All reports since 2010 could be found here.
The purpose of this study is to identify the amount and of days with clear weather and possible aurora visible due to higher solar activity reported by multiple station between Greenland and Norway.
This study IS NOT about calculating chances to see the aurora rather than building a statistical model based on historical forecasts. This study IS NOT about how many times the northern lights were actually visible.
Each year GFZ German Research Centre for GeoSciences publishes the Aurora statistics through the year. In the yearly dataset each record contains a very important 3-hour Kp-index. It describes the disturbance of the Earth's magnetic field caused by the solar wind.
So, the dataset contains 365(6) records per year. Each record is a set of daily measures divided into 8 groups of 3-hour measures:
- 0-3 range
- 3-6 range
- 6-9 range
- 9-12 range
- 12-15 range
- 15-18 range
- 18-21 range
- 21-24 range
Each 3-hour measure has its kP value as well as aP value, more information on the dataset you may find here. The dataset can be effectively transformed into the following one:
{
<date>: {
<3-hour-range>: <kP-measure>
}
}
A new format of the dataset will allow effectively work daily measures (filtering 3-hour ranges with kP value lower than requested one, finding MAX, etc.).
An ability to witness the Aurora is the result of multiple factors:
- Sun's activity that causes the disturbance
- weather (humidity, cloud cover, temperature)
- appropriate geo-positioning (meaning posted below)
- the season of the year
The value of the disturbance value (kP) is pretty obvious - the greater disturbance causing higher chances to see the Aurora. But the ability to "see" is quite challenging, mostly due to the weather.
Most of the time, the highest impact is caused by the weather. High and low clouds and the cloud cover in general, high humidity and higher temperature. The Aurora looks perfect on photos and in person only when the sky is clear with pretty low temperature and low humidity along with nearly-zero cloud cover.
But the weather is not the only factor that may impact chances to witness the Northern Lights. First of all, most of the time with a moderate sun's activity, it's highly recommended to go far north (as close as possible or far beyond the Polar Circle).
The winter season is the best one for chasing the Aurora. During winter months the length of the daytime is pretty short, up to 6 hours, or could even be shorter depending on a country's location.
This is the most important part of this project!
According to the magnetic field disturbance theory, the whole Polar Circle divided onto belts ranged from 0 to 9. Each belt has an assigned magnetic field disturbance coefficient of a 3-hour measure (known as kP). Based on the minimum (minimum enough to see the aurora with the naked eye) kP value, for the particular country within the particular belt, for each observation point (defined as the GPS point) within the country, identify particular dates with recorder kP higher than the minimum kP taking into account weather conditions through the winter night hours (9 PM to 6 AM).
Taking into account that nobody collects reports of the aurora observation at ALL locations. The only data we could rely on is - GFZ data. So, having a piece of knowledge regarding minimal values for the disturbance that leads to potential visibility of the lights, it is necessary to figure out historical weather forecasts for the whole Aurora season. With that, we could declare a simple rule:
If the solar wind will cause a great disturbance, with constant good weather (short percentage of cloud cover,
low humidity, near-to-zero chances of perception), during the winter, at nighttime,
between 9 PM to 6 AM the chances to see the Northern Lights are pretty high.
In real life, with historical weather forecasts, it is highly possible to identify the number of days, the actual dates, exact locations where the Aurora was visible with the naked eye.
FYI. In most of the northern European countries, winter months are September to April. So, we'll be using these months to check if a good weather day matched Sun's activity.
1. For each country in the rally, filter out days my country's minimum kP during nighttime.
2. For each of the GPS coordinates of a country, check nighttime hours weather of each date filtered at step 1.
3. Using historical weather forecasts, identify whether the particular day was successful to witness the Aurora.
4. Calculate the statistics: successful dates per GPS location, the over dates, the total number of days with the kP value higher than the country's minimal one.
Doesn't seem to be very complicated, does it?
Aaaaaaand the winner is... okay, not yet (or scroll down).
Due to specialties of each country taking part in the rally. It is really hard to compare the experience, but it's easy to compare numbers. As for landscape photography, it is more important to have a decent foreground, rather than just the sky.
All locations used in this comparison are heavily inspired and tested by lots of landscape photographers. So, the following number would mean more to the photographers rather than to typical tourists who don't care much about photography rather than about sightseeing.
I've been to northern Norway (Nordland) in 2019, and I kinda can confirm these results. Out of the 12 days of the trip we've seen the Aurora maybe 3-4 times somewhere between 2nd to 9th of February.
Norway had 27 days with observable northern lights, 45.76271186440678% of total dates
per-month activity (month to a number of dates with higher kP): {
FEBRUARY : <[2019-02-21, 2019-02-09, 2019-02-05, 2019-02-01] days>,
OCTOBER : <[2019-10-31, 2019-10-28, 2019-10-27, 2019-10-02] days>,
NOVEMBER : <[2019-11-23, 2019-11-22, 2019-11-21, 2019-11-11, 2019-11-06] days>,
JANUARY : <[2019-01-31, 2019-01-25, 2019-01-24, 2019-01-23, 2019-01-14, 2019-01-05, 2019-01-04] days>,
MARCH : <[2019-03-17, 2019-03-16, 2019-03-08, 2019-03-07, 2019-03-03, 2019-03-02, 2019-03-01] days>
}
The winter season of 2019 was quite in Norway, mostly due to constantly changing weather. It was snowing a lot. So, the sky remained covered with clouds most of the time.
I haven't been to Iceland in winter months yet. But things look better than in Norway. However, it doesn't mean that spotting the Aurora is very easy.
