Live from NOAA’s Space Weather Prediction Center, refreshed every minute in your browser.
Table view
| Time (UTC) | Bz (nT) | Orientation |
|---|
Data: NOAA Space Weather Prediction Center, fetched live in your
browser and refreshed each minute. The G-scale above is NOAA's call, not mine — they are the
authoritative body for space weather alerts, and if it ever disagrees with something here, believe them.
The coupling field Ey is a derived indicator I compute from the raw L1 feed
(Ey ≈ V × |Bz| when Bz is southward). It is shown because it is the physically meaningful
quantity and it moves before the official products do — but it is not an alert and carries no
authority. For actual warnings, subscribe to
SWPC's own alert service.
What you’re looking at
Bz is the one to watch. The interplanetary magnetic field arrives at Earth with some orientation, and when it points south — negative on the chart — it’s anti-parallel to Earth’s own field at the dayside. The two reconnect, the solar wind couples into the magnetosphere, and energy pours in. Northward Bz at the same wind speed does comparatively little. This is why “a huge CME is coming” isn’t the whole story: a fast CME with northward Bz is a light show, and a moderate one with strongly southward Bz can wreck your afternoon.
Ey combines the two things that matter into the quantity that actually drives the coupling:
Ey [mV/m] ≈ V [km/s] × |Bz| [nT] × 0.001 (southward Bz only)
Quiet is under 1. A decent storm is 2–3. Severe is 10 or more. Carrington, reconstructed, is somewhere near 100.
Dst measures the ring current — how much the storm has actually deflected Earth’s field. It’s the scoreboard rather than the forecast. Typical quiet is around −20 nT. The March 1989 storm that collapsed Hydro-Québec’s grid in ninety seconds hit about −589. Carrington is estimated between −850 and −1750.
What a Carrington event would actually do
There’s a persistent myth that a solar superstorm fries your electronics, and it leads people to buy the wrong thing. Geomagnetic storms induce voltage in proportion to conductor length:
induced EMF ≈ E_ground × conductor_length
Even in a severe event, the ground-level electric field is on the order of a few volts per kilometre. Run that through the geometry:
| Conductor | Length | Induced EMF | Result |
|---|---|---|---|
| Your laptop | ~1 m | millivolts | nothing |
| House wiring | ~100 m | volts | nothing |
| Transmission line | 100s of km | kilovolts | this is what breaks |
The 1859 Carrington event destroyed telegraph networks because they were thousands of kilometres of wire. Quebec in 1989 lost a grid through hundreds of kilometres of transmission line. In neither case did anything small and unplugged come to harm, and there is no recorded instance of a solar storm damaging consumer electronics.
The damage path is: geomagnetically induced current flows into transformer neutrals → drives the core into half-cycle saturation → harmonics and waste heat → relays trip, or windings cook. High voltage transformers are bespoke, largely foreign-built, and slow to replace. The threat is a grid outage measured in months, not a dead phone.
Faraday cages are for nuclear EMP’s E1 pulse — roughly 50 kV/m with a nanosecond rise time, which does couple into short conductors. A CME has no E1 component. Shielding your laptop against a solar storm is insurance against a mechanism that isn’t there. Backups, an off-site copy, and enough power to run one machine are the real answers.
“Are we overdue?”
No — and the reasoning behind the question is the interesting part.
Superstorms aren’t periodic. A ~150-year return period means roughly a 1-in-150 chance per year, every year; it does not accumulate while you wait. The sun keeps no ledger and doesn’t know 167 years have passed since 1859. “We’re overdue” is the gambler’s fallacy in a lab coat.
That said, there is a real reason for elevated concern right now, just not that one: we’re near the peak of Solar Cycle 25, and solar maximum genuinely carries higher risk than solar minimum. Right conclusion, wrong reasoning.
One more wrinkle worth knowing: recent work in Nature argues that the long-assumed saturation of geomagnetic storms — the comforting idea that the magnetosphere stops responding proportionally once driving gets extreme — may be a statistical artifact. Because the solar wind is measured at L1 and propagated to Earth, the mismeasurement grows with the extremity of the driver, so extreme values regress toward the mean and the response only looks like it flattens. Correct the bias and the coupling stays linear. If that holds, there’s no natural ceiling saving us, and the worst case is worse than the models said. It makes the grid problem bigger. It still doesn’t make your laptop couple to a kilometre-scale field.
Data courtesy of the NOAA Space Weather Prediction Center, a US government service in the public domain. This page relays their data and their alerts; it is not an alerting service and has no authority. For real warnings, use SWPC’s own Product Subscription Service — see also their alerts, watches and warnings explainer.