The Merry Dancers

Church Aurora

Aurora Borealis (1865) by Frederic Church. While Church apparently based this image in part on the notes and sketches from an Arctic-exploring friend, he likely witnessed the aurora personally at times as well, perhaps even from the land around Olana. While a straightforward landscape depiction at one level, this painting is also believed to present an allegory for the culmination of the Civil War. Courtesy Wikipedia and Smithsonian Institution. Image in the public domain (


PLEASE NOTE: There are three different ways that you can read this blog. First, read ye olde paper copy – go out and buy the Columbia Paper, and read it as our end-of-year “Perspectives on Place” contribution; second, you can read the digital narrative below and ignore the footnotes, thereby getting the gist of it all, and saving yourself the mind-twisting details (it is a holiday after all!); lastly, if you want to be hard core and take advantage of the ‘added content’, then explore the links to the footnotes – they’ll take you a level deeper into understanding our shifting aurora, watch out for the geomagnetical quicksand. However you do it, we hope you enjoy it!



On February 19th, 1831, at around 8:30pm, somebody, possibly Principal J.W. Fairchild, stood looking at the night sky from near the Hudson Academy atop Prospect Hill. A half moon hung in the dark. Notebook in hand, our observer was aurora watching, and tonight he was not to be disappointed:Brilliant…in the west and south west, shooting up in spangles towards the zenith, very much like the process of crystallization beneath the solar microscope. These consistently faded, and were succeeded by others in different lines, exhibiting at times most of the colors of the rainbow. About half-past 9, similar appearances were seen in the east and south east, meeting those mentioned at the centre above, and forming an illuminated dome of spars and spangles, the most brilliant and beautiful ever beheld.”1

On that same night, others were also peering upwards. Near Kinderhook, the aurora (aka Northern Lights) was uncommonly beautiful streaks of light. At Albany, columns were observed shooting up to the zenith from the whole northern hemisphere”. In New York City, the aurora began to be visible around 9PM, and was “peculiarly interesting… Some of the eastern coruscations were at times transiently curved, as though their middle parts were driven eastward by the impulse of the westerly breeze that was blowing at the time… A luminous band…passed near the moon, around which was one of the large haloes.” Sky watchers in Otsego, St. Lawrence, Oneida, Franklin, Herkimer, Westchester, and Kings Counties also logged awe-inspiring displays on that evening.

While particularly brilliant, this was no one-off: around the State, throughout the calendar, and across the years such night logs accumulated. These observations were not the incidental sightings of several people who happened to be out for a night-time ramble, they were mostly the duly-reported annotations of participants in one of the Country’s earliest citizen-science efforts: the network of New York State academies, which Anna described in an October Columbia Paper article. These men (and they were almost all men) had been enlisted by the Regents to gather observations on the working of the weather and other celestial processes. While not institutions of the Regents, the Regents provided limited funding for the payment of teachers and other necessities, and the academies filed annual reports justifying and describing their efforts. Beginning in 1827, they were also asked to brave the night-time chill in order to note auroral activity.2

At that time, scientists were only just beginning to turn an analytical eye to the weather, although the revolutionary idea of weather forecasting was still a couple of decades away. What the so-called Scientific Revolution had so far brought to meteorology was not its understanding, but rather the conviction that it could be understood. For generations, people had read portents into celestial events, deriving omens good or bad from the likes of haloes, meteors, eclipses, and auroras. Now ‘science’ was taking its turn. Observers were observing and patterns being sought. Who knew what mysterious threads might link aurora, magnetism, electricity, and weather? For example, based on observations made two months later to the day, a then-little-known teacher at the Albany Academy drafted a short note entitled “On a disturbance of the Earth’s magnetism, in connection with the appearance of an Aurora Borealis, as observed at Albany, April 19th, 1831”. The author was Joseph Henry and, in 1849, he was appointed the first Secretary of the nation’s premier national scientific institution – the Smithsonian, whence he spearheaded meteorological studies, including the creation of the Country’s first weather maps.3

