Expert: Philip Stahl - 10/12/2008

Question
QUESTION: Dear Expert, hello,

a quick question on the physics of auroras please - i just wanted to confirm the following please:

From what i undertsand the auroras are due to very energetic particles, like electrons, intercating with atmospheric elements like nitrogen and oxygen, exiting these atoms, and as these atoms de-excite (drop to a lower energy level) the green, red or purplish lights are emitted. Is that correct?

Now, my question relates to the spatial distribution of the initial energetic particles that caus ethe auroras in the first place. The auroras take all these beuatiful shapes, and "pattens" - do these pattern reflect the positions of the energetic electrons?? Can we say thats were those electrons are? i.e. the large-scale currents are where the auroras are observed?? i.e. the auroras ar ethe manifesattions of these lare-scale currents??thanks for your comments, elsa

ANSWER: Hello, Elsa

First, your general description is basically correct. There is just a little more detail that might be added, such as that the red aurora `is created by emission at the 630 nm (nanometer) line of oxygen and at relatively high altitudes (e.g. 200-600 km) compared to other colored auroras (e.g. green tends to form below 200 km and the oxygen line at 557 nm is excited). The key point here is that the O-atoms possess what we call “metastable decay times, e.g. of ~  200 s and the (collisional) de-excitation times are generally longer.

Another point, is that auroras can display as both “diffuse” and “discrete”. In the first case the shape is ill-defined and the aurora is believed to be formed from trapped particles originally in the magnetosphere which then propagate into the lower ionosphere via wave-particle interactions.

Bear in mind here ionospheric plasma is relatively “cold” (electron temp. < 1 eV) and dense (particle number density up to 10^5/ cm^3) while magnetospheric plasma is relatively hot (T(E) ~  500 eV) and tenuous (n(e) ~  1 /cm^3).

Diffuse auroras– like discrete-  can display assorted colors from green to red. Discrete auroras are those you referenced in the latter part of your question which show discrete shapes.  For example, when in Alaska in March of 2005, I observed a discrete green aurora that displayed lengthwise curtains then some hours later,  two perfectly symmetrical parallel green “tubes”, arcing from north to south horizon.

You asked if the patterns reflect positions of energetic electrons. Yes and “maybe”. The ‘maybe’ enters because we don’t yet have all the answers as to the complexities of specific discrete auroras and their formation. Of course, this will also await a more precise knowledge of where the generating currents are located. What we do know is that the critical “field-aligned currents” close via connection to horizontal currents in the ionosphere. (See again the properties of the ionosophere given above)

Now, bear in mind we have the (pre-existing) presence of the auroral oval surrounding the pole. If you do a google of “auroral oval” you can probably see one of the many space based images of it. The point of interest here is that the ionosopheric currents – horizontal- are distributed at two main places on the oval: the “morning side” (westward current) and the “evening side” (eastward current).

The auroral curtains such as I saw near Fairbanks perhaps stretched for several thousands of miles along the auroral curtain. Current theory postulates the thickness of these curtains as not much more than 1000m- maybe less. Trouble is, no one has an adequate hypothesis for their formation or why they are so predominant in the curtains. (Of course, changes in the field –aligned currents may explain why the curtains I observed slowly changed to “tubes” since the changing currents would elicit changing magnetic fields that could have “rolled” them into that shape. Just a guess!

What truly amazed me, is how the whole of the auroral curtain I observed moved uniformly north, then south in a kind of oscillation. Current thinking is this wasn’t caused by alterations in the distributions of the light-emitting atoms (e.g. oxygen) but rather changing direction of the electron beam itself – so if the electron beam shifts north so does the curtain. If it shifts south, so does the curtain.

A great analogy has been given – I believe by Syun Akasofu- comparing the aurora to images on a TV screen. In this case the (polar) upper atmosphere corresponds to the screen and the aurora to the image that would be projected on it, say for a TV. The electron beam in the TV (remember we are talking about the old-style cathode ray jobs!) corresponds to the electron beam in the magnetosphere. In the conventional TV, motions of the image are generated by the changing impact points of the electron beam on the screen. Similarly, with the aurora, its motions – such as moving sheets or curtains- are produced by moving impact points of the magnetospheric electron beams.

Remember that ultimately the aurora derives its power and potential from the Sun and specifically the charged particles of the solar wind. This is why the most spectacular displays are usually near sunspot maximum. Around those times the currents I noted earlier are “amped” up – no pun intended- to 10^6 A or more. To give an example, during a quiet Sun interval like we are in now the residual power for the magnetospheric generator is on the order of maybe a tenth of a megawatt. If we see a new cycle coming on and solar wind activated – we may get that power up to a million megawatts for a few hours.

Hopefully, your question has been addressed –at least to within the limits of our present knowledge. As you can guess, the aurora is an immensely complex phenomenon and much more on the spot (‘in situ”) data needs to be obtained before we can nail it down, in particular exactly how the sheet-like shapes bend and deform to give the array of discrete auroras we see.


---------- FOLLOW-UP ----------

QUESTION: Dear Sir,

i just wanted to say "Thank you" for the very clear and informative answer - i think i need to study much-more about the currents in both the ionsopshere and magnetosphere - i assume what we know comes from both measurements and simulations - has the thickeness of the aurora " curtains" ever been "measured"?? silly question perhaps - because they chnage very quickly - and therefore cannot be measured, but i was wondering whether some kind of "measurements" from image analysis has been done? or is this a silly question??

many thanks for all your help, its a fascinating topic, many thanks again, Elsa.

Answer
Hello again,

I am not aware of any direct measurements of currents being done, only of simulations (numerical) that have been done - as at the Geophysical Institute, University of Alaska-Fairbanks. What has happened is that various researchers (e.g. S.-I. Akasofu, J. Kan, C. Lee et al) have developed "dynamo" and "substorm" models for auroras, then used parameters from the models to simulate the phenomenon.

Some of the simulations are fairly good, but to be truthful, are only a very approximate method for replicating them. In most cases you don't see the "curtains" performing like they do in real life - but then remember, auroral research has only really been jump started since rocket research carrying sophisticated electronic payloads and indirect measuring devices (e.g. in Alaska, the Poker Flat range has been used for these rocket missions).

You can read some more about Poker Flat and the work done there here:

http://www.pfrr.alaska.edu/data/index.html


New instruments and new devices are being developed all the time, and we await the time a craft totally dedicated to the study of the aurora can be launched. By "dedicated" I mean expend its resources in a 24/7 time profile through at least a solar minimum all the way through a solar maximum.

When that happens, more direct measurements of currents may be forthcoming, and then better modeling.

Btw, no question is "silly". All are worthwhile, and your questions show you have already done some solid homework.

A couple of recommendations:

A good book tracing the history of the discovery of the aurora:

"The Northern Lights" by Lucy Jago

There is also a superb (30 min.) VHS video - "The Aurora Explained" - which had been available the past few years. You might google to see if it still is. It goes into all the basics, basic models, and shows real aurora as they occurred.

If you are unsuccessful in googling, you can write to the producers at:

The Aurora Color Television Project
Geophysical Institute
University of Alaska, Fairbanks  99775-7320

and inquire whether tapes or dvds are still available.

Good luck!