Mysterious electron acceleration explained

Feb 27, 2012 by David L. Chandler
An artist's rendition of Earth's magnetosphere. A magnetic tail, or magnetotail, is formed by pressure from the solar wind on a planet's magnetosphere. Image: NASA

A mysterious phenomenon detected by space probes has finally been explained, thanks to a massive computer simulation that was able to precisely align with details of spacecraft observations. The finding could not only solve an astrophysical puzzle, but might also lead to a better ability to predict high-energy electron streams in space that could damage satellites.

Jan Egedal, an associate professor of physics at MIT and a researcher at the Plasma Science and Fusion Center, working with MIT graduate student Ari Le and with William Daughton of the Los Alamos National Laboratory (LANL), report on this solution to the space conundrum in a paper published Feb. 26 in the journal .

Egedal had initially proposed a theory to explain this large-scale acceleration of electrons in Earth’s magnetotail — a vast and intense magnetic field swept outward from Earth by the solar wind — but until the new data was obtained from the computer simulation, “it used to be people said this was a crazy idea,” Egedal says. Thanks to the new data, “I don’t get that anymore,” he says.

The simulation shows that an active region in Earth’s magnetotail, where “reconnection” events take place in the magnetic field, is roughly 1,000 times larger than had been thought. This means a volume of space energized by these magnetic events is sufficient to explain the large numbers of high-speed electrons detected by a number of spacecraft missions, including the Cluster mission.

Solving the problem required a staggering amount of computer power from one of the world’s most advanced supercomputers, at the National Institute for Computational Science at Oak Ridge National Laboratory in Tennessee. The computer, called Kraken, has 112,000 processors working in parallel and consumes as much electricity as a small town. The study used 25,000 of these processors for 11 days to follow the motions of 180 billion simulated particles in space over the course of a magnetic reconnection event, Egedal says. The processing time accumulated gradually, squeezed in during idle time between other tasks. The simulation was performed using a plasma-physics code developed at LANL that rigorously analyzes the evolution of magnetic reconnection.

Egedal explains that as the solar wind stretches Earth’s magnetic-field lines, the field stores energy like a rubber band being stretched. When the parallel field lines suddenly reconnect, they release that energy all at once — like releasing the rubber band. That release of energy is what propels electrons with great energy (tens of thousands of volts) back toward Earth, where they impact the upper atmosphere. This impact is thought, directly or indirectly, to generate the glowing upper-atmosphere plasma called the aurora, producing spectacular displays in the night sky.

What had puzzled physicists is the number of energetic electrons generated in such events. According to theory, it should be impossible to sustain an electric field along the direction of the magnetic field lines, because the plasma (electrically charged gas) in the magnetotail should be a near-perfect conductor. But such a field is just what’s needed to accelerate the electrons. And, according to the new simulation, the volume of space where such fields can build up can, in fact, be at least 1,000 times larger than the theorists had thought possible — and thus large enough to explain the observed electrons.

“People have been thinking this region is tiny,” Egedal says. But now, “by analyzing the spacecraft data and doing the simulation, we’ve shown it can be very large, and can accelerate many electrons.” As a result, “for the first time, we can reproduce the features” observed by the Cluster spacecraft.

That could be important because, among other things, “these hot can destroy spacecraft,” Egedal says, which is why both the military and NASA “would like to understand this better.”

Although this analysis was specific to the phenomena in Earth’s magnetotail, Egedal says similar phenomena may be taking place in much bigger regions of magnetized plasma in space — such as in mass ejections that erupt from the sun’s corona, which occupy regions 10,000 times larger, or even regions surrounding pulsars or other high-energy objects in deep space, which are much larger still. In the future, he hopes to carry out simulations that would apply to the sun’s coronal mass ejections. “We think we can scale up the simulation” by a hundredfold, he says.

Michael Brown, a professor of physics at Swarthmore College who was not involved in this research, says Egedal “is emerging as a real leader in experimental [and] observational aspects of magnetic reconnection,” and his co-author Daughton “is the recognized leader in state-of-the-art simulations.” The new result “is very significant, and I think is surprising to the rest of the community. … I think this picture will gain more and more acceptance, and we have to go beyond” the presently accepted picture of plasmas, he says.

