An aurora (plural: auroras),[a] sometimes referred to as polar lights, northern lights (aurora borealis) or southern lights (aurora australis), is a natural light display in the Earth's sky, predominantly seen in the high latitude regions (around the Arctic and Antarctic).
Auroras are produced when the magnetosphere is sufficiently disturbed by the solar wind that the trajectories of charged particles in both solar wind and magnetospheric plasma, mainly in the form of electrons and protons, precipitate them into the upper atmosphere (thermosphere/exosphere) due to Earth's magnetic field, where their energy is lost.
The resulting ionization and excitation of atmospheric constituents emits light of varying color and complexity. The form of the aurora, occurring within bands around both polar regions, is also dependent on the amount of acceleration imparted to the precipitating particles. Precipitating protons generally produce optical emissions as incident hydrogen atoms after gaining electrons from the atmosphere. Proton auroras are usually observed at lower latitudes.
Aurora came from the Latin word for "dawn, morning light", since auroras were formerly thought to be the first light of dawn.
Most auroras occur in a band known as the auroral zone, which is typically 3° to 6° wide in latitude and between 10° and 20° from the geomagnetic poles at all local times (or longitudes), most clearly seen at night against a dark sky. A region that currently displays an aurora is called the auroral oval, a band displaced towards the nightside of the Earth. Early evidence for a geomagnetic connection comes from the statistics of auroral observations. Elias Loomis (1860), and later Hermann Fritz (1881) and S. Tromholt (1882) in more detail, established that the aurora appeared mainly in the auroral zone. Day-to-day positions of the auroral ovals are posted on the internet.
In northern latitudes, the effect is known as the Aurora Borealis or the Northern Lights. The former term was coined by Galileo in 1619, from the Roman goddess of the dawn and the Greek name for the north wind. The southern counterpart, the Aurora Australis or the Southern Lights, has features that are almost identical to the Aurora Borealis and changes simultaneously with changes in the northern auroral zone. The Aurora Australis is visible from high southern latitudes in Antarctica, Chile, Argentina, New Zealand, and Australia.
A geomagnetic storm causes the auroral ovals (north and south) to expand, and bring the aurora to lower latitudes. It was hardly ever seen near the geographic pole, which is about 2000 km away from the magnetic pole. The instantaneous distribution of auroras ("auroral oval") is slightly different, being centered about 3-5 degrees nightward of the magnetic pole, so that auroral arcs reach furthest toward the equator when the magnetic pole in question is in between the observer and the Sun. The aurora can be seen best at this time, which is called magnetic midnight.
Auroras seen within the auroral oval may be directly overhead, but from farther away they illuminate the poleward horizon as a greenish glow, or sometimes a faint red, as if the Sun were rising from an unusual direction. Auroras also occur poleward of the auroral zone as either diffuse patches or arcs, which can be sub-visual.
Auroras are occasionally seen in latitudes below the auroral zone, when a geomagnetic storm temporarily enlarges the auroral oval. Large geomagnetic storms are most common during the peak of the eleven-year sunspot cycle or during the three years after the peak. An aurora may appear overhead as a "corona" of rays, radiating from a distant and apparent central location, which results from perspective. An electron spirals (gyrates) about a field line at an angle that is determined by its velocity vectors, parallel and perpendicular, respectively, to the local geomagnetic field vector B. This angle is known as the "pitch angle" of the particle. The distance, or radius, of the electron from the field line at any time is known as its Larmor radius. The pitch angle increases as the electron travels to a region of greater field strength nearer to the atmosphere. Thus it is possible for some particles to return, or mirror, if the angle becomes 90 degrees before entering the atmosphere to collide with the denser molecules there. Other particles that do not mirror enter the atmosphere and contribute to the auroral display over a range of altitudes. Other types of auroras have been observed from space, e.g."poleward arcs" stretching sunward across the polar cap, the related "theta aurora", and "dayside arcs" near noon. These are relatively infrequent and poorly understood. There are other interesting effects such as flickering aurora, "black aurora" and sub-visual red arcs. In addition to all these, a weak glow (often deep red) observed around the two polar cusps, the field lines separating the ones that close through the Earth from those that are swept into the tail and close remotely.
