3D Mapping Doubles Estimate of Permafrost Melt's Infrastructure Cost to $261 Billion

A new study offers insight into the economic impact of melting permafrost — ground that stays below freezing for at least two consecutive years and covers about 24% of the Northern Hemisphere, including large areas of Alaska, Canada and Siberia — on Arctic communities. As global warming melts permafrost, it releases stored greenhouse gases and can severely damage infrastructure in communities built on it. The study, published in the journal Earth's Future, was led by Elias Manos, a doctoral candidate, and Chandi Witharana, an assistant professor, both in the Department of Natural Resources and the Environment, alongside collaborators at George Washington University and the Woodwell Climate Research Center.

Earlier estimates considered only buildings' two-dimensional footprint, significantly underestimating damage and rebuilding costs — a particular problem for Siberian communities, where the Soviet Union built many multistory apartment buildings to support natural resource extraction. Accounting for buildings in three dimensions, the researchers estimate that permafrost melt will cost $261 billion in infrastructure damage by the middle of the 21st century, more than double the previous estimate of $110 billion, mainly because rebuilding multistory buildings costs far more than rebuilding single-story ones.

To calculate the estimate, the researchers combined remote sensing data with artificial intelligence, training a deep learning model on a 10-terabyte set of satellite images to identify buildings, and a second model to classify them as residential or nonresidential. A separate elevation dataset called ArcticDEM let them calculate each building's height, and building codes from the U.S., Canada and Russia helped estimate total floorspace for residential buildings; the team could not calculate this for nonresidential buildings because of too much variation between them.

The researchers then coupled these building estimates with a model of permafrost degradation developed by co-author Dmitry Streletskiy of George Washington University. The study did not account for demographic trends that determine which buildings remain in use, a direction for future research. The findings have been made available to Arctic communities through the Permafrost Discovery Gateway, an international multi-institution effort, and the researchers say the same method could be extended to industrial buildings as well.