The reef is now being asked to file its own health certificate
the reef is now being asked to file its own health certificate: an explainable cnn called f net reads 7,384 underwater images and separates bleached from healthy coral at 97.6 percent accuracy, a vision framework with a temporal module forecasts bleaching from the features it already learned, and a physics informed network rebuilds depth resolved thermal fields across four great barrier reef sites from sparse temperature loggers.
the coral now files its own stress report.
Context
F-Net is an explainable CNN for bleached versus healthy coral. On 7,384 coral images it reaches 97.6 percent accuracy, with Grad-CAM style maps showing what it looks at.
The depth-resolved thermal work uses a physics-informed network on satellite sea surface temperature and sparse loggers. It was validated across four Great Barrier Reef sites in 30 holdout experiments, with 0.25 to 1.38 C RMSE at unseen depths.
F-Net classifies what a coral looks like now. The thermal network reconstructs the temperatures it is exposed to at depth. One reads damage, the other reads the cause.
Related work
- NOAA PIFSC coral bleaching dataset (Hugging Face) ↗Public bleaching imagery.
- Cross-dataset coral classification study (Stanford CS231n) ↗Tests how well coral classifiers transfer.
- PINN reef thermal code ↗Code for the thermal model.
Watch next
- How F-Net accuracy holds on other datasets.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 10:32 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 →