The altitudes where auroral emissions occur were revealed by Carl Størmer and his colleagues who used cameras to triangulate more than 12,000 auroras. They discovered that most of the light is produced between 90 and 150 km above the ground, while extending at times to more than 1000 km. Images of auroras are significantly more common today than in the past due to the increase in use of digital cameras that have high enough sensitivities. Film and digital exposure to auroral displays is fraught with difficulties. Due to the different color spectrum present, and the temporal changes occurring during the exposure, the results are somewhat unpredictable. Different layers of the film emulsion respond differently to lower light levels, and choice of film can be very important. Longer exposures superimpose rapidly changing features, and often blanket the dynamic attribute of a display. Higher sensitivity creates issues with graininess.
David Malin pioneered multiple exposure using multiple filters for astronomical photography, recombining the images in the laboratory to recreate the visual display more accurately. For scientific research, proxies are often used, such as ultra-violet, and color-correction to simulate the appearance to humans. Predictive techniques are also used, to indicate the extent of the display, a highly useful tool for aurora hunters. Terrestrial features often find their way into aurora images, making them more accessible and more likely to be published by major websites. It is possible to take excellent images with standard film (using ISO ratings between 100 and 400) and a single-lens reflex camera with full aperture, a fast lens (f1.4 50 mm, for example), and exposures between 10 and 30 seconds, depending on the aurora's brightness.
Auroras frequently appear either as a diffuse glow or as "curtains" that extend approximately in the east-west direction. At some times, they form "quiet arcs"; at others they evolve and change constantly. These are called "active aurora".
The most distinctive and brightest are the curtain-like auroral arcs. Each curtain consists of many parallel rays, each lined up with the local direction of the magnetic field, consistent with auroras being shaped by Earth's magnetic field. In-situ particle measurements confirm that auroral electrons are guided by the geomagnetic field, and spiral around them while moving toward Earth. The similarity of an auroral display to curtains is often enhanced by folds within the arcs. Arcs can fragment or 'break-up' into separate, at times rapidly changing, often rayed features that may fill the whole sky. These are the 'discrete' auroras, which are at times bright enough to read a newspaper by at night. and can display rapid sub-second variations in intensity. The 'diffuse' aurora, on the other hand, is a relatively featureless glow sometimes close to the limit of visibility. It can be distinguished from moonlit clouds by the fact that stars can be seen undiminished through the glow. Diffuse auroras are often composed of patches whose brightness exhibits regular or near-regular pulsations. The pulsation period can be typically many seconds, so is not always obvious. Often there black aurora i.e. narrow regions in diffuse aurora with reduced luminosity. A typical auroral display consists of these forms appearing in the above order throughout the night.
In addition, the aurora and associated currents produce a strong radio emission around 150 kHz known as auroral kilometric radiation AKR, discovered in 1972. Ionospheric absorption makes AKR only observable from space. X-ray emissions, originating from the particles associated with auroras, have also been detected.
Aurora noise, similar to a hissing, or crackling noise, begins approximately 70 meters (230 feet) above the Earth's surface and is caused by charged particles in an inversion layer of the atmosphere formed during a cold night. The charged particles discharge when particles from the Sun hit the inversion layer creating the noise.
A full understanding of the physical processes which lead to different types of auroras is still incomplete, but the basic cause involves the interaction of the solar wind with the Earth's magnetosphere. The varying intensity of the solar wind produces effects of different magnitudes, but includes one or more of the following physical scenarios.
The details of these phenomena are not fully understood. However it is clear that the prime source of auroral particles is the solar wind feeding the magnetosphere, the reservoir containing the radiation zones, and temporarily magnetically trapped, particles confined by the geomagnetic field, coupled with particle acceleration processes. Furthermore, in 2016 more than fifty citizen science observations (under the aegis of Aurorasaurus) observed a heretofore unknown type of aurora which was dubbed "STEVE" (a reference to the animated film Over the Hedge in which one of the characters randomly names a previously unknown creature "Steve"); subsequent reportage went viral, and STEVE remains under active scientific investigation.
