Is the Aurora There When No One Is Looking?

Presented bilingually. Click to toggle: Scientific Presentation Plain English

(Nature paper published June 24, 1996)

For centuries scientists and amatuers have been studying the northern lights. Like the stars, they could only be seen in darkness, but were assumed to still be there although hidden in sunlight. Our research results show that by and large auroral displays impressive enough to be seen with the naked eye actually only occur in darkness. In other words, when the sun is out, and neither scientist nor other admirer of the spectacle is watching, the northern lights are much rarer (about 1/3 as common).

What causes the aurora:

First of all there are actually two types of aurora, diffuse and discrete. The Earth is surrounded by hot ions and electrons, at about 1-10 million degrees. The energy to heat these charged particles comes from the solar wind interacting with the Earth's magnetic field. As these particles meander around in space (although generally moving along the Earth's magnetic field lines) some of them strike the upper atmosphere, causing a glow just as in florescent tube. Because electrons are so much lighter than ions, they move faster, and cause most of the aurora. This type of faint glow is the diffuse aurora, and it is always present.

However, this background process does not constitute the spectacular show known as the northern lights. Although 1-10 million degrees might sound hot, most displays visible from the ground involve electron beams that are still more energetic than that, and that are highly localized. The result is narrow curtains or ribbons of light that are highly dynamic. There are a great many theories on what causes the discrete aurorae, but no consensus. Certain facts are known, such as that strong electric fields appear above the ionosphere (about 5000-15000 miles up) causing the acceleration, but this just pushes the issue back a step: what causes these electric fields?

How can you tell whether the aurora is there in sunlight?

By studying the electron beams that cause the aurora, instead of the aurora itself. The Air Force has its own private set of weather satellites. Because the Air Force is also interested in space weather, these satellites include detectors which measure the hot particles surrounding the Earth (Dr. D. Hardy of Phillips Laboratory built and calibrated them). When electrons have been accelerated by electric fields it leaves a characteristic signature in their spectrum. Most electrons will have an energy close to the voltage drop associated with the electric field.

We studied 152 million measurements of the electron population taken by 5 satellites over a 9 year interval, and counted every case of acceleration by electric fields. This is what we found for their distribution:

The distribution of discrete aurora. Local time varies around the circle (noon is at top). Latitudes increase towards the center of the circle, which is the magnetic pole.

Only at night?

Well the last figure shows that discrete aurora can be found at any local time, including daytime. However this plot counts all electron acceleration events, including those much too weak to be visible to the eye. What follows is a plot of the probability of electron acceleration events intense enough to produce real northern lights:

The distribution of intense discrete aurora.

Notice that the intense aurora is highly concentrated in the dusk-to-midnight sector. (As a consequence, already appreciated by many who live at high latitudes, there is little point in staying up long past midnight if you want to see a good auroral show). But if electron acceleration events occur everywhere, at all times of the day, why are these intense events which produce the northern lights so concentrated in the dusk-to-midnight sector?

Also, how do we know sunlight itself has any effect?

To answer the last question, we separated the observations taken in sunlight from those in darkness. Remember that the aurora occurs at high latitudes, so that for example in the summer hemisphere it might be sunlit even at midnight, or in the winter hemisphere in darkness even at noon. So here is a direct comparison of the probability of observing intense aurora in darkness and sunlight respectively:

The probability of finding northern lights in darkness

The probability of finding northern lights in sunlight

Darkness wins by a wide margin (a factor of 3 in the peak dusk-to-midnight sector).

So how do you explain it all?

Currents (indirectly driven by the solar wind) want to flow between near-Earth space and the ionosphere. But for a current to flow, conductivity must exist. This conductivity can be created in two ways, from ultra-violet sunlight and from the diffuse aurora. Below is a figure giving our estimate of the typical conductivity around equinox from these two sources:

The conductivity of the ionosphere with currents from space superposed (red lines).

The only place where conductivity is low, but the currents are required by solar wind-Earth interaction is from dusk to dawn. Therefore this is the region that is unstable to the creation of discrete aurora. Once the intense northern lights do appear, they themselves create the conductivity needed to solve the problem, and allow the currents to flow.

Return to Auroral Particles and Imagery Page

Return to Main Stories Page

Examine Available Data Sets

Send science questions/comments to Dr. P. Newell
Send WWW questions/comments to webmaster