In good scientific form, the Regents tried to standardize the work of their collaborating observers so that the reports would be more comparable across geography and time. In 1833, they published aurora observation instructions assembled by the illustrious British Society for the Advancement of Science: during a one-hour observation period set to begin at 10PM, various characteristics were to be noted such as opacity, breadth, velocity relative to the stars, lateral motion, “defects in symmetry” and elevation with the aid of a theodolite. However, despite such analytical instructions, the aurora continued to simply enthrall. On January 14th, 1837, for example, a Kinderhook observer described the show asBrilliant, fantastic and very changeable; arcs, radiations, flashes and lurid banks On the third of September 1839, an Albany viewer noted Splendid; the entire heavens lighted up with long massy rays of a rich silver hue, radiating from the zenith, and forming a dome of magnificent proportions. Deep crimson mass in east and west alternately, which formed a striking contrast with the long lines of white light with which it at times mingled… Light so strong at times it cast shadows.” On 18 November 1848, an observer in New York City reported, The Merry Dancers very numerous.” 4

Each of the Regents reports was filled with such notes, and a 19th-century compiler estimated that, at one academy or another, the aurora were noted on about 50 nights per year. While the Regents may have struggled to derive standardized information, it’s clear that many had the opportunity to wax poetic about these light shows. Fast forward to the present, and how many of you have seen aurora from your backyards? Aside from any mystery about their origins and interconnections with other terrestrial and celestial phenomena (connections which are, by the way, still debated), one of the questions that taps most persistently at the skull of a modern reader of these accounts is, Where are the aurora today? The answer to that question tells us something about the aurora and perhaps something about ourselves.5

One of the explanations for why we see fewer auroras today is, simply, that there are fewer. This is because the Sun is fickle and the poles have wanderlust. The current scientific explanation for the aurora is that a flow of protons and electrons emanating from the Sun as the solar wind interacts with the Earth’s atmosphere, exciting atmospheric atoms which release light as they subsequently calm down. Because of the interactions with the Earth’s magnetism, an auroral halo forms in a roughly 350-mile wide band about 10-20° from magnetic north (or south). The stronger the ‘hose’ of the solar wind, the brighter is that halo and the farther from the poles the auroras are visible. While some of the high ‘floods’ of solar wind are caused by unpredictable solar flares, others are associated with turbulence on the Sun’s surface which, in turn, can be indexed by counting sun spots, those dark blemishes visible on the Sun’s skin. More sun spots will, in general, mean more aurora. There is a continuous record of sun spot abundance going back into the 1700s, and so we have a way of numerically comparing then and now in terms of one force behind the aurora. Inspection of those records reveals a roughly eleven-year cycle in sunspots and shows that we are indeed on the waning arm of one of the weakest recorded sunspot cycles.6

On top of this, we are getting farther from the magnetic North Pole. The magnetic North Pole and the rotational North Pole – the one heralded by the North Star – are not the same. Indeed, the point towards which your compass directs you has diverged from the rotational North Pole for all of its recorded history. Furthermore, the magnetic North Pole, to the chagrin of navigators, wanders. Today, it is moving towards Siberia at around 35 miles per year. In 1831, the year it was first pin-pointed, the magnetic pole lay at about 70° N, 96° W (a location in northern Canada some 1400 miles from the rotational North Pole and 2100 miles from us); today it is found at roughly 86° N, 159° W (a spot in the Arctic Ocean about 250 miles from the rotational North Pole and 3200 miles from us). The auroral halo has moved with it. Picture a classical monk with his tonsure of hair around a bald pate. With a good barber, the ring of hair will perfectly encircle the top of his head, and a fly sitting on either ear gets a roughly equal view of the furry higher reaches. Now suppose that, after a few glasses of wine, the barber is a bit off center – he keeps the radius of the hair ring constant, but tilts its center point to the left. The fly on the monk’s left ear may then be brushing the hairs from its eyes, while the fly on the right ear may be convinced the monk is now bald. If we assume the ring of hair is the aurora, the monk’s head is the globe, and we observers are flies, then we were the left-ear fly in the 19th century but are now heading towards being the right-ear fly. The auroral halo is receding from our view.7

