R. . J. Livesey

The Aurora Section regularly receives enquiries from the public as to where and when they might go and see the aurora borealis. This paper looks at the currently available observing statistics in an attempt to answer this question.

The polar aurora

The polar aurora is virtually a permanent oval of auroral activity encircling the geomagnetic pole. Akasofu¹ puts the diameter on the average at about 4000 km, increasing to 6000 km when the oval is brighter and decreasing to around 3000 km when the oval is faint. The oval is centred on the geomagnetic pole but shifted to the dark side by about 300 km. In terms of angles of geomagnetic latitude from the geomagnetic pole, the oval lies (at a mean value) on the dayside of the geomagnetic pole at about 16^o and on the night side at about 20-30^o. The Earth rotates under this oval and the loci of all points on the Earth's surface that pass under the midnight sector of this oval define the auroral zone where the aurora is most frequently visible to ground based watchers.

The University of Tromso in north Norway operates an auroral observatory² and the district is a suitable location from which to see the polar aurora. Dr K. Henriksen³ of the observatory has given the following statistics for this site. The probability of seeing the aurora is close to 100% on any cloud free night between October and March. The worst observing conditions apply to October and November when snow or rain frequently falls and the probability of seeing aurora is only about 10 to 20%. In March the probability of seeing aurora rises to 50% when the poor Atlantic weather tends to track to the south of Tromso. Just as in the UK cloud cover at Tromso can be very variable and it is possible to experience 15 consecutive nights of continuous cloud.

Dr Alastair Simmons has spent several winter holiday periods at Tromso in January and February and has been particularly fortunate with the weather and his auroral observing. On the other hand Peter White has visited Tromso on three consecutive winters and seen precisely nothing of the aurora due to total cloud cover while in one of these years his visit coincided with a period of minimal geomagnetic activity.

When a geomagnetic substorm develops, there is a west-wards surge in the evening sector of the auroral oval followed by a migration of activity polewards. Several couples from time to time have spent holidays in Iceland in September during active solar periods and have been sucoessful in~seeing one or two events in the course of a fortnight. This activity is associated with southwards tuming of the north~outh cornponent of the interplanetary magnetic field and its interaction with the geomagnetosphere and magnetotail.

| Top Of Page |

The storm and quiet aurorae of lower latitudes

The development of a major auroral and geomagnetic storm⁴ to make the aurora visible at lower geomagnetic latitudes related to the UK is caused by impact of shock waves and high speed solar wind streams, combined with a southerly turning of the north-south component of the interplanetary magnetic field, upon the Geomagnetosphere and magnetotail. The extent of the equatorwards migration depends upon the strength of the magnetic storm.
Two types of auroral activity have been identified, the quiet glows and homogeneous arcs of higher latitudes seen especially in the declining years of the sunspot cycle and the active storms with rayed bands, rays, flickering and flaming forms that tend to relate to the sunspot cycle. Quiet aurorae are more often seen from the Moray Firth northwards while active storms, most frequently seen in the north of the UK, often extend farther south and in extreme cases are visible in the south of England and on occasion down to the Mediterranean.

The Aurora Section statistics enable the frequency of auroral activity to be calculated for various latitudes in the UK. In Table 1 are given the frequencies with which auroral light in any form of activity could be seen within a year given that cloud would not interfere with the observations. Strong moonlight adversely affects auroral observation but the times of full Moon are readily determined and avoided.

Quiet aurorae are associated with high speed streams of particles emanating from coronal holes on the sun and not necessarily the storm aurorae associated with shock events and transient solar events. Quiet diffuse aurorac, red in colour, are features found to the equatorward side of the discrete storm aurorae comprising rayed forms. These quiet red aurorae, called stable auroral red arcs, (SARC) are sometimes seen in England when the discrete aurora is observed in Scotland.

| Top Of Page |

The effect of cloud on auroral observations

Cloud statistics measured by the writer at Newton Mearns south of Glasgow over a period of years suggested that on the average only one night in five would be sufficiently cloud free to enable auroral light to be detected. Dr Alastair Simmons undertook a similar exercise at Milagavie north of Glasgow and came to the conclusion that only some 15% of his nights were sufficiently cloud free. The United States Air Force has published cloud cover data for the entire world. Assuming that an auroral watcher wishes to be able to see the sky at an altitude of 10^o above the horizon it is possible to calculate the percentage frequency with which the watcher will have a cloud free line of sight. Figures for the UK are included in Table 1.

It is the practice of the British Meteorological Office to keep cloud statistics in terms of the number of eighths of the sky that is cloud covered, the unit of measurement being called the octa. Mr David Wheeler, senior meteorological officer at RAF Kinloss on the Moray Firth has very kindly given figures for the period l980-l989⁵. During the hours of darkness the mean percentage time when cloud cover was 1-2 octas amounted to 18%, from 3-6 octas it was 25.6% and for 7-8 octas it was 56.4%. These figures are in general agreement with measurements at Aberdeen and Kirkwall except that the percentage for 3-6 octas is much higher than almost any other meteorological station in the UK. The fact that the Moray coast has an uninterrupted sea horizon to the north, is in a reasonably high geomagnetic latitude and has more favourable skies than some observing sites tends to the view that it is a good site for auroral observers. This is brought out by the efforts of the increasing number of auroral reports being submitted by new observers in that area.

| Top Of Page |

Comparison of auroral observations made on both sides of the atlantic

By good fortune, the Aurora Section has been approached by two observers who have made a long series of observations over a period of years and have made their respective records available to the BAA. Dr Michael Hapgood of the Rutherford Appleton Laboratory. has made calculations which showed that both observers were sitjiated at the same corrected goemagnetic latitude and therefore their records could be intercompared.

