The kelp forest is now being asked to file its own recovery map
the kelp forest is now being asked to file its own recovery map: ucla runs deep learning over satellite imagery to draw statewide california kelp maps that catch changes old tools missed after the 2014 2016 heatwave, a vision transformer learns eelgrass presence from 8,300 annotated underwater images, and fujitsu's ocean twin identifies seagrass beds in turbid water at over 85 percent accuracy.
the meadow now files its own blue carbon census.
Context
The January 6, 2026 kelp paper uses a VGG16-U-Net on Planet Dove imagery to create California-wide annual canopy maps at 3 m from 2017 to 2024. The study reports 89.59% overall test accuracy and 86.4% for kelp, with wave and seawater confusion addressed through temporal filtering. UCLA’s March 18, 2026 account connects the maps to monitoring persistence after the 2014-to-2016 marine heatwave.
SeagrassFinder builds more than 8,300 annotated underwater images and tests several architectures for eelgrass presence or absence. Vision Transformers exceed 0.95 AUROC on the final test set. Coverage estimation is described as preliminary.
Fujitsu’s Ocean Digital Twin page claims image restoration in turbid seawater and identification of seaweed/seagrass species and coverage above 85% accuracy. That percentage is vendor-reported; the page does not provide an independent test protocol.
The kelp maps measure floating canopy after the heatwave, not all submerged kelp biomass. Near-daily satellite acquisition is not proof that every published map is a real-time product. AUROC above 0.95 is not 95% classification accuracy. Underwater eelgrass classification, satellite kelp canopy mapping and Fujitsu’s ecosystem recognition are distinct tasks. None alone establishes carbon-credit quantity or permanence.
Watch next
- Seagrass transfer tests in new waters and seasons. Independent Fujitsu class-level results under measured turbidity. Error propagation from mapped area to blue-carbon estimates.
Sources
- High-resolution Planet Dove data identify local drivers of kelp canopy persistencepreview-www.nature.com
- How AI deep learning is helping UCLA scientists protect California’s coastal ecosystemsnewsroom.ucla.edu
- SeagrassFinder: Deep Learning for Eelgrass Detection and Coverage Estimation in the Wildarxiv.org
- Ocean Digital Twin (Fujitsu Research Portal)en-documents.research.global.fujitsu.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 06:03 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 →