The work was supported by grants from NASA and the National Science Foundation.

Explore further: New microscope collects dynamic images of the molecules that animate life

Related Stories

Ultrafast substorm auroras explained (w/ video)

Sep 06, 2011

(PhysOrg.com) -- From time to time, sudden releases of energy in Earth's magnetosphere lead to major disturbances that result in bright auroral displays over the planet's polar regions. These auroras are caused by a phenomenon ...

A tool to touch the sun

Oct 25, 2011

A scientist at the Harvard-Smithsonian Center for Astrophysics (CfA) is leading an effort to design an instrument that would be the first to come directly into contact with the sun’s fiery atmosphere ...

Voyager 1 measures magnetic mayhem

Jun 01, 2011

When Voyager 1 passed into the heliosheath in 2004, it became the first man-made object to explore the remote edge of the Sun’s magnetic influence. Launched by NASA on September 5, 1977, the probe was ...

Recommended for you

Cooling with molecules

Oct 22, 2014

An international team of scientists have become the first ever researchers to successfully reach temperatures below minus 272.15 degrees Celsius – only just above absolute zero – using magnetic molecules. ...

User comments : 18

Adjust slider to filter visible comments by rank

Display comments: newest first

ichigo kurosaki
4 / 5 (4) Feb 27, 2012
damn.i love this science.......
Kinedryl
1.6 / 5 (8) Feb 27, 2012
computer, called Kraken, has 112,000 processors working in parallel and consumes as much electricity as a small town. The study used 25,000 of these processors for 11 days
It's nice, but one Tesla card has a computational power of 512 CPUs in parallel and its consumption doesn't differ from common graphic card so much. You can have as many as 8 Tesla cards in one chasis, which contains ~ 2000 CPUs after then and its consumption (800 W) doesn't exceed the average consumption of microwave owen. Which calls the effectiveness of Kraken computation facility into question. http://www.geeks3...mputing/
tkjtkj
1 / 5 (2) Feb 27, 2012
re: "..Solving the problem required a staggering amount of computer power from one of the world's most advanced supercomputers, at the National Institute for Computational Science at Oak Ridge..."
Minor point perhaps, but those who edit so many of these physorg articles do need some slaps on the wrist!
In this case, stating that the sim *required* a 112,000 cpu Kraken (where in reality it only used less than 25% of that number ) is just dumb.
antialias_physorg
4.9 / 5 (12) Feb 27, 2012
It's nice, but one Tesla card has a computational power of 512 CPUs

CPU power alone is not where it's at. It's also memory and interconnectivity.

The Kraken Specs look like this:
Cray Linux Environment (CLE) 2.2
A peak performance of 1.17 PetaFLOP
112,896 compute cores
147 TB of compute memory
A 3.3 PB raw parallel file system of disk storage for scratch space (2.4 PB available)
9,408 compute nodes

Each node has:
Two 2.6 GHz six-core AMD Opteron processors (Istanbul)
12 cores
16 GB of memory
Connection via Cray SeaStar2 router


Comparing that to just putting up a few boxes with great graphics cards is...naive.
Deathclock
3.9 / 5 (7) Feb 27, 2012
computer, called Kraken, has 112,000 processors working in parallel and consumes as much electricity as a small town. The study used 25,000 of these processors for 11 days
It's nice, but one Tesla card has a computational power of 512 CPUs in parallel and its consumption doesn't differ from common graphic card so much. You can have as many as 8 Tesla cards in one chasis, which contains ~ 2000 CPUs after then and its consumption (800 W) doesn't exceed the average consumption of microwave owen. Which calls the effectiveness of Kraken computation facility into question. http://www.geeks3...mputing/


So... you're saying you don't know anything about supercomputers?

*edit* AA beat me to it ;)
holoman
5 / 5 (9) Feb 27, 2012
The article is about magnetic fields, plasma, and the
acceleration of ionized particles.

The number of computers and speeds necessary for the simulation
are less important than the plasma concept published.
Lurker2358
1.7 / 5 (6) Feb 27, 2012
And, according to the new simulation, the volume of space where such fields can build up can, in fact, be at least 1,000 times larger than the theorists had thought possible and thus large enough to explain the observed electrons.