These celestial and geophysical processes probably account for much of the auroral drought at our latitudes, but we ourselves may also be contributing. Foremost amongst our own contributions is probably light pollution, the erasing of the nighttime sky by our ever-more powerful lights. In the Academy records, for example, Erasmus Hall, located near Prospect Park in what is today Brooklyn, provides some of the most vivid descriptions of the Northern Lights. One need only compare the candlepower of a mid-19th century gas street lamp (ca. 13 candlepower) with those of a modern street lamp (potentially measured in the 1000s) to understand that the neighborhood of Erasmus Hall was surely a darker place during that era. Such is true not only of city locations but also of more rural spots, where the light auras of nearby villages or cities, or of the commercial or residential cluster down the road tinge the nighttime sky and so can mask faint auroral glows.8

Finally, while we can pin some of the blame for the apparent rarity of modern aurora on sun cycles, drifting poles, and light pollution, perhaps we are also short in the wonder that fuels observation. Those academicians, stamping their feet, clutching their pencils, and, no doubt, pulled by the tasks of the day ahead or behind, weren’t just out to see a show, they were out to discover. They believed that through patient, coordinated observations, perhaps with compass (to detect auroral-induced magnetic variation) and theodolite in hand, they could start unraveling the aurora’s secrets, revealing mysteries which had puzzled generations. Curiosity and the idea that new knowledge could be derived from the observations of the ‘common person’ was heady stuff and likely a potent spur for getting up from beside the fire. Imagine looking at the night sky and seeing more questions than answers; and imagine believing that some of those answers might be at your own finger tips. One of the greatest challenges to learning today is, I think, the perception that we know it all. We don’t, but sometimes the enticing corners of unknown which can be illuminated by our own senses get buried beneath a dulling hubris.

It’s unlikely that many of us who stay in the County during this upcoming days will see the aurora, and yet who cannot wish that their holidays might sparkle just a bit more given a visit from the Merry Dancers. We wish you all such a visit and, more than that, we hope that as you travel through the natural world in 2016, you have the health and peace of mind to really wonder.9

This image of the Aurora was created by Étienne Léopold Trouvelot, a Frenchman who worked in Boston from about 1852 to 1882. The subtitle states "As observed March 1, 1872, at 9h. 25m. P.M.", presumably from near Boston. An accomplished astronomical artist, Trouvet is best known today as the man who introduced Gypsy Moth into Massachusetts. Reportedly, he alerted others to the potential problem, but none took him seriously.

This image of the Aurora was created by Étienne Léopold Trouvelot, a Frenchman who lived in Boston from about 1852 to 1882. The subtitle states “As observed March 1, 1872, at 9h. 25m. P.M.”, presumably from near Boston. An accomplished astronomical artist, Trouvelot is best known today as the man who introduced the Gypsy Moth into Massachusetts. Reportedly, he alerted others to the potential problem, but none took him seriously. Image courtesy of Wikimedia and the New York Public Library.



1) ^ Each year the academies submitted their reports, including aurora sightings, to the Regents. These reports were compiled and published in the Annual Report of the Regents. Periodically, these annual reports were gathered together and published in a multi-year volume. The first such volume was published in 1855; and the second in 1872.

There are many beautiful auroral videos on line; here are two of my favorites, feel free to suggest your own: Ole Salomonsen’s Polar Spirits, much of which may be accelerated time-lapse photography, and these two (by Garðar Ólafsson and Ronn Murray) which are in real time and so perhaps give you a better feel for what the 19th century observers were probably seeing. This space-station footage shows a unique, if somewhat ‘distant’, view of the Aurora.

2) ^ For more on the network of academies, see our web page on this project.

3) ^ The Smithsonian has a web page profiling Thomas Henry and his role in early meteorological studies. Henry’s early article on magnetism is available here. While it only touches upon developments in the US, Peter Moore’s captivating book on the development of 19th century meteorology, The Weather Experiment, is a fine read that provides relevant background on the state of meteorological thinking at this time. For an example of contemporary pattern searching using these records, see Joslin’s Meteorological Observations and Essays (1836).

4) ^ Eager to try Aurora observation yourself? Check out the British Association for the Advancement of Science’s Instructions for Observers of Aurora Borealis as published in the 1834 Annual Report of the Regents. Despite its jovial, spontaneous sound, “Merry Dancers” was actually a repeated descriptive term in the Aurora accounts, and I’m not sure if it referred to all Aurora or to a particular class of Aurora. One hint on contemporary usage comes from the 1838 Annual Report, in which a Mr. Haskins, a Buffalo-based observer states,”All these [auroral streamers] rose, faded, and were renewed again and again, with great rapidity; but they did not exhibit any of that tremulous motion sometimes denominated Merry Dancers” [italics in original]. As this blog describes, the term may have originated in the northern isles of the UK, although it’s not hard to believe that it popped up more than once.