Dave Wheeler has operated a synoptic meteorological observatory on Fair Isle since 1974 and has calculated that, excluding minor glows, on the average he would expect to see 3.5 times more aurorae than he does were it not for the effect of cloud at his location. For the purpose of comparison, in the period 1981-1989 Dave logged 224 aurora event nights.

Jay Brausch has maintained a dedicated watch for aurorae at Glen Ullin in North Dakota USA where the aurora is seen in darker skies in summer due to his lower geographic latitude than is the case in the UK where summer twilight is a problem. In the period 1981-1989 Jay logged 732 aurora event nights and commented on the clarity of the atmosphere in North Dakota for observing.

Referring again to the United States Air Force cloud statistics the percentage time to obtain cloud free lines of sight were compared for an altitude of 300, bearing in mind that at the two observatories under consideration, aurorae at high altitudes are quite common. At Fair Isle the highest and lowest clear sight probabilities were calculated to be 38 to 26% respectively whereas at Glen Ullin the comparable figures were 70 to 35% respectively.

| Top Of Page |

Discussion

What advice can be given to holidaymakers and others wishing to see their first aurora? For a high probability that the aurora is active the watcher must travel as close to the auroral zone as possible⁶. Otherwise the watcher has to be content with looking for the storm aurora or the repeating quieter corona stream aurora in north Scotland particularly immediately before sunspot maximum or in the middle of the declining years of the sunspot cycle. Watchers should avoid the twilight of summer in the UK and the midsummer sun in northern Norway. Anybody who is travelling in North America might consider North Dakota in particular, even in summertime, as a likely vantage point. The period around full Moon anywhere is to be avoided. There is a higher probability of observing aurorae at the equinoxes although this may not be so for a given year, depending upon solar conditions. In the UK there is a higher probability of seeing the aurora at 2200 UT, at local magnetic midnight, when the Sun, the geomagnetic pole and the observer are in the same straight line. However, storms may develop at any time and it depends upon the location of the observer on the rotating Earth relative to the nightside tail of the auroral oval as to what is seen of any auroral activity.

Weather and cloud are the big unknowns for although weather statistics can suggest better periods of the year to observe in, seasons vary from year to year. November anti-cyclones tend to be cloudy so that a weather pattern associated with clear skies in summer may give poor visiblity- at other times of the year. To the casual traveller it is a matter of luck. Recendy there was the case of an English couple who raced up to SkYe for a three day visit and were blest with a clear night and an active aurora. The writer has spent several business trips in Shetland, seen clear skies and no aurora, but on the other hand has experienced what is jocularly referred to in reports to Bracknell as 'Force 10 Fog', a combination of fog and wind blowing at Force 10 on the Beaufort Scale, not exactly the best of weather in which to observe aurora.

| Top Of Page |

Conclusion

The best place so far identified to see the storm aurora with a minimum of cloud cover appears to be North Dakota. Given that the aurora is most probably present and the only problem is to hope for suficient gaps in the cloud cover, north Norway seems to be the place to go. The auroral zone of highest auroral probability passes down across north Norway and the Lofoten Islands, across to Iceland, just south of Greenland and over to James Bay south of Hudson's Bay. Based upon recent observations the best place to see the storm aurora would appear, in the UK, to be along the Moray coast with a clear view north to seaward free of town lighting. The longer the visit the better but there is nothing to beat becoming a local resident.

| Top Of Page |

Acknowledgements

Grateful acknowledgement is made to Dr Simmons, Neil Bone and Martin Mobberley for constructive criticism in the preparation of this paper. The observational work of Jay Brausch, Dave Wheeler, Dave Rutherford and all members of the Aurora Section and the Meteorological Office stations contributing to Aurora Section records are fully acknowlodged and as listed in the Section Annual Reports.

References

1. The Sotar Wind and the Earth, Ed. Akasofu, S.-I., and Kamide, Y., Geophysics and Astrophysics Monographs, Terra Scientific Publishing Company, Tokyo, 1987.
2. The Northern Light from Mythology to Space Research, Brekke, A., and Eggeland, A., Springer, Berlin, 1983.
3. Personal Correspondence, Henriksen, K., 1990.
4. 'The dynamic aurora', Alsasofu, S.-I., Scientific American, 260, 5,
1989.
5. Personal Correspondence, Rutherford, D. G., 1990.
6. Physics of the Upper Atmosphere, Bates, D. R., Ed, Ratcliffe, Academic press, 1960.

J. Br. Astron. Assoc. 101, 4, 1991

| Top Of Page |