The immediate cause of the ionization and excitation of atmospheric constituents leading to auroral emissions was discovered in 1960, when a pioneering rocket flight from Fort Churchill in Canada revealed a flux of electrons entering the atmosphere from above. Since then an extensive collection of measurements has been acquired painstakingly and with steadily improving resolution since the 1960s by many research teams using rockets and satellites to traverse the auroral zone. The main findings have been that auroral arcs and other bright forms are due to electrons that have been accelerated during the final few 10,000 km or so of their plunge into the atmosphere. These electrons often, but not always, exhibit a peak in their energy distribution, and are preferentially aligned along the local direction of the magnetic field. Electrons mainly responsible for diffuse and pulsating auroras have, in contrast, a smoothly falling energy distribution, and an angular (pitch-angle) distribution favouring directions perpendicular to the local magnetic field. Pulsations were discovered to originate at or close to the equatorial crossing point of auroral zone magnetic field lines. Protons are also associated with auroras, both discrete and diffuse.
Auroras result from emissions of photons in the Earth's upper atmosphere, above 80 km (50 mi), from ionized nitrogen atoms regaining an electron, and oxygen atoms and nitrogen based molecules returning from an excited state to ground state. They are ionized or excited by the collision of particles precipitated into the atmosphere. Both incoming electrons and protons may be involved. Excitation energy is lost within the atmosphere by the emission of a photon, or by collision with another atom or molecule:
Oxygen is unusual in terms of its return to ground state: it can take three quarters of a second to emit green light and up to two minutes to emit red. Collisions with other atoms or molecules absorb the excitation energy and prevent emission. Because the highest atmosphere has a higher percentage of oxygen and is sparsely distributed such collisions are rare enough to allow time for oxygen to emit red. Collisions become more frequent progressing down into the atmosphere, so that red emissions do not have time to happen, and eventually even green light emissions are prevented. This is why there is a color differential with altitude; at high altitudes oxygen red dominates, then oxygen green and nitrogen blue/red, then finally nitrogen blue/red when collisions prevent oxygen from emitting anything. Green is the most common color. Then comes pink, a mixture of light green and red, followed by pure red, then yellow (a mixture of red and green), and finally, pure blue.
Bright auroras are generally associated with Birkeland currents (Schield et al., 1969; Zmuda and Armstrong, 1973), which flow down into the ionosphere on one side of the pole and out on the other. In between, some of the current connects directly through the ionospheric E layer (125 km); the rest ("region 2") detours, leaving again through field lines closer to the equator and closing through the "partial ring current" carried by magnetically trapped plasma. The ionosphere is an ohmic conductor, so some consider that such currents require a driving voltage, which an, as yet unspecified, dynamo mechanism can supply. Electric field probes in orbit above the polar cap suggest voltages of the order of 40,000 volts, rising up to more than 200,000 volts during intense magnetic storms. In another interpretation the currents are the direct result of electron acceleration into the atmosphere by wave/particle interactions.
Ionospheric resistance has a complex nature, and leads to a secondary Hall current flow. By a strange twist of physics, the magnetic disturbance on the ground due to the main current almost cancels out, so most of the observed effect of auroras is due to a secondary current, the auroral electrojet. An auroral electrojet index (measured in nanotesla) is regularly derived from ground data and serves as a general measure of auroral activity. Kristian Birkeland deduced that the currents flowed in the east-west directions along the auroral arc, and such currents, flowing from the dayside toward (approximately) midnight were later named "auroral electrojets" (see also Birkeland currents).