5) ^ For papers describing apparent links between sun activity (as reflected in aurora) and concurrent climate, see this paper (summarized here) by two NASA scientists linking water levels in the Nile with northern European auroral observations between 622 and 1470 AD. Another paper, by a researcher at Duke University, compares climate cycles and auroral cycles.

6) ^ For background on the aurora and their connection with solar storms, see this easy-to-understand video. It gives an overview of the connection between solar storms, sunspots, and the Earth’s magnetic field. As is often the case, Wikipedia contributors do a nice job, its pages on sunspot cycles and its table of historical cycles provide some textbook-style background; either source will provide you with a sketch of current and bygone sunspot cycles.

For those of you leery of links (or lazy), here are two graphs showing sunspot cycles:

Sunspot cycles since about 1600. The Maunder Minimum was an intriguing late 17th century sunspot lull; Wikipedia offers a good summary of it. In this image 1928 is marked because, when this graph was made in 2009, scientists believed it would be similar to the upcoming cycle. (Where they right? See below.) Notice how sunspot activity in the mid-1800s seem relatively high, although there was also a mid-20th century peak. From;


The latest sunspot number progression plot

The scientists who predicted that the most recent cycle would peak at levels similar to the 1928 peak were not far off; this cycle seemed to peak with a sunspot index of about 75 sunspots. From


7) ^ Before getting into the deep and mucky intricacies of the priest’s wandering tonsure (, there’s a useful number to become familiar with: the Kp index. This index is, so far as I can understand, a number representing on the scale of 0 (very low) to 9 (very high) the current strength of solar wind being experienced by the Earth. Apparently the “p” stands for planetary and the “K” stands for the German word Kennziffer which means, ta-da, ‘characteristic number’. There, doesn’t that help?!  Basically, it seems to be a measure of how much the lines of magnetic force around the Earth are deviating from orientations typical of calm solar weather. Think of the surface of a pond: when there’s no wind, that surface is perfectly flat, however, as the wind builds, so too does the angle assumed by the waves of water. That’s not a perfect analogy, but as the solar wind increases, the magnetic lines are pushed further from the norm, and so higher Kp, means higher solar wind energy.

The reason Kp is so useful to us is that, because it is directly related to solar wind energy, it is also, more or less, directly related to aurora strength and, hence, the chance of seeing the aurora at any particular place on Earth. Maps such as this indicate the Kp value at which aurora might be visible at a given latitude. Using that map, one can predict that, at our latitude, aurora would not be visible unless Kp reached about 7 or higher, quite a strong solar storm. For our purposes (this doesn’t hold at the highest latitudes), the farther one is from the magnetic north or south pole, the higher the Kp needed to view an aurora. But, you may ask, how the heck am I supposed to know the current Kp values, it’s not as if they are regularly given on the six o’clock weather? Luckily for aurora lovers, there are regular forecasts. The NOAA space weather program, for example, gives the latest Kp index on this page (for the everything-and-the-kitchen-sink version of these data, see this page). As I write this, the Kp index is around 3, and there’s no point in my running outside to scan for aurora.

How kind, you may think, for the government to cater to aurora lovers. Their motives are more practical: geomagnetic storms not only cause aurora, but can also cause radio interference and, at high levels, badly damage power grids. During the greatest recorded solar storm in 1859, not only were there terrific aurora, but some telegraph operators were able to communicate with each other relying only on the solar electrical energy gathered by their transmission lines. Other operators were less fortunate and received bad shocks and/or had their equipment disabled by the storm. For more on that 1859 solar ‘hurricane’, you can read this account.

OK, so the last part of our puzzle is this: if we know that the Kp value (i.e., the intensity of solar storm) needed in order to make aurora visible increases with distance from the magnetic north pole, and we know that the magnetic north pole has wandered away from us over the last 200 years or so (e.g., see this map or, with bells and whistles, here), then what Kp value was needed in order for there to be visible aurora at our latitude in 1831, when the magnetic north pole was first directly determined, and how much more likely did that make a visible aurora for, say, an observer in Kinderhook?

