AuroraIs 2027 a good year to see the northern lights?
For northern Norway, Sweden, Finland, Iceland, Alaska or northern Canada: yes. Those places sit under the auroral oval, where there's aurora on most dark nights whatever the Sun is doing, and the weather decides more than the solar cycle does. The winter of 2027–28 should be close to the last few.
For Scotland, southern Scandinavia or the northern US, the answer changes. The aurora only reaches those latitudes during strong geomagnetic storms, and those get rarer as the Sun calms down towards the end of the decade.
Is the solar maximum over?
Yes. The Royal Observatory of Belgium, which keeps the official sunspot record, puts the peak of Solar Cycle 25 in October 2024, at a smoothed sunspot number of about 161 (SIDC). NASA and NOAA had already said in October 2024 that the Sun had entered its maximum period (NOAA SWPC).
A cycle runs about eleven years. Cycle 25 began in December 2019, so the next minimum should come around 2030, and the next maximum in the mid-2030s.
Did the aurora get worse after the peak? Our camera data
We expected it to. It didn't.
Aurora's forecast is checked against all-sky cameras in northern Norway and Sweden, which record the sky all night, every night. Here is how often each camera saw clearly visible aurora, September to March, over the three winters around the peak:
| Camera | 2023–24 | 2024–25 (the peak) | 2025–26 (after the peak) |
|---|---|---|---|
| Tromsø | 31% | 24% | 41% |
| Skibotn, inland from Tromsø | 39% | 40% | 45% |
| Kiruna | 39% | 47% | 48% |
In Tromsø, the winter of the solar maximum was the worst of the three. The reason was the weather: only 50% of nights that winter had a usable sky at Tromsø airport, against 68% the winter before. On the clear nights, the camera saw aurora about as often in both winters (43% and 44%).
Then in 2025–26, with the Sun past its peak, Tromsø's camera had its best winter. On clear nights it saw aurora 58% of the time.
Three winters is a short record, and Kiruna's first winter is only half a season (88 nights). We won't claim that activity rose after the peak. What the data does show is narrower and more useful: at these latitudes, a cloudy winter moved the numbers far more than the peak of the solar cycle did.
The camera figures come from the UEC Tromsø all-sky camera classifier (Nanjo et al. 2022); the cloud figures from airport weather reports. Definitions are on our methodology page.
Why doesn't the aurora drop straight after solar maximum?
Two kinds of solar activity drive the aurora.
Coronal mass ejections are huge clouds of plasma thrown out by the Sun. They cause the big storms, the ones that paint the sky red over Germany or Texas, and they're most common near the maximum. The storm of 10–11 May 2024 was the first G5 (extreme) storm since 2003.
High-speed solar wind streams pour out of coronal holes, and they become more common as the cycle declines. A coronal hole can last for months and swing past the Earth every 27 days or so as the Sun rotates. The storms these streams cause are usually minor to moderate, which is plenty for Tromsø, Abisko or Fairbanks, and not enough for London.
Big storms don't vanish in the declining years either. The Halloween storms of October 2003, among the strongest on record, came more than three years after Cycle 23's peak.
So who should go sooner rather than later?
- If you're hoping to see it from home in Scotland, Denmark or the northern US, the next few winters are your best chance until the mid-2030s. Strong storms get rarer each year towards 2030.
- If you're planning a trip to the Arctic, don't rush it because of the solar cycle. Pick the place and the month by cloud and darkness. Our month-by-month guide and how many nights you need matter more than the year.
- If you keep putting the trip off until "some year": the cycle isn't a reason to wait. The winter you actually go is the one that counts.
What changes for forecasts after the peak
Coronal-hole streams come back with the Sun's rotation, so active nights become a little easier to see coming, sometimes a few weeks ahead. Kp still won't tell you if you'll see anything, because it ignores clouds (more on that).
Aurora's Tonight forecast for each place combines the solar wind, Kp, darkness, the moon and a cloud forecast. That works the same in a quiet year as in a peak year. A forecast, not a guarantee.