The Earth is constantly immersed in the solar wind, a rarefied flow of hot plasma (a gas of free electrons and positive ions) emitted by the Sun in all directions, a result of the two-million-degree temperature of the Sun's outermost layer, the corona. The solar wind reaches Earth with a velocity typically around 400 km/s, a density of around 5 ions/cm3 and a magnetic field intensity of around 2-5 nT (for comparison, Earth's surface field is typically 30,000-50,000 nT). During magnetic storms, in particular, flows can be several times faster; the interplanetary magnetic field (IMF) may also be much stronger. Joan Feynman deduced in the 1970s that the long-term averages of solar wind speed correlated with geomagnetic activity. Her work resulted from data collected by the Explorer 33 spacecraft. The solar wind and magnetosphere consist of plasma (ionized gas), which conducts electricity. It is well known (since Michael Faraday's work around 1830) that when an electrical conductor is placed within a magnetic field while relative motion occurs in a direction that the conductor cuts across (or is cut by), rather than along, the lines of the magnetic field, an electric current is induced within the conductor. The strength of the current depends on a) the rate of relative motion, b) the strength of the magnetic field, c) the number of conductors ganged together and d) the distance between the conductor and the magnetic field, while the direction of flow is dependent upon the direction of relative motion. Dynamos make use of this basic process ("the dynamo effect"), any and all conductors, solid or otherwise are so affected, including plasmas and other fluids. The IMF originates on the Sun, linked to the sunspots, and its field lines (lines of force) are dragged out by the solar wind. That alone would tend to line them up in the Sun-Earth direction, but the rotation of the Sun angles them at Earth by about 45 degrees forming a spiral in the ecliptic plane), known as the Parker spiral. The field lines passing Earth are therefore usually linked to those near the western edge ("limb") of the visible Sun at any time. The solar wind and the magnetosphere, being two electrically conducting fluids in relative motion, should be able in principle to generate electric currents by dynamo action and impart energy from the flow of the solar wind. However, this process is hampered by the fact that plasmas conduct readily along magnetic field lines, but less readily perpendicular to them. Energy is more effectively transferred by temporary magnetic connection between the field lines of the solar wind and those of the magnetosphere. Unsurprisingly this process is known as magnetic reconnection. As already mentioned, it happens most readily when the interplanetary field is directed southward, in a similar direction to the geomagnetic field in the inner regions of both the north magnetic pole and south magnetic pole.
Earth's magnetosphere is shaped by the impact of the solar wind on the Earth's magnetic field. This forms an obstacle to the flow, diverting it, at an average distance of about 70,000 km (11 Earth radii or Re), producing a bow shock 12,000 km to 15,000 km (1.9 to 2.4 Re) further upstream. The width of the magnetosphere abreast of Earth, is typically 190,000 km (30 Re), and on the night side a long "magnetotail" of stretched field lines extends to great distances (> 200 Re). The high latitude magnetosphere is filled with plasma as the solar wind passes the Earth. The flow of plasma into the magnetosphere increases with additional turbulence, density and speed in the solar wind. This flow is favoured by a southward component of the IMF, which can then directly connect to the high latitude geomagnetic field lines. The flow pattern of magnetospheric plasma is mainly from the magnetotail toward the Earth, around the Earth and back into the solar wind through the magnetopause on the day-side. In addition to moving perpendicular to the Earth's magnetic field, some magnetospheric plasma travels down along the Earth's magnetic field lines, gains additional energy and loses it to the atmosphere in the auroral zones. The cusps of the magnetosphere, separating geomagnetic field lines that close through the Earth from those that close remotely allow a small amount of solar wind to directly reach the top of the atmosphere, producing an auroral glow. On 26 February 2008, THEMIS probes were able to determine, for the first time, the triggering event for the onset of magnetospheric substorms. Two of the five probes, positioned approximately one third the distance to the moon, measured events suggesting a magnetic reconnection event 96 seconds prior to auroral intensification.
Geomagnetic storms that ignite auroras may occur more often during the months around the equinoxes. It is not well understood, but geomagnetic storms may vary with Earth's seasons. Two factors to consider are the tilt of both the solar and Earth's axis to the ecliptic plane. As the Earth orbits throughout a year, it experiences an interplanetary magnetic field (IMF) from different latitudes of the Sun, which is tilted at 8 degrees. Similarly, the 23 degree tilt of the Earth's axis about which the geomagnetic pole rotates with a diurnal variation, changes the daily average angle that the geomagnetic field presents to the incident IMF throughout a year. These factors combined can lead to minor cyclical changes in the detailed way that the IMF links to the magnetosphere. In turn, this affects the average probability of opening a door through which energy from the solar wind can reach the Earth's inner magnetosphere and thereby enhance auroras.
