Elevation data plays a crucial role in the work we do here at Ag-Informatics. One dataset that we often turn to is the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM). Some background on the mission:
“SRTM flew aboard the Space Shuttle Endeavour in February 2000, mapping Earth’s topography between 56 degrees south and 60 degrees north of the equator. During the 11-day mission, SRTM used an imaging radar to map the surface of Earth numerous times from different perspectives. The combination of these radar data were processed at JPL to produce a global topographic map created by bouncing radar signals off Earth’s surface and back to the shuttle.”
As the name suggests, a DEM is a representation of the Earth’s surface. DEMs are produced as gridded data where each cell in the grid is an aggregation of an area on the ground and the value of the cell is the elevation of that area. These data are often produced and manipulated in GeoTIFF format such that each pixel in the image corresponds to a grid cell. These files are essentially a regular TIFF image with additional metadata describing the spatial extent of the data such that they can be positioned on a map.
SRTM was collected at a resolution of 30 meters, meaning every pixel in the dataset represents a 30m by 30m area. However, since its release, the 30m product has only been openly available for the US, with the rest of the global dataset having restricted availability. A 90m product was made available for the rest of the world, and although downsampled to a lesser resolution, this product is still immensely popular and is one of the highest quality elevation sources available.
For our products and analyses focused on the US, we are fortunate to have excellent data freely available to us in the National Elevation Dataset (NED), which is generally available at 10m resolution and at 3m resolution in some areas. When working in some countries or regions it may not be possible to obtain high quality openly available data. SRTM enables analyses in these areas to be conducted from an excellent and consistent data source. It is invaluable when working on a global scale or with an area of interest that encompasses multiple countries. In fact, in some cases may be the only source available!
With that in mind, we were elated when the White House announced that the complete global 30m SRTM data would now be released over the next year. You can read that announcement here, under the heading “Releasing Powerful New Data to Enable Planning for Resilience”:
The sheer number of additional pixels is staggering. SRTM data is downloaded in tiles covering an area of 1 degree longitude by 1 degree latitude. Each 90m tile is 1201-by-1201 for a total of 1,442,401 pixels. The same region covered by a 30m tile would be 3601-by-3601 for a total of 12,967,201 pixels. The number of pixels increased by a factor of 9! The complete SRTM dataset is 14,546 tiles, meaning the 90m data is 20,981,164,946 pixels in total, and the 30m data totals over 188 billion pixels, 188,620,905,746 to be exact!
Take a look at this example of a 30m pixel and a 90m pixel in relation to the field at Memorial Stadium here in Champaign, IL. Visualizing the difference in resolution in this way allows you to see just how much potential variation is being aggregated into each pixel, and what a difference the increase can make. Now just imagine the difference a 10m or 3m pixel would make!
We are thrilled to put this data to work in creating new products and improving our existing products and analyses. If you want to talk SRTM or anything else, shoot me an email, firstname.lastname@example.org.