How can you see the northern lights (aurora borealis) in Iceland?

Below are some of the most commonly-asked questions by visitors to Iceland regarding aurora-watching. The answers are simply intended to provide practical advice based on my own personal experiences of aurora-watching in Iceland. Please click on one of the questions below to find the relevant information. If you have a question that is not answered here then please consider emailing me. Then I can add your question to the list to help future visitors maximise their chances of seeing the aurora.


Q1. What is the best time of year to see the aurora in Iceland?

It is generally accepted that aurora activity worldwide is highest around the equinoxes, i.e. during September and March of each year. However, while your chances may be generally higher during those months, there is still no guarantee of activity. Excellent displays of aurora can also occur at any time over the winter between October and February. Additionally, auroras can be seen as early as mid-August and as late in the season as mid-April. Bear in mind that coming to Iceland at the very start (mid to late August) or the very end (early to mid April) of the aurora season will limit your potential aurora-watching opportunities to a relatively short period in the very middle of the night (i.e. 01:00 in the morning), which may not suit your plans. Naked eye views of auroras at those times will also require reasonably strong activity in order to beat the twilight skies.

If you visit over the summer between late April and early August then you have absolutely zero opportunity to see the aurora. The nights simply do not become dark enough during the summer at these latitudes.

Q2. What time will the aurora start tonight?

It is not possible to answer this question. Aurora activity fluctuates continuously according to conditions on the sun and in the earth's atmosphere. If it is sufficiently dark in Iceland then auroras could potentially be visible at any time. Peak times for activity are often between 20:00 and 03:00, but the aurora can certainly be active earlier or later than that if it is dark enough. During the winter I have personally observed vivid auroras as early in the evening as 18:00 and as late in the morning as 07:30. You can find information on the hours of darkness HERE. You should look for the times of "astronomical twilight" or "night" as these are the conditions when it is sufficiently dark to observe the aurora (although very strong aurora may also be visible through the twilight on either side of astronomical twilight).



Q3. I am visiting Iceland in *insert your dates* will I see the aurora?

Auroras are generated by activity originating from the sun. For example, coronal holes that appear on the surface of the sun can allow hot gases to flow outwards as high solar winds, while the coronal mass ejections (CME) resulting from sunspot flares can emit strong bursts of solar particles. Sometimes these events are earth-directed, causing auroras. However, since coronal holes and sunspots continuously form and disappear, there are limits to how far ahead in time auroras can be predicted.

Auroras within the next three days: The most reliable predictions of aurora activity have only a 3-day timeframe. This is because the transit time between events (such as CMEs) on the sun's surface and the earth's atmosphere is typically between one and three days, depending on circumstances. Predicted 3-day forecasts of aurora activity are issued by space weather organisations. However, the actual realised occurrence of auroras around the earth sometimes differs from these predictions by periods of many hours, and some predicted events fail to occur at all. Predictions are, by their very nature, uncertain, but the 3-day forecasts do provide guidance to the approximate periods when you might expect aurora activity to be highest.

Auroras within the next 27 days: The sun takes 27 days (just under four weeks) to complete one full rotation. Consequently, features on the sun that have caused auroras on earth (e.g. sunspots) may rotate back into an earth-facing position 27 days later, as the sun completes a full rotation around its axis. As a result, space weather research organisations also generate a predictive 27-day outlook of potential aurora activity, based on the activity that was recorded on earth during the suns previous rotation. It should be understood that features such as sunspots and coronal holes form, alter and disappear continuously, and so there is no guarantee that aurora activity will be the same on subsequent rotations. Nevertheless, the 27-day forecast is a useful tool in estimating likely occurrence, particularly for the larger and more stable coronal holes which may provide auroras on several consecutive rotations.

Auroras more than 4 weeks into the future: Predicting aurora activity more than 27 days ahead is not possible, and consequently if you book a trip to Iceland to see aurora months in advance then nobody can offer any degree of certainty about whether or not you will see it. It will be a lottery! However, you can improve your chances by making sure that you visit at the most suitable time of year and follow the other practical advice on this page.

Remember weather on earth! Seeing auroras also depends on the weather on earth, especially cloud. The occurrence of cloud is the single largest factor influencing the success of aurora-chasing in Iceland, and can sometimes obscure the sky for many consecutive nights. For this reason, if you are travelling to Iceland specifically to see the aurora then it is sensible to visit for at least one week. You might be lucky and see it during a 3-night mini-break, but your odds will certainly be increased by staying for longer.