The electrons responsible for the brightest forms of aurora are well accounted for by their acceleration in the dynamic electric fields of plasma turbulence encountered during precipitation from the magnetosphere into the auroral atmosphere. In contrast, static electric fields are unable to transfer energy to the electrons due to their conservative nature. The electrons and ions that cause the diffuse aurora appear not to be accelerated during precipitation. The increase in strength of magnetic field lines towards the Earth creates a 'magnetic mirror' that turns back many of the downward flowing electrons. The bright forms of auroras are produced when downward acceleration not only increases the energy of precipitating electrons but also reduces their pitch angles (angle between electron velocity and the local magnetic field vector). This greatly increases the rate of deposition of energy into the atmosphere, and thereby the rates of ionisation, excitation and consequent emission of auroral light. Acceleration also increases the electron current flowing between the atmosphere and magnetosphere.
One early theory proposed for the acceleration of auroral electrons is based on an assumed static, or quasi-static, electric field creating a uni-directional potential drop. No mention is provided of either the necessary space-charge or equi-potential distribution, and these remain to be specified for the notion of acceleration by double layers to be credible. Fundamentally, Poisson's equation indicates that there can be no configuration of charge resulting in a net potential drop. Inexplicably though, some authors still invoke quasi-static parallel electric fields as net accelerators of auroral electrons, citing interpretations of transient observations of fields and particles as supporting this theory as firm fact. In another example, there is little justification given for saying 'FAST observations demonstrate detailed quantitative agreement between the measured electric potentials and the ion beam energies...., leaving no doubt that parallel potential drops are a dominant source of auroral particle acceleration'.
Another theory is based on acceleration by Landau resonance in the turbulent electric fields of the acceleration region. This process is essentially the same as that employed in plasma fusion laboratories throughout the world, and appears well able to account in principle for most - if not all - detailed properties of the electrons responsible for the brightest forms of auroras, above, below and within the acceleration region.
Other mechanisms have also been proposed, in particular, Alfvén waves, wave modes involving the magnetic field first noted by Hannes Alfvén (1942), which have been observed in the laboratory and in space. The question is whether these waves might just be a different way of looking at the above process, however, because this approach does not point out a different energy source, and many plasma bulk phenomena can also be described in terms of Alfvén waves.
Other processes are also involved in the aurora, and much remains to be learned. Auroral electrons created by large geomagnetic storms often seem to have energies below 1 keV, and are stopped higher up, near 200 km. Such low energies excite mainly the red line of oxygen, so that often such auroras are red. On the other hand, positive ions also reach the ionosphere at such time, with energies of 20-30 keV, suggesting they might be an "overflow" along magnetic field lines of the copious "ring current" ions accelerated at such times, by processes different from the ones described above. Some O+ ions ("conics") also seem accelerated in different ways by plasma processes associated with the aurora. These ions are accelerated by plasma waves in directions mainly perpendicular to the field lines. They therefore start at their "mirror points" and can travel only upward. As they do so, the "mirror effect" transforms their directions of motion, from perpendicular to the field line to a cone around it, which gradually narrows down, becoming increasingly parallel at large distances where the field is much weaker.
The discovery of a 1770 Japanese diary in 2017 depicting auroras above the ancient Japanese capital of Kyoto suggested that that storm may have been 7% larger than the Carrington event, which affected telegraph networks. 
The auroras that resulted from the "great geomagnetic storm" on both 28 August and 2 September 1859, however, are thought to be the most spectacular in recent recorded history. In a paper to the Royal Society on 21 November 1861, Balfour Stewart described both auroral events as documented by a self-recording magnetograph at the Kew Observatory and established the connection between the 2 September 1859 auroral storm and the Carrington-Hodgson flare event when he observed that, "It is not impossible to suppose that in this case our luminary was taken in the act." The second auroral event, which occurred on 2 September 1859 as a result of the exceptionally intense Carrington-Hodgson white light solar flare on 1 September 1859, produced auroras, so widespread and extraordinarily bright, that they were seen and reported in published scientific measurements, ship logs, and newspapers throughout the United States, Europe, Japan, and Australia. It was reported by The New York Times that in Boston on Friday 2 September 1859 the aurora was "so brilliant that at about one o'clock ordinary print could be read by the light". One o'clock EST time on Friday 2 September, would have been 6:00 GMT and the self-recording magnetograph at the Kew Observatory was recording the geomagnetic storm, which was then one hour old, at its full intensity. Between 1859 and 1862, Elias Loomis published a series of nine papers on the Great Auroral Exhibition of 1859 in the American Journal of Science where he collected worldwide reports of the auroral event.
