Skip to main content

Mapping the Arctic with Delta

An aerial photo of the Arctic Tundra wetlands. In this photo, half of the wetlands are covered in snow.

The word permafrost can make this land type sound eternal. And the word does mean what it sounds like: land that is permanently frozen, even in summer months. According to the National Snow and Ice Data Center (NSIDC), permafrost is under a quarter of the land in the northern hemisphere. To be considered permafrost, the land needs to be frozen for at least two years. Much of our permafrost has existed in a frozen state for thousands of years. But climate change has made a lot of land that was once permafrost thaw out. This can create a number of cascading effects that researchers are trying to figure out.

One such team of scientists is using GPU-based supercomputer, Delta, at the National Center for Supercomputing Applications (NCSA) to help them study how thawing permafrost affects structures. Building on top of permafrost used to be relatively safe. But as the permafrost layers thaw, the ground begins to shift. Buildings and roads that are built on this ground may be in danger of collapsing. To prepare for the worst, governments need to be aware of how permafrost will change the land underneath the infrastructure. But there’s another problem to tackle as well – there is no comprehensive survey of all the structures on permafrost. Chandi Witharana, an assistant professor of natural resources and the environment at the University of Connecticut, is working with a team to create a dataset of all the structures built atop permafrost so that researchers have something more comprehensive to work from when studying permafrost thaw. However, 25% of land in the northern hemisphere is a vast area to survey for images. Cataloging all that imagery would be a monumental task. That’s where Delta comes in.

“Understanding the risks to infrastructure in ice-rich areas due to thawing permafrost is crucial for adapting to climate change in the Arctic; it’s also important no matter what the climate is doing because infrastructure increases risks to permafrost thaw,” Witharana said. “But to do this, we need detailed maps of buildings and other structures, which are often lacking, so our study used ACCESS allocations on Delta to make mapping efforts less costly and more efficient.”

Witharana is familiar with Delta’s capabilities, having worked with the Woodall Climate Research Center’s project tracking permafrost thaw. For this research, Witharana’s team began the process of developing a foundational artificial intelligence (AI) model capable of identifying human-built structures from high-resolution satellite imagery of the Arctic. Witharana’s team published a paper on their work titled A multi-objective comparison of CNN architectures in Arctic human-built infrastructure mapping from sub-meter resolution satellite imagery in the International Journal of Remote Sensing.

“We trained and tested close to 100 different deep-learning models to gather an adequate number of samples for statistical testing, much of which occurred simultaneously as we could submit multiple jobs at once,” explained Elias Manos, a graduate student at Connecticut and lead author on the paper. “Without ACCESS allocations on the Delta supercomputer at NCSA, this work would likely not have been possible.”

“Arctic imagery work like this is exactly why advanced research computing and data resources and service available via ACCESS are important assets to the U.S. research community,” said John Towns, NCSA’s deputy director and principal investigator for the ACCESS Coordination Office. “We are excited to see where this work goes and encourage collaborations with researchers as they address their big data challenges.”

If you’d like to know more, read the original story on the ACCESS website here: Using Deep Learning to Understand Permafrost Thaw


NCSA combines next-generation processor architectures and NVIDIA graphics processors with forward-looking user interfaces and file systems to create Delta, a powerful computing and data analysis resource that is part of the national cyberinfrastructure ecosystem through ACCESS. The project partners with the Science Gateways Community Institute to empower broad communities of researchers to easily access Delta and with the University of Illinois Division of Disability Resources &

Back to top