Frames are displayed as they are downloaded.
The Aurora Borealis, often referred to as the Northern Lights, is an inspiring display of red, pink, blue, light green, and white lights typically visible in the nighttime sky at high latitudes. These displays are commonplace in Alaska, Canada, and the northernmost states of the Lower Forty-Eight. Aurora are also occasionally visible farther south during times of geomagnetic activity. More info on seeing the aurora is available at "The Auroral Page" or through the auroral activity forecast Web page at the Geophysical Institute, University of Alaska.
What causes the aurora? For some time now, we have known that the primary means by which aurora are produced is through the bombardment of the Earth's upper atmosphere by high-speed electrons which flow down the Earth's magnetic field lines, as shown in the following Java animation:
How and where do these electrons get their high energy? This is the big question, the question which has not yet been definitively answered. There are a number of things we do know, leading, of course, to other questions whose answers we don't know. It is clear, for example, that the electrons are getting their energy ultimately from the solar wind (high speed particles which are constantly emitted by the Sun). We don't know, though, how this energy is transferred to these "auroral" electrons. The process cannot be a simple one, because the Earth is, for the most part, shielded by its magnetic field environment, called the magnetosphere. We think that some fraction of the electrons responsible for the nighttime aurora gain their energy in two regions of the magnetosphere called the plasma sheet and the plasma sheet boundary layer, Again, however, there is controversy. These regions are thought to be where the electrons are energized, because, if the magnetic field lines which pass through the auroral displays are followed out away from the Earth, many think that these regions are where the field lines end up. This ought to be where the electrons come from, then, since electrons tend to flow along magnetic field lines. The problem is, magnetic field lines are invisible, so we can't tell for sure where the field lines go. Additionally, it is likely that many of the auroral electrons gain energy in a region closer to where the aurora appear, called the auroral acceleration region. This region is again on the field lines which pass the through the auroral displays but is much closer to the Earth: between one and two Earth radii (1-2 Re) above the Earth (4000 to 8000 km altitude), whereas the other regions are 6 Re or more away. There appears to be an electric field in this region, oriented parallel to the magnetic field. Its existence is also disputed, although acceptance has grown considerably in the last decade or so. Clearly, this electric field, if it exists, will tend to accelerate electrons as they flow towards the Earth. The physics of the auroral acceleration region is the central interest of our research. There are, again several areas of confusion. First of all, not only can the existence of a parallel electric field lead to accelerated electrons, accelerated electrons can also produce an parallel electric field. This happens through a process generally referred to as "anomalous resistivity." When the electrons reach a certain speed after having been accelerated, they cause instabilities such as the ion acoustic instability and the lower hybrid instability. The turbulence created by these instabilities looks like electrical resistivity to the flow of the electrons. When current flows through a resistance, Ohm's law says there exists a potential drop. This potential drop is the origin of the parallel electric field.
All this raises the possbility that the electric field is not causing acceleration at all; instead, it might be other way around. Consideration of this problem leads to the second problem: if electric fields are producing acceleration, what's generating the electric field? If accelerated electrons are generating the parallel electric field, we're back to the same question: what's accelerating the electrons?
Physicists are trained to respond in a certain way when confronted with situations such as this, where the equation for "A" depends on "B" and vice versa. We solve both equations -simultaneously-. When terms are substituted for "A" and "B", in this case, "the parallel electric field" and "the parallel electron dynamics," the term typically used is "self-consistency." We solve for the parallel electric field and electron acceleration simultaneously, in a self-consistent fashion. Actually trying this leads to another problem, though: nothing happens! That is, when a self-consistent simulation in initialized with no electric field and no net electron drift, neither is subsequently generated.
The solution to this problem is actually fairly obvious. What's missing is a -source-. We shouldn't expect anything to happen if we don't have anything in the model to represent the energy input from the solar wind. Earlier studies have included sources by variously imposing voltage drops across, or constant current boundary conditions on, the simulation system. These constraints do indeed represent sources, but since they are imposed, they are not being determined -self-consistently-. Such simulations thus cannot answer the question of what is causing what.
We have attempted to improve on this situation by reasoning as follows. We don't know the details of what the source should be, because, as pointed out above, there is still much debate as to how energy is transferred from the solar wind to various parts of the magnetosphere, including the auroral acceleration region.
(To be continued...)
| Here's a phase space movie from a particle simulation of a propagating Alfven wave in the auroral acceleration region: |
|
|
Alfvén wave movie (641K) Frames are displayed as they are downloaded. |