While the aurora can be seen almost anywhere, your choice of location for photography should consider light pollution. In this image the clouds are lit orange by light pollution from Reykjavik, there is a strong beam of light from the Imagine Peace Tower, and there are horizontal streaks of light created by passing traffic during the long exposure. Whether each of these is desirable is a matter of personal taste!



Q5. I saw a report that the aurora was visible in Selfoss. I am in Reykjavik, should I drive to Selfoss?

No. See Q4. If the aurora is visible in Selfoss (or any other location) then it should also be visible throughout most of Iceland, as long as the sky is clear enough. To put this into context, auroras forming north of Iceland can even sometimes be viewed in Scotland (600 km to the south). I would only recommend that you jump in your car and drive to where someone else has reported it if your own location is completely clouded. But by the time you reach the other location, then the activity may well have changed or died down. Your best bet is just to get to your nearest cloud-free area as quickly as you can and look for yourself.



Q6. In the UK I rarely see aurora with my eyes but mostly via my camera. Is that also true in Iceland?

No. Iceland is situated at a much higher latitude than the UK or other "mid latitude" locations, and occurs within the oval around the northern magnetic pole where most aurora activity occurs. Auroras are very often visible to the naked eye in Iceland, which is why a booming tourist industry has been established around winter aurora-watching (it is doubtful whether most tourists would be satisfied with only camera views!). I almost always search for auroras by naked eye, and if I see them then I head out with the camera. Faint auroras can, however, look a lot like light wispy cloud to the naked eye, so it is sometimes helpful to use your camera to be certain.



Q7. My aurora phone app says that In 1 hour and 34 minutes, the aurora is predicted to be Kp 5 so should I rush out?

Firstly, I would always advise that you actually go outside and check the sky with your eyes, regardless of what your phone app or other people might be telling you. Your eyes will always provide you with the best available real-time indication of aurora activity (and of clouds!). Secondly, the use of predictive Kp values by most phone apps is extremely unreliable for the reasons I explain in Q8. As a result, I would not recommend relying on this method for your aurora-hunting.



Q8. Most aurora phone apps use Kp value, but what does it mean and how should I use it?

It is important to understand what the Kp value actually is, and why it doesn't represent an absolute reliable method of determining real-time, or future, aurora activity. The term Kp originates from the German term “planetarische Kennziffer,” which translates as planetary index. It is the global geomagnetic storm index, and ranges from 0 (very little geomagnetic activity) to 9 (extreme geomagnetic storming). The Kp values are generated as summarised global 3-hr measurements of K (a measure of how much geomagnetic disturbance there is at a particular location). To account for local variations between locations, the planetary Kp-index is calculated from the magnetic variation data recorded by 13 magnetometers worldwide.

With regard to the specific use of Kp values for real-time aurora-hunting, there are several inherent limitations:

  • Kp values are not generated in real-time. Rather, they are scientific values that can, by their definition, only be generated at the end of a 3-hr period based on data that has been recorded from worldwide magnetometers. Consequently, the Kp-index was not designed to be a real-time indicator of auroral activity, or a reliable forecaster of future activity.

  • Since Kp values represent an average of the measurements recorded from 13 magnetometers located across the globe, they do not reflect the finer-scale increases and decreases in aurora activity at particular locations. There is strong daily variation in the K index recorded at individual magnetometers, due to increased activity close to magnetic midnight. Therefore, the averaged global Kp value may not accurately reflect what aurora activity took place at your exact location.

  • Kp values are scientifically-calculated using a 3-hour timeframe. This is simply too long a duration to represent the growth and recovery of individual aurora substorms at particular locations, which happen over much shorter timeframes. The fact that Kp values are calculated in 3-hour periods, means that phone apps which tell you to the exact minute what a predicted Kp value is are unrealistic.

  • Predicted Kp values were not created as a real-time resource for aurora-hunters. Rather, they are generated by the NOAA SWPC as "worst case scenario" measurements to forewarn the users that are potentially affected by geomagnetic storms, for example electrical power grids, spacecraft operations, and users of radio signals that reflect off of or pass through the ionosphere. Consequently, the predicted Kp values are deliberately-issued as maximum possible values and are usually higher than the values actually achieved in reality for a given period.

As a result of these limitations, the Kp values and ovation model forecasts provided by many aurora apps are not a reliable method of ascertaining real-time aurora activity. If you doubt this, then it will soon become apparent to you when you fail to see spectacular aurora on nights of predicted Kp5, or when you go to bed because your app told you it was only Kp2 but you miss an amazing aurora shown by other peoples' photographs on social media. I can promise you that this will happen, because it happened multiple times to me before I began to read more background information on the aurora and understand what Kp values actually meant. I suggest several much better sources of information to use for your real-time aurora-watching in Q9.