That aurora is thought to have been produced by one of the most intense coronal mass ejections in history. It is also notable for the fact that it is the first time where the phenomena of auroral activity and electricity were unambiguously linked. This insight was made possible not only due to scientific magnetometer measurements of the era, but also as a result of a significant portion of the 125,000 miles (201,000 km) of telegraph lines then in service being significantly disrupted for many hours throughout the storm. Some telegraph lines, however, seem to have been of the appropriate length and orientation to produce a sufficient geomagnetically induced current from the electromagnetic field to allow for continued communication with the telegraph operator power supplies switched off. The following conversation occurred between two operators of the American Telegraph Line between Boston and Portland, Maine, on the night of 2 September 1859 and reported in the Boston Traveler:
Boston operator (to Portland operator): "Please cut off your battery [power source] entirely for fifteen minutes."
Portland operator: "Will do so. It is now disconnected."
Boston: "Mine is disconnected, and we are working with the auroral current. How do you receive my writing?"
Portland: "Better than with our batteries on. - Current comes and goes gradually."
Boston: "My current is very strong at times, and we can work better without the batteries, as the aurora seems to neutralize and augment our batteries alternately, making current too strong at times for our relay magnets. Suppose we work without batteries while we are affected by this trouble."
Portland: "Very well. Shall I go ahead with business?"
Boston: "Yes. Go ahead."
The conversation was carried on for around two hours using no battery power at all and working solely with the current induced by the aurora, and it was said that this was the first time on record that more than a word or two was transmitted in such manner. Such events led to the general conclusion that
The effect of the aurorae on the electric telegraph is generally to increase or diminish the electric current generated in working the wires. Sometimes it entirely neutralizes them, so that, in effect, no fluid is discoverable in them. The aurora borealis seems to be composed of a mass of electric matter, resembling in every respect, that generated by the electric galvanic battery. The currents from it change coming on the wires, and then disappear: the mass of the aurora rolls from the horizon to the zenith.
An aurora was described by the Greek explorer Pytheas in the 4th century BC.Seneca wrote about auroras in the first book of his Naturales Quaestiones, classifying them, for instance as pithaei ('barrel-like'); chasmata ('chasm'); pogoniae ('bearded'); cyparissae ('like cypress trees'), and describing their manifold colors. He wrote about whether they were above or below the clouds, and recalled that under Tiberius, an aurora formed above the port city of Ostia that was so intense and red that a cohort of the army, stationed nearby for fire duty, galloped to the rescue. It has been suggested that Pliny the Elder depicted the aurora borealis in his Natural History, when he refers to trabes, chasma, 'falling red flames' and 'daylight in the night'.
In the traditions of Aboriginal Australians, the Aurora Australis is commonly associated with fire. For example, the Gunditjmara people of western Victoria called auroras puae buae ('ashes'), while the Gunai people of eastern Victoria perceived auroras as bushfires in the spirit world. The Dieri people of South Australia say that an auroral display is kootchee, an evil spirit creating a large fire. Similarly, the Ngarrindjeri people of South Australia refer to auroras seen over Kangaroo Island as the campfires of spirits in the 'Land of the Dead'. Aboriginal people in southwest Queensland believe the auroras to be the fires of the Oola Pikka, ghostly spirits who spoke to the people through auroras. Sacred law forbade anyone except male elders from watching or interpreting the messages of ancestors they believed were transmitted through an aurora.
Bulfinch's Mythology relates that in Norse mythology, the armour of the Valkyrior "sheds a strange flickering light, which flashes up over the northern skies, making what Men call the 'aurora borealis', or 'Northern Lights' ". There appears to be no evidence in Old Norse literature to substantiate this assertion. The first Old Norse account of norðrljós is found in the Norwegian chronicle Konungs Skuggsjá from AD 1230. The chronicler has heard about this phenomenon from compatriots returning from Greenland, and he gives three possible explanations: that the ocean was surrounded by vast fires; that the sun flares could reach around the world to its night side; or that glaciers could store energy so that they eventually became fluorescent.
