Skip to main content

Blue Waters processes final installment of ArcticDEM mapping initiative

The Arctic has historically been one of the most remote and under mapped regions on Earth. Two years ago, the Obama Administration issued an executive order to create better digital elevation models for an area that spans almost twice the size of the continental United States. The data would be used to provide greater context for the monitoring ice caps and assist with a wide array of decision-making. Through partnerships with the National Science Foundation and the National Geospatial Intelligence Agency (NGA), ArcticDEM was born and this month it has released the sixth and final installment of some of the most accurate digital elevation data for 32% if the Arctic.

Paul Morin, Director of the Polar Geospatial Institute at the University of Minnesota, delivered on that promise with the help of the Blue Waters supercomputer at the National Center for Supercomputing Applications at the University of Illinois.

“Blue Waters was particularly suitable for this project not only because of its size, but also because of how it’s managed,” said Morin, “This collection of HPC resources, both human and hardware are nimble enough that they can be called on a moment’s notice to address an impossible problem like this.”

In order to get this extremely accurate elevation data, commercial satellites from NGA orbited above the Arctic capturing high-resolution images. Blue Waters then processes these enormous data sets, using Surface Extraction from TIN-based Searchspace Minimization (SETSM) software, developed at Ohio State University, and produces stereoscopic images of the Earth. To date, ArcticDEM has produced high resolution terrain data for 97.4% of the Arctic, covering over 70.5 million square kilometers, using Blue Waters. These DEMs are publicly accessible at

Awards and other uses

The ArcticDEM Project was recently awarded the HPC User Innovation Award by Hyperion Research for their response to a need for high quality elevation data in remote locations, the availability of technology to process big data, and the need for accurate measurement of topographic change. However, the technology of ArcticDEM and Blue Waters extends far beyond the Arctic. “Elevation models in the past have often been produced within the bounds of a single country. What we’re doing is ignoring national boundaries. We’re in one of the most remote places on Earth, and we can construct a dataset for a very large region with high resolution that’s publicly available,” said Morin.

That data is used to predict sea level rise, coastal erosion, national security, civil engineering, and aircraft safety, and provide more information about natural disasters.

In June, a colossal landslide triggered a tsunami near the Northwestern village of Nuugaatsiaq, Greenland leaving several homes washed into the ocean and four casualties in its wake. “It was very useful for us to make a map, with better height information just a day or two after the landslide happened, so it gave the emergency response and everyone involved to get a clearer geographic information,” said Eva Mätzler, project manager of remote sensing for Asiaq.

The landslide is believed to be over 3,000 feet long and 900 feet wide, and was large enough to falsely register as a 4.0 magnitude earthquake on the seismometer. It was initially unclear what caused the landslide, but researchers have since discovered melting permafrost and ice sheets that threaten unstable slopes and a significant risks of future tsunamis if the cliffs fail.

The final release of ArcticDEM is expected to be announced in Spring 2018, and will focus on updating information of previous releases, including an improved filtering methodology that was used for Release 6.

Disclaimer: Due to changes in website systems, we've adjusted archived content to fit the present-day site and the articles will not appear in their original published format. Formatting, header information, photographs and other illustrations are not available in archived articles.

Back to top