The pacific is now being asked to file its own enso forecast
the pacific is now being asked to file its own enso forecast: a geospatial transformer reaches skillful predictions sixteen months out from spring by coupling the tropical basins, a glass box dynamical deep learning model extends that to nineteen months and hindcasts the 2015 super el nino, and a cmip6 trained cnn finds historical forcing lifts enso predictability by 14 percent.
the ocean now files its own phase.
Context
The geospatial transformer is GL-Geoformer, from Science Advances, Tropical basin interactions reduce spring predictability barrier of ENSO in a deep learning model. The abstract says incorporating tropical basin interactions substantially reduces the spring predictability barrier, giving skillful ENSO predictions up to 16 months ahead when initiated in spring, with pacemaker experiments on the Indian Ocean Dipole and Atlantic Nino.
The glass-box model is in npj Climate and Atmospheric Science (2026), built with dynamical system deep learning, a transparent multivariate ENSO model with skillful predictions up to 19 months ahead that hindcasts the onset, intensity and decay of the 2015-2016 super El Nino more than a year in advance. The CNN study is in Science Advances: with a leave-one-out strategy on CMIP6 historical and preindustrial control runs, ENSO predictability is enhanced by 14.0 plus or minus 1.8% under historical anthropogenic forcing.
All three numbers match their abstracts. They are three separate studies on different data and different skill definitions, so 16 and 19 months are not a like-for-like ranking of the two models. The 14 percent is a change in predictability measured in climate model simulations, not a gain in forecast skill for real events. The note's phrase 'more than a year' for the 2015 hindcast comes from the abstract excerpt, and the excerpt read is cut off at that point.
Watch next
- Skill metric and threshold behind each lead time. Whether the 19-month result holds for other events.
Sources
- Tropical basin interactions reduce spring predictability barrier of ENSO in a deep learning model (Science Advances)science.org
- Skillful and interpretable ENSO prediction via a glass-box dynamical-deep learning model (npj Climate and Atmospheric Science)nature.com
- Deep learning reveals enhanced ENSO predictability under historical anthropogenic forcing (Science Advances)science.org
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 09:19 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 →