Iceland had 35 days with observable northern lights, 59.32203389830509% of total dates
per-month activity (month to a number of dates with higher kP): {
NOVEMBER : <[2019-11-29, 2019-11-17, 2019-11-06] days>,
DECEMBER : <[2019-12-19, 2019-12-11, 2019-12-09] days>,
MARCH : <[2019-03-28, 2019-03-17, 2019-03-16, 2019-03-08, 2019-03-07] days>,
JANUARY : <[2019-01-31, 2019-01-25, 2019-01-23, 2019-01-17, 2019-01-16, 2019-01-15, 2019-01-06] days>,
OCTOBER : <[2019-10-31, 2019-10-29, 2019-10-28, 2019-10-27, 2019-10-26, 2019-10-10, 2019-10-02, 2019-10-01] days>,
FEBRUARY : <[2019-02-18, 2019-02-14, 2019-02-13, 2019-02-10, 2019-02-09, 2019-02-05, 2019-02-03, 2019-02-02, 2019-02-01] days>
}
35 days to 59 days overall, a pretty damn good stats. But take a closer look, it was required to stay long enough to see the Aurora. 1st week of February was CRAZY: Feb 1-3, 3-5, 5-10. It means that you'd need to stay for at least 2 weeks to raise your chances to see Aurora. But 2 weeks in Iceland is quite challenging, especially, if you're not doing the photography as the variety of activities is quite limited and the weather could be really challenging.
Sweden is probably the southernmost country of those that take part in 2019 rally. It means that Aurora must be very strong to be visible, even though that lots of photographers do Northern Lights tours to Abisko.
Sweden had 17 days with observable northern lights, 45.945945945945944% of total dates
per-month activity (month to a number of dates with higher kP): {
DECEMBER : <[2019-12-19] days>,
NOVEMBER : <[2019-11-24, 2019-11-23] days>,
OCTOBER : <[2019-10-29, 2019-10-27] days>,
FEBRUARY : <[2019-02-21, 2019-02-14, 2019-02-05] days>,
MARCH : <[2019-03-31, 2019-03-17, 2019-03-16, 2019-03-01] days>,
JANUARY : <[2019-01-31, 2019-01-24, 2019-01-23, 2019-01-05, 2019-01-04] days>
}
In 2019, Sweden was not the best location to see the Aurora, if not to say - the worst. Only a few days, for a long period. It doesn't sound fun, at all.
I was quite surprised to see these numbers. I had a feeling that Nordland's is way more up north (Alta, Nordcap), but I was wrong. I'm happy to be wrong! First of all, I haven't been to Finland yet, but according to this research, Finnish Lapland is the best place to witness the Aurora.
Finland had 59 days with observable northern lights, 100.0% of total dates
per-month activity (month to a number of dates with higher kP): {
DECEMBER : <[2019-12-19, 2019-12-11, 2019-12-09] days>,
NOVEMBER : <[2019-11-29, 2019-11-24, 2019-11-23, 2019-11-22, 2019-11-21, 2019-11-17, 2019-11-11, 2019-11-06] days>,
MARCH : <[2019-03-31, 2019-03-28, 2019-03-17, 2019-03-16, 2019-03-15, 2019-03-14, 2019-03-08, 2019-03-07, 2019-03-03, 2019-03-02, 2019-03-01] days>,
JANUARY : <[2019-01-31, 2019-01-25, 2019-01-24, 2019-01-23, 2019-01-19, 2019-01-17, 2019-01-16, 2019-01-15, 2019-01-14, 2019-01-06, 2019-01-05, 2019-01-04] days>,
OCTOBER : <[2019-10-31, 2019-10-29, 2019-10-28, 2019-10-27, 2019-10-26, 2019-10-25, 2019-10-24, 2019-10-21, 2019-10-10, 2019-10-05, 2019-10-02, 2019-10-01] days>,
FEBRUARY : <[2019-02-28, 2019-02-27, 2019-02-21, 2019-02-18, 2019-02-14, 2019-02-13, 2019-02-10, 2019-02-09, 2019-02-08, 2019-02-05, 2019-02-03, 2019-02-02, 2019-02-01] days>
}
100% - WOW. This is impressing.
It does seem to be obvious - the winner has nearly 100% match between good weather days and high magnetic field disturbance.
FINLAND
And the Winner is: Finland, Congrats!
Chances to see the Aurora were: 100.0
With total number of clear sky nights vs high kP nights: 59 vs 59
The best month: FEBRUARY
The best location: Kilpisjärvi [69.0443388,20.803093]
First of all, I'm not trying to predict the Sun's activity rather than to figure out if there's any pattern between years and countries. The GFZ provides data since 1938, but the historical weather reports are available only for the last 10 years (GPS-based). So, I'm kinda limited in a number of years to compare: 2010 to 2019.
According to the reports I've calculated, until 2018 both Iceland and Finland were sharing the No. 1 place as The Best Country for the Northern Lights. However, as I stated above, Iceland is No.1 if you stay long enough, for at least 2 weeks as the Sun's activity and good weather days are spread through months. The same barely applies to Finland.
Until 2013, March was the best month to see the Lights in Iceland. Since then, the period of time has changed to sometime between October and December.
Since 2018, Finland became No.1 country for the Northern Lights. Maybe cause of the number of storms going through northern Atlantic, maybe climate changed? But Finland is the mainland, which means the temperature is low, all clouds eventually turning themselves into snow showers, which leads to clear skies.
Solar activity and magnetic field yearly records provided by: GFZ German Research Centre for GeoSciences.
Historical Data provided by DarkSky weather service.