The permafrost is now being asked to keep its own slump ledger
the permafrost is now being asked to keep its own slump ledger: a 2026 arctic challenge asks teams to map retrogressive thaw slumps from multimodal satellite data, a landsat time series method found 3,273 slumps on the qinghai tibet plateau while cutting very high resolution imagery needs by 90 percent, and a hybrid model ingests 62.71 million field measurements with 3.3 billion satellite observations to track frozen ground carbon.
the tundra now files its own thaw report.
Context
Retrogressive thaw slumps are places where thawing ground collapses and keeps retreating, and they release carbon that was frozen. Mapping them by hand needs very high resolution imagery, which is costly and patchy. The 2026 GeoAI Arctic Challenge turns this into a shared benchmark with public data, starter files and a submission format.
The Qinghai-Tibet Plateau result uses a Landsat time series to find 3,273 slumps while cutting the need for very high resolution imagery by 90 percent, according to the paper. The third result is about carbon: the zero-curtain, the stretch where soil sits near 0 C while water freezes or thaws, matters because permafrost soils hold roughly 1,700 billion tonnes of organic carbon.
Competitions and open benchmarks make methods comparable. A Landsat-first method is cheaper to run repeatedly than one built on commercial imagery, so it suits long term monitoring.
Related work
- A Geo-Foundation Framework for Retrogressive Thaw Slump Detection (ISPRS Archives) ↗Foundation model approach to the same task.
- A multi-scale vision transformer multimodal GeoAI model for mapping Arctic permafrost features ↗Related multimodal model.
Watch next
- Challenge results, once teams report them on the shared test set.
- Whether slump maps made from Landsat alone transfer from the Tibetan Plateau to the Arctic.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 14:16 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →