Permafrost – ground that has been permanently frozen for two or more years – makes up a considerable amount of the Earth, around 15% of the Northern Hemisphere.
Permafrost is important for our climate and contains large amounts of biomass stored as methane and carbon dioxide, making tundra soil a carbon sink.
As global warming heats the Earth and causes soil to thaw, the permafrost carbon cycle is expected to accelerate and release soil-contained greenhouse gases into the atmosphere, creating a feedback loop that will contribute to climate change.
Over the past two decades, much of the Arctic has been mapped with great precision by commercial satellites.
With funding and support from the U.S. National Science Foundation as part of the “Navigating the New Arctic” program, a team was given access to archives of over 1 million image scenes taken in the Arctic.
One of the most distinctive features of permafrost are ice wedges, which produce recognizable polygons in satellite imagery.
“The ice wedges form from the freezing and melting of soil in the tundra,” said researcher Anna Liljedahl. “Some of them are tens of thousands of years old.”
Starting in 2018, researcher Chandi Witharana began using neural networks to detect polygons from thousands of Arctic satellite images. To do so, Witharana and his team first had to label 50,000 individual polygons, hand drawing their outlines and classifying them as either low-centered or high-centered.
Training the model with the labeled images, they fed the satellite imagery into a neural network and tested it on unlabeled data. After a few initall challenges, the team was seeing accuracy rates between 80 and 90%.
The next important phase of the research project will come when the researchers analyze satellite imagery from different years and times of year. Comparing the state of the ice-wedge polygons can show trends such as how fast the landscape is changing.
https://phys.org/news/2022-02-arctic-permafrost-satellites-supercomputers-deep.html