3D PRINTING OF ELEVATION DATASETS: ASTER & SRTM DEMS, LIDAR DSMS & DTMS

Digitized elevation data for most of the Earth’s land surface are readily available for download from sites such as The National Map, CGIAR-Consortium for Spatial Information, Land Processes Distributed Active Archive Center (LP DAAC) or Open Topography, as well as other sources. Depending on the platform that collected the data, resolutions range from 90 m per pixel to < 1 m per pixel for some LiDAR data.

Software packages such as ArcGIS and Global Mapper, along with many others, are able to read these data files and produce pseudo-3D hillshaded images. Using LiDAR point cloud data, these programs can produce 3D views on a monitor. Until recently, printing has been limited to 2D prints. The advent of relatively inexpensive 3D printers with quality resolution has changed the picture.

Elevation data processed in ArcGIS, Global Mapper and Accutrans3D can be printed on an Ultimaker2 3D printer using the printer-supplied Cura software. Print times vary with the size of the 3D print and the resolution. Ultimaker2 has four native resolutions: Fast (0.2 mm), Normal (0.1 mm), High (0.06 mm) and Ultra high (0.04 mm). Using Normal (0.1 mm) resolution, most print samples take 6 - 16 hours to print, depending on the size. The printer has a maximum model volume of 223 x 223 x 205 mm (L, W, H).

Elevation data can be printed either horizontally (real world view) or vertically (within the limits of the build volume). Printing horizontally results in layering that mimics contour lines and can be useful in teaching topo map/contour concepts. This layering can mask fine detail features, however, so the most detailed prints are achieved by printing vertically rather than horizontally at the same resolution.

Scaling and vertical exaggeration are necessary for most datasets. Scaling is easily accomplished within the Cura printer software with a single button (“scale to fit build volume”) command. Most projects require a vertical exaggeration of 1.5 - 5 X to produce a useful print. This, too, is easily accommodated in the Cura software. Vertical exaggeration is very much a “season to taste” value. Unfortunately, with print times running many hours, it is impossible to use a “test print” to determine quickly what the ideal vertical exaggeration is for a given dataset.