The topsoil is now being asked to file its own erosion report
the topsoil is now being asked to file its own erosion report: deep learning segmentation models extract erosion gullies from high resolution imagery across china's loess hills, a quarterly land cover record covering one hundred phases from 2000 to 2024 shows how water driven soil loss has been cut across the plateau, and an explainable model trained on 150,000 samples reproduces soil loss maps with near perfect fit and identifies the drivers.
the earth now files its own sediment ledger.
Context
The Loess Plateau is one of the best-known cases of large scale soil erosion control, so it is also a good test for mapping methods. The first result extracts erosion gullies from high resolution imagery with deep learning segmentation. The second is a dataset: a 25-year, quarterly land change map of the plateau at 10 metres resolution, published in Earth System Science Data, which shows how land cover has changed across 100 phases from 2000 to 2024.
The third trains an explainable model on 150,000 samples to reproduce soil loss maps and rank what drives them, which is where land managers decide what to act on.
Gully extraction is a feature-level map. Land cover change is a time series. Soil loss modelling is a risk surface. The sources are useful together because each answers a different question: where the damage is, what changed, and why.
Related work
- Erosion gully extraction in Southwest China using MSAR-Net ↗Comparable deep learning gully mapping in another region.
- Towards accurate mapping of loess waterworn gully ↗Related loess gully mapping method.
Watch next
- Linking the land cover record to measured sediment in rivers.
- Whether explainable drivers agree with long-term field plots.
Sources
- 25-year, quarterly land change maps of China's Loess Plateau (Earth System Science Data)essd.copernicus.org
- Towards accurate mapping of loess waterworn gully by integrating google earth imagery and DEM using deep learningexa.ai
- Integrating Open-Source Earth Observation, RUSLE and explainable machine learning (preprint)researchsquare.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 15:17 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 →