The wetland is now being asked to keep its own water ledger
the wetland is now being asked to keep its own water ledger: a model trained on annotated imagery from wetlands all over the globe watches coastal and inland marshes change, a segment anything variant maps wetlands from sparse point labels across seasons, and weakly supervised nets refreshed minnesota's statewide wetland inventory without hand drawn labels.
the fen now files its own extent report.
Context
Swamp-AI (February 13, 2026) trains wetland segmentation on geographically diverse annotated imagery. Its three test locations yield reported average overall accuracy of 93.7% and intersection-over-union of 74.6%. The task is wetland versus non-wetland, not a full wetland-type inventory.
WetSAM uses Sentinel-2 image time series, temporal adapters to Segment Anything, sparse point labels and region-grown pseudo-labels. Its study manually annotates approximately 2,000 to 4,000 points per area and tests four classes across eight areas.
The Minnesota inventory paper’s author-institution record describes weak labels extracted from existing land-cover products. Change detection selects stable pixels and SNIC objects refine the labels before training U-Net++ and DeepLabV3+. The abstract reports average F1 of 91.3% across seven classes for U-Net++.
Sparse or weak labels do not mean no human annotation. WetSAM manually labels points; Minnesota reuses previously produced thematic maps rather than creating dense labels from scratch. Swamp-AI’s globally varied training and three-location testing do not establish performance everywhere. The Minnesota evidence read here is the author-institution abstract, not the full publisher methods. These studies support mapping approaches, not proof of a completed government inventory adoption.
Watch next
- Full Minnesota evaluation protocol and operational adoption record. Change-detection tests on new years and unseen wetlands. Class-specific omission errors beyond overall accuracy.
Sources
- Swamp-AI: a deep learning model for monitoring wetlands change across the globenature.com
- Wetland mapping from sparse annotations with satellite image time series and temporal-aware segment anything modelarxiv.org
- Cost-effective statewide wetland inventory update using weakly supervised deep learning: A case study in Minnesota, USA (SUNY author-institution record)researchconnect.suny.edu
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 06:33 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 →