Universe Today
(This is Part 2 of a series on cosmic dust. Read Part 1 first.)
And if you're trying to do any sort of astronomy or cosmology whatsoever, dust just gets in the way. It contaminates. It RUINS.
Dust messes with light in really complicated ways. And it's EVERYWHERE. If you go out into some random patch of interstellar space, you're not going to have a lot of stuff around you. And about 99% of that stuff will be simple hydrogen and helium. The rest is generally categorized as "dust", which is microscopic clumps of heavier elements like carbon, oxygen, and iron.
Yes, all those elements are made inside of stars. So whenever someone is waxing poetic about "stardust" you have permission from me to roll your eyes in a really exaggerated way. Because stardust is AWFUL.
The smallest grains are just a few molecules, so only nanometers across, while the largest, the true giants of the dusty cosmos are...a thousandth of a millimeter across.
Now what these grains do to light that is so dang annoying is that they don't just do one thing, in one way, all across time and space. Each dust grain is slightly unique. It has different elements in different combinations in different arrangements. And sometimes there's a little bit of dust, and sometimes there's a lot. And what the grains are made of and how they're arranged affects what the dust grain does to light.
Some wavelengths get scattered. The photon comes in, bounces off the grain, and heads off in some random new direction. The light doesn't disappear, but it gets redirected, which means the image you were trying to make of that distant galaxy is now a blurry smear, and some of the light you're collecting wasn't actually coming from where you think it was.
Some wavelengths get absorbed. The photon comes in, the grain swallows it, and the energy goes into heating the grain up. That photon is gone, full stop. If you were trying to measure how bright something was, well, it's now dimmer than it should be.
And some wavelengths get EMITTED. That heated-up grain has to dump its energy somewhere, and it does, by glowing in the infrared. So now your "empty" patch of space is shining at you in wavelengths the original star never produced. You're seeing light, but it's coming from the wrong thing.
And one dust grain will have one behavior, and then its neighbor will have a different behavior.
And yeah, 1% of basically nothing is even less than basically nothing. By density dust is barely there. But there's a lot of space in space (which is why we call it space) so by volume there's a TREMENDOUS amount of dust. If we're looking at something 1000 light-years away, that light has to travel through 1000 light-years of dust, and different kinds of dust in different kinds of densities along different lines of sight.
A typical galaxy like the Milky Way contains roughly 100 million solar masses of dust. If you gathered up all the dust into one spot, you could build 100 million suns. (Which wouldn't work very well as a sun, but that's a different story.)
And it's not like the dust just goes away. You can get out your interstellar vacuum cleaner (I'm thinking of Spaceballs here) but it won't help. Why? Because just like in your house, THERE'S MORE DUST AROUND THE CORNER. Giant stars blowing off their outer envelopes. Supernovae going off. Dust GROWING ON ITS OWN just by sticking to its buddies.
And all this just messes everything up.
It means that no observation in astronomy is raw. Every single photon we receive from the depths has been filtered through who knows how much dust. The dust dims, selectively reduces wavelengths, and scatters light. It can be said that the history of astronomical discoveries is a) not realizing how much dust was in the way of something, then b) slowly figuring it out the hard way.
In general, dust makes almost everything dimmer and redder than it normally is. Blue light scatters more easily off small grains, so it gets selectively removed from a sight line, leaving the longer-wavelength red light to pass through. This is exactly the same effect that makes sunsets red on Earth, just operating over light-years instead of a few kilometers of atmosphere. So anytime you need to measure the distance and/or color of something, if you don't properly account for the dust, you're going to be wrong.
This is one of the reasons, if not THE reason, that measurements of stellar distances prior to 1930 or so were all way off. Astronomers had a method that should have worked beautifully: find a star whose intrinsic brightness you know, measure how bright it appears, do some quick algebra, and you've got the distance. Brighter than expected means closer. Dimmer means farther. The whole technique runs on a single hidden assumption, that nothing is in the way between you and the star. Robert Trumpler eventually showed that this assumption was very, very wrong. There's a galactic plane full of dust between us and pretty much everything we care about, and dust makes stars look dimmer, which makes them look farther, which means our entire map of the Milky Way was stretched out and distorted. The galaxy was actually smaller than people thought. The Sun was in a different position. Whole structural conclusions had to be redrawn. Just from dust.
And pretty much every single debate and discussion in astronomy and cosmology somehow involves dust. See a faint red galaxy? Well are you SURE it's a faint red galaxy? Or is it a bright blue galaxy surrounded by dust? See a distant supernova? Well are you SURE you know how distant it is? Or is it actually closer than you think, just dimmed and reddened by foreground material?
"I need to model the dust" is the never-ending headache of every astronomer. The basic approach is to combine laboratory measurements of how different dust compositions interact with light, then observe your target across many wavelengths simultaneously. Each wavelength is affected differently. Visible light gets scattered hard, near-infrared somewhat less, far-infrared barely at all. By comparing what you see across the spectrum, and fitting the differences against your laboratory models, you can solve for how much dust is in the way and what kind it is. Then you subtract its effect from your data. It works. Mostly. Sometimes. With caveats. With error bars that nobody likes.
And of course dust can be a major source of embarrassment. In 2014, the BICEP2 collaboration announced that they had detected the imprint of primordial gravitational waves in the cosmic microwave background. This would have been one of the biggest discoveries in cosmology in decades, direct evidence of inflation in the very early universe. Press conferences. Champagne. Drafts of Nobel Prize speeches, probably. Then the Planck satellite team came along and pointed out that the patch of sky BICEP2 had been staring at had a lot more galactic dust in the foreground than the team had assumed. The "primordial" signal was, to a very large extent, just polarized emission from local dust grains in our own galaxy doing their thing. The discovery quietly evaporated. Whole careers spent grappling with the cleanup. Dust did that.
Okay this apology isn't going so well, is it?
In Part 3, I have to start admitting that the stuff that ruins astronomy might also be doing something I owe it for.
