Astronomy/cosmology questions...

[Lorizael]

I guess I can see your point on the aether thing. Isn't the Bell labs radio telescope discovery of the background noise from The Big Bang a similar example? Something ubiquitous that eventually had no other credible alternative explanation?

Yeah, but on a pretty short timescale. Penzias and Wilson had this brand new Bell Labs antenna that picked up the same amount of noise at a particular wavelength no matter what direction they looked. They did everything they could to remove sources of noise, which iirc involved scaring off some birds who had been crapping on the antenna with a shotgun. When they were super confident the instrument was working correctly and actually detecting something out in space, they called up a physicist, Robert Dicke, to ask if he had any idea what it might be. Dicke, who had been planning to build an antenna to search for the cosmic microwave background after he and Jim Peebles and others had made precise predictions about it, got off the phone, turned to his colleagues, and said, "Well boys, we've been scooped."

[Buster's Uncle]

The Worst Theoretical Prediction in the History of Physics
#Science #Cosmology #Astronomy #GeneralRelativity #CosmologicalConstant #QuantumFieldTheory

[Lorizael]

Heh. This is a story physicists like to tell. 

If you look at quantum physics, it says that the vacuum of space should have energy itself. And since most of space is a vacuum, that would seem to work as a candidate for dark energy. But when you try to calculate how much energy might be in the vacuum (there's not necessarily a definitively correct way to do this), it comes out to be ~120 orders of magnitude (that is, a 1 followed by 120 0s) more powerful than the observed effect of dark energy. So that's a pretty bad prediction.

[Buster's Uncle]

[Lorizael]

Since this is kind of the thread where I answer people's astronomy questions, here's a question I've got for all of you. I'm applying to grad school right now--a science writing master's program--and I need to get letters of recommendation. I've gotten 2 from the astronomy professors I TA'ed for last year, but I need a third one.

My initial thought (Option A) was to ask an astronomy professor I had a class with a couple years back who is in the department I work for now. I remember her praising my papers, and we're friendly and run into each other pretty frequently now. But one of my TA professors asked if I was getting a letter from the writing side of things, and the more I think about it the more that sounds like it might be a good idea. The problem is that my last class was a year ago and I was never super close with any instructor, either in the astronomy department or elsewhere. But there are a few candidates I've come up with.

(Option B) Fall 2016, I took a class on the philosophy of language that involved a lot of writing. That's the most recent class I've taken, so there's a better chance of the instructor remembering me and having positive things to say. Downside is he's just a grad student, and apparently having the Dr. before your name is a biiig plus for letters of recommendation.

(Option C) Spring 2016, philosophy of mathematics. That was a small class (~10 people) that involved a lot of writing and active discussion every class that I actually participated in. I think there's a good chance the instructor will remember me, but it was a year and a half ago. Upside is he's now a postdoc and can put Dr. in front of his name (but is no longer at the university).

(Option D) Also Spring 2016, literature of science and technology. A lot of writing that superficially sounds similar to what the grad program would be like (not really, because the instructor mostly had us write about post-modernism and globalism for some reason), and the instructor was quite complimentary of my writing. Con: larger class, not sure if he will remember me; still a PhD candidate.

Anybody have any thoughts or advice to offer? Monkeys.

[Rusty Edge]

Not that I ever did that kind of application, but Option C sounds to me like it would stand out in a good way. Me reading it, I'd notice the Phd now somewhere else and conclude, either Lori is such a standout that he has a lasting contact with this guy, OR Lori is diligent enough to be able to get references from other places, not just the convenient undergrad ones.

[Lorizael]

Yeah, that's what I'm leaning toward. Will have to see if I can find his new contact information...

[Buster's Uncle]

Hey Lori - I had an idea for you.  You asked a few months ago about qualifications to be a science writer/popularizer.

-So, what if when I see a journal link in a science article, I post it, and you try writing it up.  You could even do easy and hard versions.  I have a journalism background, and would be glad to chip in on style, and I'm sure all of us would be glad to ask questions and give you a feel for what needs dwelling on and what doesn't - at lest for a somewhat science-literate audience most likely to read such things.

It would also generate blog articles, of course.

Here's two links on an astronomical topic...

http://iopscience.iop.org/article/10.3847/1538-4357/aa9789
http://aip.scitation.org/doi/full/10.1063/1.4964481

[Lorizael]

That sounds like a good idea. It will have to wait a little bit, though. I need to be writing application essays for the grad program thing I'm applying to.