If you wish to know more about why relying on Kp-based predictions is not recommended for real-time aurora-chasing then please check out the blog Why Kp is a poor indicator for auroral alerts.



Q9. What information do you use for aurora-hunting in Iceland?

I use a range of sources, including:

  • The NOAA 3-day and 27-day predictive forecasts. These give me a general feeling for which dates are most likely to produce enhanced aurora activity. I treat these forecasts purely to alert me to the 24-48 hr periods with the highest potential.

  • The cloud forecast from the Icelandic Met Office. I don't pay too much attention to their aurora forecast, which is based entirely on predicted Kp values (see Q8). I consult the cloud forecast only to give me a very approximate idea of what to expect in terms of clear skies. The actual cloud cover observed on an evening can vary considerably from the forecast. In general there tends to be more cloud at my location than predicted. However, on occasion I have seen a 100% cloud forecast, when the night turned out to be completely clear. So it is always advisable for you to check outside at regular intervals and monitor the cloud yourself.

  • The Glendale Skye aurora alert app. Unlike other aurora-alert apps, the Glendale app is based on real-time data collected from satellites and magnetometers, and on years of first-hand observations by Andy Stables (app designer) of the correlations between aurora activity in the UK and those data. During the 2017/18 aurora season I helped to develop this app for use in Iceland, and there are now Iceland-specific alerts to tell you whether an aurora is likely to be visible based on real-time data. I strongly recommend using this free app for your aurora-chasing in Iceland, and by reporting your own observations on it then you will help to improve it for other people too. I have this app running most evenings, and it emits an audible alarm when an aurora substorm commences.

  • The magnetometer located at Leirvogur in Iceland. This is the only magnetometer located in Iceland and provides real-time information on deviations in the magnetic field which are usually indicative of aurora. Some basic interpretation of the magnetometer is provided in Q10. I highly recommend that you familiarise yourself with this magnetometer display and monitor it while you are in Iceland.

  • Other real-time datasets that are important factors in generating aurora displays, such as the solar wind speed/density and the north-south direction (Bz) of the interplanetary magnetic field (IMF). The solar wind and Bz can be monitored at websites such as SpaceWeatherLive. However, they are now also available in one place on the Glendale Skye aurora alert app. The faster and denser the solar wind, the more likelihood there is for aurora displays. Additionally, the direction of the IMF needs to turn southwards to facilitate solar wind particles entering the atmosphere and creating geomagnetic storming. Consequently, a negative (southerly) Bz value such as -1 nT means a much higher chance of aurora than a northerly one, and the more negative the value the greater the potential for aurora.

  • My eyes! Over the course of a typical evening I go outside every 15-20 min and check the skies for both cloud cover and aurora activity. Both vary continuously. It is very common that during one check I see nothing, and then 20 min later there is a faint aurora arc appearing and I know it is time to go out. Sometimes I can see the aurora forming in the sky even before the magnetometer reacts. If you want to see the aurora then this is your single most effective method!

  • Reports from other people on some social media groups, primarily on the Aurora Iceland Facebook Group.



Q10. I don't understand the magnetometer. What do all of those lines mean?


A magnetometer is a device that measures deviations in the Earth's magnetic field which can be indicative of auroral activity. The magnetometer readings are measured in nanotesla (nT), which is a unit of measurement for magnetic field strength. Ambient levels result in a flat horizontal line, whereas a sharp dip in the line is indicative of activity. The bigger the dip in nT, the greater the level of auroral activity is likely to be. So, for example, the image to the right shows the Leirvogur magnetometer readings for the evening 21st/22nd October 2017, the aurora activity of which is shown as photographs in my corresponding aurora diary. In this graph it can be seen that baseline levels of activity were recorded up until 22:00 on the 21st and again after around 03:00 on the 22nd. However, between 22:00 and 03:00 there are strong fluctuations in the data which are indicative of changes in the magnetic field strength. Often, the strongest auroras by eye in Iceland correspond with direct vertical dips in the magnetometer (rather than necessarily the greatest -nT values), such as the one shown by the arrow in the diagram.

Magnetometer data from Leirvogur on the evening 21st/22nd October 2017.



Q11. The aurora is being reported in the UK, but I am in Iceland and I don't see anything! Why?

Sometimes, during very strong geomagnetic storms, the aurora oval that encircles the northern magnetic pole can shift entirely southwards which causes the aurora to be overhead in the Faroe Islands or other regions to the south of Iceland. On such occasions there can be "red alerts" on aurora apps and enhanced viewing of the aurora for UK-based observers looking northwards. At the same time in Iceland the aurora may be located more towards the southern horizon rather than overhead. For this reason, and rather counter-intuitively, the strongest geomagnetic storms do not necessarily produce better aurora activity in Iceland than a lower-strength geomagnetic storm.



