Aurora Forecasts Explained: Kp, Bz and What NOAA Actually Predicts

Auroras happen when charged particles from the Sun hit Earth's magnetic field and rain into the upper atmosphere, making oxygen and nitrogen glow. Forecasts estimate how strong that particle flow is likely to be, and where on Earth the glow will be pushed.

Every number in an aurora forecast is an estimate with error bars. The honest way to use it: space weather tells you if a storm is likely; your local sky tells you if you will see anything.

The Kp index is a 0–9 planetary measure of geomagnetic disturbance, compiled from ground magnetometer stations. It is the headline number in most aurora forecasts.

The practical scale:

• Kp 0–2 — quiet; aurora stays near the polar regions.
• Kp 3–4 — active; aurora can dip toward mid-latitudes on the poleward side.
• Kp 5 (G1 storm) — minor storm; aurora visible from southern New Zealand and the northern US/Scandinavia.
• Kp 6 (G2) — moderate; aurora reaches further toward populated latitudes.
• Kp 7 (G3) — strong; aurora photographed from Wellington, southern Australia, northern Europe.
• Kp 8–9 (G4–G5) — severe to extreme; rare, and potentially visible from very low latitudes.

NOAA uses the G-scale (G1–G5) for storm alerts, and it maps roughly onto Kp: G1 ≈ Kp 5, G2 ≈ Kp 6, G3 ≈ Kp 7, G4 ≈ Kp 8–9, G5 ≈ Kp 9. Kp is a proxy, not a local measurement — your actual visibility depends on your geomagnetic latitude, the storm's structure and your local sky.

The Bz value is the north–south component of the interplanetary magnetic field (IMF) carried by the solar wind. It is the single most useful number after Kp, because it controls whether the solar wind can couple with Earth's field.

• Bz negative (southward) — the solar wind's field points opposite to Earth's, magnetic reconnection happens, energy flows in, and auroras intensify. This is what you want to see.
• Bz positive (northward) — the fields align, coupling weakens, and even a fast solar wind may produce little aurora.

A forecast that quotes only Kp is incomplete. Watch the Bz trend too: a strongly negative Bz that persists for hours is a much better signal than a Kp number that spikes for one three-hour window.

Two more numbers fill out the picture:

• Solar wind speed (km/s) — typical values run 300–500 km/s. Sustained speeds above 600 km/s often signal a fast stream from a coronal hole, which can produce several days of unsettled conditions.
• Density and arrival time — a coronal mass ejection (CME) arrives hours after it is launched, and forecasters can only estimate the arrival window. The classic failure mode: a big CME is predicted, the window passes, and nothing happens — or a smaller event arrives early.

This is why every forecast carries uncertainty. NOAA SWPC publishes watch, warning and alert products for exactly this reason: watches are days out, warnings hours out, alerts when activity is actually detected at the ACE or DSCOVR spacecraft about an hour upstream of Earth.

The US NOAA Space Weather Prediction Center (SWPC) is the reference source most aurora tools — including Darkest Hour — draw from. The products you will see:

• 3-day Kp forecast — a rolling estimate of Kp for the next three days, updated a few times daily. Good for choosing a night this week.
• 27-day outlook — a rough seasonal look based on solar rotation. Useful for spotting a likely active week, useless for tonight's decision.
• Geomagnetic storm watches and warnings — watches for expected CME arrivals, warnings when a storm threshold is crossed.
• Aurora oval maps — where the aurora is likely to be visible right now, based on current activity.

SWPC is excellent at when the Sun is active; it is deliberately cautious about whether you will see it — that part depends on your cloud, moon and horizon.

Here is where most people get burned: they see a G3 storm warning, drive to a dark site, and see nothing — because the sky was cloudy, the moon was up, or the storm was pushed north/south of their latitude.

The complete decision needs four conditions at your exact site:

1. Geomagnetic activity — Kp high enough for your latitude, ideally with a negative Bz.
2. Cloud cover — the sky must be clear. This is the most common killer.
3. Darkness — the moon must be down or low. A full moon washes out all but the strongest aurora.
4. Horizon — you need an unobstructed view toward the pole (south in New Zealand, north in the northern hemisphere).

Darkest Hour's aurora forecast page puts Kp, cloud, moon and darkness on one screen so you can see the whole decision at once — and the tonight check turns it into a Go / Worth checking / Wait verdict for your coordinates.

The workflow that actually works:

Days out: watch the SWPC 3-day Kp forecast and the 27-day outlook. If an active window is coming, pick your site and plan your moonless night — the week around new moon is your baseline.

A day out: confirm the storm watch is still current. Check the expected cloud for your site. If both look plausible, commit to the drive.

The night of: check the current Kp and Bz trend, then check your local sky — cloud, moon phase and rise/set, darkness — right before you leave. Re-check when you arrive.

Expectation setting: most nights with an active forecast produce a subtle glow, not a curtain show. That is normal. The bright, structured displays you see in photos are the exception, not the rule.

When you want to compare a few candidate sites or scan the week for the best aurora-observing night, the Pro planner ranks nearby sites against current provider forecasts and finds your clear windows. See Pro pricing — and for the difference between northern and southern lights, read aurora australis vs aurora borealis.