The tornado is now being asked to give its own heads up
the tornado is now being asked to give its own heads up: mit lincoln lab trains models on the tornet radar set for 5, 10 and 15 minute lead times before a funnel forms, a frequency aware network on dual polarization radar pushes strong case detection past 40 minutes, and ncar runs daily ai hazard forecasts for tornadoes hail and wind a week out.
the vortex now files its own warning.
Context
MIT Lincoln Laboratory’s Intelligent Tornado Prediction Engine page confirms models trained on TorNet storm images before tornadogenesis at 5-, 10- and 15-minute lead times. TorNet contains more than 200,000 radar images and is intended as an open research benchmark.
NCAR’s experimental medium-range convective-hazard system uses two AI numerical-weather-prediction emulators to generate 100 ensemble members. NCAR reports preliminary skill improvements up to a week ahead. Its example map gives the probability of at least one tornado, hail or wind report within 40 kilometres of a point over a 24-hour period.
A benchmark prediction at a chosen lead time is not a guarantee of warning every tornado that far in advance. MIT describes potential decision support, not an autonomous operational replacement for forecasters. NCAR’s week-ahead regional hazard probabilities are different from predicting an individual funnel’s location and formation time.
A title for a dual-polarization enhanced detection network was located, but the primary paper and the specific greater-than-40-minute strong-case result were not verified. The frequency-aware and timing claims remain unsupported here, not refuted. NCAR’s page confirms real-time availability but does not establish the note’s exact daily update cadence.
Watch next
- Primary paper and event selection behind the greater-than-40-minute claim. Operational false alarms, probability calibration and severe-weather skill by forecast day.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 October 2026 at 07:07 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 →