Q12. The sky is clear, why can't I see anything?

If you are certain that there is definitely nothing to see with either your eyes or via your camera, then it is possible that the aurora simply isn't active. There are some nights where activity is simply very low, and the aurora is located far to the north of Iceland.



Q13. Can I still see the aurora when it is a full moon?

Yes. Aurora in Iceland are often of sufficient strength and intensity to be visible even when a bright moon is present, as indicated by the image to the right. As a result, it isn't really necessary to specifically plan your trip around the moon phase. Information on current and future moon phases in Reykjavik is available HERE.

Personally, I prefer to have some moonlight (especially around a 50% to 75% moon) in my aurora photographs, as the additional light greatly benefits the appearance of the foreground. As a crude illustration of that, take a look at the images below. Both were shot at identical settings of ISO 1600, 14 mm and quite a long exposure of 24 seconds, and I have applied no processing to them at all except for a lens profile to correct distortion. The first image, from the 19 September 2017, was taken the day before a new moon. The second image, from 3 October 2017, was taken two days before the full moon. You can see how the moonlight completely lifts the foreground, while the aurora is still clearly visible. The moonlit image also provides much more scope for exposure adjustments during processing, since the exposure can be reduced if necessary during processing without ill-effect, whereas increasing the exposure of the image taken during the new moon phase will introduce a great deal of unwanted "noise" and graininess.

This image from 29 September 2018 shows a bright moon to the right, and a moderate-strength aurora showing up well despite the moonlight.


19 September 2017. The day before a new moon.

3 October 2017. Two days before the full moon.



Q14. Will I see the aurora in colour in Iceland?

The colours in auroras are caused by electrical particles (protons and electrons) from the sun colliding with gas molecules in the earth's upper atmosphere (ionosphere) which causes the excited molecules to emit light. The colour of the light is dependent on the type of gas and also the altitude above earth at which the collision occurs. For example, oxygen atoms cause red and green colouration at altitudes above and below 200 km altitude respectively. Excited nitrogen molecules produce pink colouration at altitudes below 100 km.

The perceived colour of an aurora by a human (or a camera) varies according to factors that include the sensitivity of that person's eyes (or camera), the intensity/structure of the aurora, the altitude of the aurora, and the spatial distance from the aurora. Weak auroras or those observed at great distance may appear mostly a faint white, and lack obvious colour to the naked eye (but usually appear green to a camera). However, auroras of greater intensity can be seen in colour by the human eye.

In Iceland, almost all of the auroras that I observe appear as shades of green to the naked eye. Occasionally I see them in white towards the start and end of a period of higher activity. I also regularly observe pink with my naked eye at the lower edge of the aurora during periods of intense activity, and also purples at the top of high rays. What colours you will see with your naked eye is dependent on the type and intensity of aurora at the time, and also on the sensitivity of your own eyes. For example, I have stood next to my partner watching an aurora, and sometimes I can see pink fringes by eye while he does not.



Q15. I have seen lots of photographs of extremely colourful aurora in Iceland. Can I expect to see that with my eyes?

Yes...and No. As explained in Q14 the aurora can frequently be seen in colourful shades of green to the naked eye in Iceland. However, the appearance of pinks and other colours is mostly limited to the stronger periods of aurora activity.

With regard to aurora photographs, a camera will usually record a more intense and vivid aurora than was observed by naked eye, because long exposure photography captures the scene over multiple seconds compared with the instantaneous second-by-second processing of your eye. Cameras do also sometimes capture colours that you don't observe with your naked eye. In Iceland, I find that my photographs occasionally contain pink and purple colours that I could see only as pale greens or whites by eye. Unfortunately, many aurora images shared on social media have also been heavily enhanced during the processing stage on computer, which can emphasise even more the intensity and colours of the aurora. Heavy processing results in end images that may not accurately represent what was visible by eye, which, while a matter of personal choice, does tend to unrealistically raise people' expectations of what they will see themselves.

For these reasons it is advisable to limit your expectations of seeing a lot of colour. You may be lucky and see green aurora with your eyes. You may be very lucky indeed and visit Iceland during high aurora activity where the aurora will be extremely bright to the eye with amazing colours. On such nights you cannot miss it! Worst-case scenario is that you may see only a diffuse whitish glow in the sky, if the aurora is weak.



Q16. How do I know if it is an aurora or not? What does an aurora actually look like?