Answering that will involve a series of ‘back-of-the-envelope’ calculations whose crudity would probably make a good space scientist shudder. (Actually, a couple of them gave me input on this, but my mistakes remain my own.)

Here are the basics:

In 1831, when Ross first pinpointed its exact location, the magnetic North Pole was found at about 70.09°N, 96.77°W. Today, best estimates put it at around 86.3°N, 160°W. These locations are, respectively, about 2095 and 3230 miles from our present location in Hillsdale, NY. In other words, relative to the magnetic pole, we are now about 1135 miles farther south than we were in 1831.

However, if one gets down to the nitty-gritty, as I did with the help of those kind folks at USGS and NOAA, it’s not, strictly speaking, just the magnetic pole’s location that determines the likelihood of us seeing aurora. It’s all a bit more complicated but can be reduced to a number called “corrected geomagnetic latitude”; a short-hand for determining how far north we are relative to the geomagnetic (rather than geographic) North Pole,  NASA has a handy-dandy web page for calculating one’s corrected geomagnetic latitude for various years.

Using that tool, in turns out that our modern “corrected geomagnetic latitude” is about 51°N, whereas it was about 55.3°N in 1900 (the earliest date this tool accepts and, given how little the poles wandered between 1831 and 1900, probably not too far from the 1830s-1840s value). In other words, we were, from an auroral perspective, about 5° (or 350 miles) further north back then.

If we return to that map of the Kp’s needed in order to have visible aurora, we can make the following hypothetical modification,

Kp map

Our effective latitude, relative to the geomagnetical North Pole, in the 1800s vs. today. We were, in terms of aurora viewing, about 350 miles further north.

Be careful to interpret this map correctly. Relative to the geographic North Pole, we basically didn’t move over this period, so this does not say we were  experiencing southern-Canadian seasons in the 1800s. However, relative to the geomagnetic fields, our relocation has been more dramatic. Following this logic and using the above map, in the 1800s we were probably experiencing aurora skies like those in Minneapolis or southern Canada today. In other words, we saw aurora at a Kp of around 5, whereas today we need a Kp of 7 or higher.

Continuing on this vein, how much more likely were local residents to see aurora in the 1800s vs. 2015? To answer that we have to ask how frequent various Kp values are – in other words, how likely is it for the Earth to be buffeted by a Kp7 solar wallop vs a stiff Kp5 gale? Luckily for us, two authors named Zawalick and Cage wrote just such a paper (in 1971 in volume 76 of the Journal of Geophysical Research, pages 7009-7012) for the period 1932-1970. If we use their Kp frequency table as a reasonably accurate reflection of the likelihood of Kp reaching various levels, then we can conclude that Kp levels reach 5 about 8% of the time, whereas they reach 7 about 1% of the time. In other words, based on this calculation, visible aurora were about 8 times more likely in the 19th century than today. Of course, these calculations are frighteningly crude and, as the above narrative suggests, many other factors also influence the abundance of visible aurora, including sun spots, light pollution, weather, etc. However, it seems likely that wandering magnetic poles may have played a part in the 19th century abundance of aurora reports.

Indeed, some evidence in the historical accounts suggests, we may have been, geomagnetically speaking, even further north during 1830-1850: the summary of aurora records for that period states that they were seen in New York on about 50 nights per year; that would be about 14% of the nights. If that’s an accurate estimate (and it may not be), it would suggest that aurora were visible at Kp values of between 4 and 5, suggesting that our “corrected geomagnetic latitude” was perhaps closer to 57°N. Got that?

8) ^ For more on the history of NYC lighting, see this paper.

9) ^ Finally, those of you who have read this far deserve a treat, one more image (below) and this, my favorite aurora forecast web page. Enjoy, happy aurora hunting and please let us know what you see!


An engraving perhaps from New York State, from page 271 of American Progress or, The Great Events of the Greatest Century, published in 1890 by Richard Devens and available on-line through

An engraving, perhaps from New York State, from page 271 of the humbly titled American Progress or, The Great Events of the Greatest Century, published in 1890 by Richard Devens and available on-line through See pages 269-275 of that book for an exciting historical narrative of the November 1837 aurora.


A big thanks to Josh Rigler of USGS and Rob Steenburgh of NOAA for helping me piece this together.