In the 1740s Benjamin Franklin theorized in his paper Aurora Borealis, Suppositions and Conjectures towards forming an Hypothesis for its Explanation that an aurora was caused by a concentration of electrical charge in the polar regions intensified by the snow and moisture in the air:
May not then the great quantity of electricity brought into the polar regions by the clouds, which are condens'd there, and fall in snow, which electricity would enter the earth, but cannot penetrate the ice; may it not, I say (as a bottle overcharged) break thro' that low atmosphere and run along in the vacuum over the air towards the equator, diverging as the degrees of longitude enlarge, strongly visible where densest, and becoming less visible as it more diverges; till it finds a passage to the earth in more temperate climates, or is mingled with the upper air?-- Benjamin Franklin
Observations of the rhythmic movement of compass needles due to the influence of an aurora were confirmed in the Swedish city of Uppsala by Anders Celsius and Olof Hiorter. In 1741, Hiorter was able to link large magnetic fluctuations with an aurora being observed overhead. This evidence helped to support their theory that 'magnetic storms' are responsible for such compass fluctuations.
A variety of Native American myths surround the spectacle. The European explorer Samuel Hearne traveled with Chipewyan Dene in 1771 and recorded their views on the ed-thin ('caribou'). According to Hearne, the Dene people saw the resemblance between an aurora and the sparks produced when caribou fur is stroked. They believed that the lights were the spirits of their departed friends dancing in the sky, and when they shone brightly it meant that their deceased friends were very happy.
During the night after the Battle of Fredericksburg, an aurora was seen from the battlefield. The Confederate Army took this as a sign that God was on their side, as the lights were rarely seen so far south. The painting Aurora Borealis by Frederic Edwin Church is widely interpreted to represent the conflict of the American Civil War.
Both Jupiter and Saturn have magnetic fields that are stronger than Earth's (Jupiter's equatorial field strength is 4.3 gauss, compared to 0.3 gauss for Earth), and both have extensive radiation belts. Auroras have been observed on both gas planets, most clearly using the Hubble Space Telescope, as well as on Uranus and Neptune.
The auroras on the gas giants seem, like Earth's, to be powered by the solar wind. In addition, however, Jupiter's moons, especially Io, are powerful sources of auroras on Jupiter. These arise from electric currents along field lines ("field aligned currents"), generated by a dynamo mechanism due to the relative motion between the rotating planet and the moving moon. Io, which has active volcanism and an ionosphere, is a particularly strong source, and its currents also generate radio emissions, which have been studied since 1955. Using the Hubble Space Telescope, auroras over Io, Europa and Ganymede have all been observed.
Auroras have also been observed on Venus and Mars. Venus has no magnetic field and so Venusian auroras appear as bright and diffuse patches of varying shape and intensity, sometimes distributed over the full disc of the planet. A Venusian aurora originates when electrons from the solar wind collide with the night-side atmosphere.
An aurora was detected on Mars, on 14 August 2004, by the SPICAM instrument aboard Mars Express. The aurora was located at Terra Cimmeria, in the region of 177° East, 52° South. The total size of the emission region was about 30 km across, and possibly about 8 km high. By analyzing a map of crustal magnetic anomalies compiled with data from Mars Global Surveyor, scientists observed that the region of the emissions corresponded to an area where the strongest magnetic field is localized. This correlation indicated that the origin of the light emission was a flux of electrons moving along the crust magnetic lines and exciting the upper atmosphere of Mars.
The first ever extra-solar auroras were discovered in July 2015 over the brown dwarf star LSR J1835+3259. The mainly red aurora was found to be a million times brighter than the Northern Lights, a result of the charged particles interacting with hydrogen in the atmosphere. It has been speculated that stellar winds may be stripping off material from the surface of the brown dwarf to produce its own electrons. Another possible explanation for the auroras is that an as-yet-undetected body around the dwarf star is throwing off material, as is the case with Jupiter and its moon Io.