Faint, colourless patches of aurora can be easily confused with cloud, especially when the clouds occur in streaks across the sky. Sometimes the only way to be absolutely certain is to take a photograph, which should then reveal whether the patch is green or not. The structure and colour of auroras varies enormously, both between nights and within a single substorm as the aurora progresses through its growth, expansion and recovery phases. You could see anything from a uniform arc, to flickering rays, to a sky full of dancing bands and overhead corona. I have provided examples of these different types of aurora.



Q17. Which direction should I look in?

From mid latitudes such as the UK, you would always look directly north when searching for an aurora. This is not always the case in Iceland, where auroras can potentially be located to the east, west, overhead or, less commonly, located entirely south of your location. As a general rule I would still always recommend looking northwards, since my personal experience is that auroras in Iceland are most commonly located across the northern half of the sky rather than to the south. This is especially the case for weak auroras, which may be apparent as a low arc along the northern horizon. However, sometimes, during moderate to stronger activity, there can be bands of aurora directly overhead, continuing over the horizon at their north-east and north-west limits. Occasionally, when the aurora is particularly strong, activity shifts southwards and you would need to look to the southern sky to observe it from Iceland. My advice therefore is to always start with searching northwards, but also check 360 degrees around the entire sky.

An aurora along the northern horizon, taken in Iceland on 23 October 2017.



Q18. Should I go on a northern lights coach or boat tour? Which is best?

I have never personally been on a northern lights tour in Iceland, so I can offer no personal advice on them. In general, I think it very much depends on the type of person you are, your aims and whether you have sufficient previous experience of aurora-watching to be able to enjoy finding it for yourself. This is a matter of personal choice.

There are times when a coach tour may be the best option, most obviously when the roads are snowy/icy so that you aren't confident to drive yourself, or when you have limited time and there is a lot of cloud cover. Tour operators will try to drive you to areas where the sky is clear. If you do opt for a coach tour, then most of the operators offer you a free second trip if you don't see the aurora. I would not recommend going on a boat tour if the main aim of your trip is photography. Boats move and you will not be able to use a tripod unless the sea is extremely calm, resulting in blurry photographs. Many of the tour operators have websites showing their aurora diaries, where you can get a feel for what they usually see, the types of photographs they acquire, and whether or not it is for you.

There are also a number of independent small group aurora tour guides in Iceland, who will drive you far from Reykjavik (where necessary) to find cloud gaps and aurora. You are also likely to get the best photography tips on this type of tour (remember to ask about that, if photography is your priority). A simple internet search will reveal many aurora tour operators.



Q19. What about hiring a car and looking by myself?

If you are confident about driving in Iceland then this would be the first choice for aurora-chasing in my opinion as it gives you the maximum flexibility. Also an option are the camper van rentals, which mean that you could easily relocate to different parts of Iceland depending on cloud forecasts. The main limitation to self-drive is simply the weather, and people being nervous or inexperienced at driving on snow and ice. For that reason, this option may be best from late August through to October, before the worst of the winter weather has arrived. Please pay attention to weather forecasts and see http://www.road.is/ for up-to-date information on road conditions.



Q20. I don't have transport and I am limited to Reykjavik. Can I still see the aurora?

Yes. The aurora is frequently seen and photographed from the city. Some of the better locations for aurora-watching comprise areas with open views to the north across the sea, including the viking ship "Sun Voyager" sculpture on Reykjavik seafront, the harbour near the Harpa conert hall, the Grótta lighthouse, and the entire footpath along the seafront between the harbour and Grótta island. These areas can become quite busy with other aurora-chasers.



Q21. I'm thinking of buying a camera/lens to photograph the aurora, which one should I get?

Please see my how to photograph the northern lights in Iceland page for guidance on aurora photography in Iceland. Ideally you will need a DSLR camera and the widest and fastest lens that you can afford.



Q22. I want to try and photograph the aurora on my phone, can you recommend an app?

People increasingly ask about the possibilities to photograph the aurora using their iphone or android phone or similar. There are apps available that will help with this - just look for any app that facilitates long exposure photography. You will certainly also need a tripod, because hand-holding a camera for even a couple of seconds will result in tiny movements that will cause the stars and other features to blur. However, my advice ends there because I would never personally opt to use a phone to photograph the aurora. Sure, you will likely be able to get something that shows a little green in the sky. But the image will almost certainly be blurred, grainy and darker than you would like it to be. If your aim is simply to take a photo to put on social media that proves you saw an aurora, then this may be fine. If your aim is to get a photograph that you might want to display on your wall as a memory, then my advice is to ditch the phone and get a suitable DSLR camera or even a modern smaller compact camera (such as the Sony RX100 which I review HERE).