robert Hijmans (University of Berkeley) and Andy Nelson (University of Leeds), have imported the SRTM data, and performed some processing to fill in these no data. This process involved the production of contours, and the re-interpolation of these derived contours back into a DEM. We then used the interpolated DEM values to fill in the original no data holes from the crude SRTM data. This was done using Arc/Info and an AML script which is available for download. Questions regarding the use of this AML should be directed to Andy Jarvis.

Data Availability

[Click to enlarge]

The hole-filled SRTM data is available for download from the Consortium of Spatial Data (CSI) website, found at the CGIAR SRTM Data website. Anyone interested in further information should contact Andy Jarvis.

Evaluation of SRTM 90m DEM data

Project scientists are already using this DEM data. We have performed an analysis of data quality, comparing in detail 90m SRTM data with a DEM produced from digitizing a 1:10,000 cartographic map of the region of Dapa in Valle del Cauca, Colombia. Using contours alone, we produced a 25m high resolution DEM, and compared this in absolute and relative terms with the SRTM data product for the same region. In absolute terms the two DEMs compare very closely, but it is evident that the 1:10,000 cartographic map still produces greater topographic detail. The ridges and valleys are more pronounced, and in some cases small streams are identified in the cartographically produced DEM which are not recognizable in the SRTM data. Some preliminary hydrological modeling is made to highlight this, showing that the topographic index (TOPMODEL) differs somewhat between DEMs, and fails to pick out some small streams in the SRTM data.

The results of this evaluation are available in detail in a Powerpoint presentation entitled "Comparison of SRTM derived DEM vs. topographic map derived DEM in the region of Dapa, Valle del Cauca, Colombia". The associated data is also available in a zip file with an ArcView project file for easy access.