[Lorizael]

http://iopscience.iop.org/article/10.3847/1538-4357/aa9789
http://aip.scitation.org/doi/full/10.1063/1.4964481

Oh boy. So these papers deal with magnetohydrodynamics (MHD), which is a complicated field I know almost nothing about. With what I do know and can piece together from some digging, here's the story.

Directly observing or performing experiments on phenomena for which MHD is relevant is hard. MHD deals with the movement of electrically conductive fluids, so we're talking about the molten core of the Earth, or the plasma in the solar atmosphere, or the accretion disks of supermassive black holes. Because studying these directly is problematic, physicists have developed a lot of simplifying models using the equations of fluid dynamics and electromagnetism.

One explosive feature we see in this field is something called magnetic reconnection. Because you've got charged particles moving around in plasmas, you also have magnetic fields. Magnetic reconnection is a process by which energy stored in a magnetic field is released as heat or kinetic energy when the shape (topology) of the field changes in a dramatic way. Solar flares, for example, are very likely a result of reconnection, so understanding the process is important. These papers discuss problems with an older model of reconnection and the ways in which their newer model solves them.

In simplistic, ideal MHD, where there's no fluid viscosity or electrical resistivity or anything like that, plasmas will flow along magnetic field lines just like metal filings following field lines around a bar magnet. Even if there are shocks or waves in the plasma, the topology of the lines won't change and everything just keeps going. But we know that in real situations, when magnetic field lines get pushed around and tangled up, they can snap and form new lines. This is magnetic reconnection, and if it happens very rapidly (which seems to be the case), the particles following along with their lines suddenly jerk around and start colliding, and all that turmoil releases energy.

Because ideal MHD cannot account for this, physicists have developed models to explain the phenomenon. A classic one is the Sweet-Parker current sheet. So imagine you've got two magnetic field lines sitting parallel to each other but pointing in opposite directions.

S---------->N
N<----------S

The model says that electric current will start to flow in a thin sheet between the two field lines. But these current sheets are susceptible to tearing, because we're dealing with a very chaotic environment. When that happens, the sheet collapses and the magnetic field lines--brought closer together by the current sheet--can break and reconnect in a different orientation. (Below, the lines pinch off in the middle and turn around.)

S-----\ /---->N
N<---/ \------S

The problem is that on the scales of astrophysical phenomena (like solar flares), the Sweet-Parker model says magnetic reconnection should take weeks or months and be kind of a gradual thing, rather than the fast, explosive events we actually see. So a lot of work has been done trying to figure out how to make fast reconnection happen.

These two papers dig into the theoretical details of a current sheet model in which plasmoid instabilities cause sheets to collapse. Plasmoids are little magnetic bubbles of plasma that can form in a current sheet as it's growing. These isolated magnetic bundles separate chunks of the current sheet, leading to many smaller sheets connecting the two magnetic lines rather than one large one. The smaller sheets collapse more quickly, allowing fast reconnection and explosive energy release.

The big idea behind the model used in these papers is a least-time principle. Plasmoid instabilities initially grow at a linear rate. Eventually, the process stops being linear and gets weird. Their least-time principle argues that whichever plasmoid instabilities get through the linear phase of growth first will become dominant, and so its their properties which determine how the current sheet evolves and breaks down.

The big result from the theoretical work done in these papers is a set of scaling laws, which show how the properties of plasmoid instabilities change at different scales. Scaling laws in physics are very often power laws, which means that if you double x, the properties of x go like 2^n, where n is some constant power. Their results show that this is not the case for plasmoid instabilities and their scaling laws are more complicated, requiring different rules at different scales. One of the results to come out is that current sheets thin quickly, and this plays a big role in how plasmoid instabilities grow.

After detailing the scaling laws and factoring in things like viscosity and resistance, they show that magnetic reconnection can happen very quickly, on timescales comparable to what would be necessary for, say, solar flares.

There's plenty more work to be done here in terms of (a) adding in more variables to make the model more realistic and (b) testing the model in plasma physics experiments. You can do this on the ground, but you can also do this in satellites that park themselves in our magnetosphere and monitor activity to see if plasmoid instabilities really grow the way the model suggests.

[Buster's Uncle]

Ok.  Now, try to condense it to a 500-word Jack Horkheimer script, only less lamely unfunny.

[Rusty Edge]

How goes the grad school application?

[Lorizael]

Just waiting for my letters of recommendation to come in now. I have to send the whole thing in by January 15 and then wait until April 1 for a decision.

[Lorizael]

Ah phooey. Didn't get in. Now to figure out what to do with my life again.

[Unorthodox]

Become an electrical engineer. 

Rocket companies are ALWAYS hiring electrical engineers.