Mainstream theory fail number MTF-2273.1.14b.
mgmirkin
4.2 / 5 (5) Feb 27, 2012
So, what they're saying is "the theory was wrong, magnetic field-aligned electric currents ARE possible." No kidding... Couldn't have told you that!

http://adsabs.har...70.2235H
http://www.geofis...amar.pdf
http://www.agu.or...09.shtml
http://adsabs.har...24..461L

It's been known for a while that magnetic field-aligned electric fields *are* possible and *do* exist... There's just a blind eye to most things electrical in astronomy, et al.

But, let's not let a little REALITY intrude on a perfectly good theory. ;)

This isn't "news" it's just "news to them."

P.S. 'Magnetic Reconnection' is just an exploding electric circuit:
http://plasmafron...tic.html

Done.
mgmirkin
3 / 5 (2) Feb 27, 2012
So, what they're saying is "the theory was wrong, magnetic field-aligned electric currents ARE possible."


Oops, that should say "magnetic field-aligned electric *fields*," NOT *currents*.

Sorry, didn't catch they typo in time to edit the original.

Though "magnetic field-aligned electric currents" are also known to exist and are also known as "Birkeland currents" after Norwegian physicist Kristian Birkeland, who predicted them WRT the auroras & inflowing currents from space, subsequently confirmed by satellite measurement in the 70's...

As you were...
fmfbrestel
4 / 5 (1) Feb 27, 2012
The number of computers and speeds necessary for the simulation
are less important than the plasma concept published.


Yes, thank you. This is an excellent job of scientific team looking at the observations, seeing a model that is failing to explain that observation, and correcting the model.
Lurker2358
1.7 / 5 (6) Feb 27, 2012
Yes, thank you. This is an excellent job of scientific team looking at the observations, seeing a model that is failing to explain that observation, and correcting the model.


Now all the various cosmologists need to have the integrity to re-assess all other major theories and interpretations in space regarding electric current and electric fields to take this into consideration, seeing as how it could have mild to moderate effects on everything from star formation to neutron stars and black hole physics.

It may even provide a modest correction to the estimates of mass for black holes and neutron stars, their jets or other anomalous acceleration events.
StarGazer2011
2 / 5 (4) Feb 27, 2012
so now we have a HYPOTHESIS based on computer simulations which needs to be TESTED with actual physical OBSERVATIONS!
Seriously, treating simulation output as though it was an observation is just getting boring and its totally anti-science.
fmfbrestel
2.3 / 5 (3) Feb 27, 2012
Seriously, treating simulation output as though it was an observation is just getting boring and its totally anti-science.


Serious, people seeing the word simulation and then skipping the rest of the article in order to troll about how horrible simulations are is just getting boring and its totally anti-intelligence.

This means a volume of space energized by these magnetic events is sufficient to explain the large numbers of high-speed electrons detected by a number of spacecraft missions, including the Cluster mission.
Callippo
1 / 5 (1) Feb 27, 2012
The animation of this simulation. Not sure if it's worth of 11 days of electricity consumption of small city though.... http://www.youtub...hOuAdbYM
Lurker2358
1.8 / 5 (5) Feb 27, 2012
so now we have a HYPOTHESIS based on computer simulations which needs to be TESTED with actual physical OBSERVATIONS!
Seriously, treating simulation output as though it was an observation is just getting boring and its totally anti-science.


At least a simulation is better than simply accepting the model or someone's interpretation of the model at face value.
HannesAlfven
1 / 5 (3) Feb 28, 2012
By the way, for the record, laboratory magnetic reconnection experiments stop working the second you shut the electrical power off. You cannot run the experiment without an electrical current, so what sense does it make to ignore them in the theory?
StarGazer2011
1 / 5 (1) Mar 07, 2012
@fmfbrestel: I read the whole article, must have missed the mention of the Cluster Mission. So there is one data point supporting the hypothesis. It still needs more empirical support, specifically a place where the current theory and the new one diverge significantly.
Maybe i do tend to assume simulation is being treated as observation even when its not, must be the climate :P Thanks for pointing it out so politely.
tkjtkj
5 / 5 (1) Mar 07, 2012


At least a simulation is better than simply accepting the model or someone's interpretation of the model at face value.


Has there ever existed any complex sim that did not include its author's biases ?