Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
amazing news if you like the subject
Options
-
15-02-2018 7:50pm#1
Comments
-
-
-
-
-
-
Advertisement
-
-
-
-
The thing that hits home for me is they can detect moons, we have not found any outside our own solar system before.riffmongous wrote: »Did they though? The snippet just says 'objects' with the same mass as the moon, which could also mean dwarf planets. Has any access to the full article?
The article preprint is up on arXiv here. It's hard to get one's brain around what this "detection" actually means. I think people might be disappointed to find it's nothing to do with identifying particular planets, moons, or anything else. Here's my poor layman's superficial attempt to explain it. As is often the case with scientific papers, the amount of information in the paper itself is tiny compared to the amount of background material you need to understand it. So most of the explanation will be background stuff.
Gravitational lensing occurs when light coming to us from some source object is bent by the gravity of some intervening massive object, typically a galaxy or galaxy cluster. It is very roughly analagous to the behaviour of a refractive glass lens. Light rays are bent as they pass through the lens, geometrically altering the size or shape of the image that is formed.
The analogy shouldn't be taken too far. We typically construct glass lenses such that the amount of bending of a light ray depends on its distance from the optical axis (the line through the centre of the lens), so that the whole bundle of light rays is brought to a single focus and forms an image. The geometry of gravitational lensing is different, so that we can get multiple (possibly deformed) images. In the limiting case of the source object and lens being perfectly aligned along our line of sight, we get the source stretched out into a circle around the lens -- a so-called Einstein ring. One of the coolest (and certainly happiest) space pictures is this Hubble telescope image of multiple Einstein rings caused by a pair of cluster galaxies:
A major difference between classical optics and gravitational lensing is that the latter can make the overall image brighter. If you look at planets through a telescope, a frustrating feature is that the more you turn up the magnification (by changing eyepieces to shorter focal lengths) the dimmer the image gets. The constant amount of light collected by the telescope is spread over a larger solid angle so the surface brightness decreases. Not so with gravitational lensing. In fact, surface brightness is conserved no matter how much the image is magnified. This counterintuitive feature isn't easy to explain without complicated mathematics and it may even seem to violate conservation of energy. But we'll skip over it by noting that with an ordinary lens we are focusing light rays that radiate isotropically (equally in all directions) in space, whereas a gravitational lens can redirect light toward us grabbed from a larger than normal volume of space.
Now, when we actually gaze out into space looking for examples of gravitational lensing, they are few and far between. Because they rely on precise alignments of the background source object and the foreground lensing object, and space is almost completely empty on average as evidenced by the darkness of the sky, chance alignments are rare. Furthermore, galaxies are incredibly distant and do not move perceptibly so there are no new alignments coming into existence even over long timescales.
This is where we turn our attention to microlensing. When we see a strong lensing effect it typically has an Einstein radius of about one arcsecond if the source and the lens are galaxies, and maybe ten times that for galaxy clusters. (The Einstein radius is the angular radius of the Einstein ring if there is one, and one arcsecond = 1/3600 degree). Microlensing is gravitational lensing caused by much smaller masses, individual stars or even smaller. For lensing caused by stars in the Milky Way the Einstein radius is measured in milliarcseconds, and for stars in other galaxies in microarcseconds.
Although the microlensing effect is much tinier, there is the compensating advantage that it occurs much more frequently. If we look at a background object through an intervening galaxy, the billions of individual stars in that galaxy have a much higher chance of transiting the source object. Although the motions of the individual stars are tiny, the very large numbers mean that chance alignments will come into existence fairly regularly.
Now, in the case of microlensing we don't see any geometrical changes. Whereas for strong lensing we might see double images, Einstein crosses or rings, with microlensing the effect is much too small for that.
But remember how lensing can increase the total brightness of the lensed object? We can measure that even if we can't optically resolve the lensing event. This is similar to how astronomers measure changes in brightness of variable stars such as Cepheids. Those stars are actually changing in physical size but most of them are much too far away for the size to be apparent. Nevertheless, the brightness variation can be seen even for unresolved objects. Microlensing events are way, way below the resolving power of telescopes, but brightness changes can still be measured if we look at the lensing system over a period of time.
An additional way to zoom in on brightness changes is to only look at changes in particular wavelengths of light. Suppose the background source object is a quasar. As you know, a quasar is the supermassive black hole at the centre of an active galaxy, i.e. one where the black hole is actively feeding on material falling into it. The quasar light comes from an incredibly tiny region -- relative to the galaxy -- near the black hole's accretion disc. The synchrotron radiation created by this whirling mass increases in frequency the closer you get to the central engine.
So when we look at high-frequency X-rays from a quasar we are zooming in on an unbelievably tiny region, maybe only one or two orders of magnitude bigger than our solar system, even though it may be billions of light years away. This creates incredibly precise alignments with foreground lensing objects which can create high magnifications (by factors of several) and can be measured through changes in brightness in the emission lines in the X-ray spectrum. The preciseness of the alignment also means that it changes over mere days, weeks, or months. As with most things in the real world, it's a bit messier than that in practice. Quasars are intrinsically variable so to measure the extrinsic variability caused by lensing needs additional techniques that I will come back to later.
Ok, so now we have set the scene for how we can detect microlensing events through observing the light curve of quasar lensing systems in the X-ray spectrum. There is one more physical phenomenon that we need to explore. This is the concept of optical caustics. The basic concept is simple. Remember how you used to incinerate ants with a magnifying glass? Caustic means burning, that's where the word comes from. Caustics in the present context are the overlapping bright regions created by lensing.
A combination of curved surfaces can produce complicated patterns. Above left is the caustic produced by the overlapping light rays from the front and back curved surfaces of a water glass. Ripples on a water surface can produce more complicated caustics. Imagine you're the barracuda in the right hand image (he's not easy to spot but he is there). Suppose you stay in one place while the pattern of ripples changes above you. You are going to see bright transient flashes due to the shifting pattern of caustics. Effectively it's a one dimensional projection of the two dimensional pattern.
Ok, so finally: the Chandra X-ray telescope is the barracuda. A distant quasar is the light source. The ripples are the masses of stars and planets in an intervening galaxy which individually microlense the quasar. Because the short wavelength X-ray light must pass extremely close to a stellar or planetary sized mass to be lensed, it is only occasionally that we will get multiple stars or planets in the line of sight to cause multiple lensing and caustics. The frequency with which we see the caustics then tells us the distribution of stars and planets in the intervening space, and the size of the spikes in the light curve tells us about their masses.
Here's another good visualisation which you can see a fuller description of by following the youtube link. This is microlensing caused by a binary pair of stars of unequal mass, represented by the yellow circle and tiny yellow dot. The green circle is the geometric position of a background quasar and the red shapes which vary in number are the lensed images of the quasar. The geometric positions at which caustics will occur are represented by the light blue lines. At the bottom you can see our barracuda's-eye-view of what's going on. This is a graph measuring the total brightness coming from all the quasar images together. The caustics are manifested in the two spikes in the light curve.
And finally finally (!) we come to the "discovery" mentioned by the OP. What did the researchers actually detect? They took ten years worth of Chandra observations of a particular strongly lensed quasar. This exhibits four separate strongly lensed images. Now, intrinsic variations in the quasar light due to, say, accretion disc outbursts, appear in all four images separated by a time delay corresponding to the different paths travelled by the light producing each image. These can be subtracted out, leaving just the extrinsic variations due to microlensing events. These are measured and the size and frequency of caustics is analysed.
Next they construct a mathematical model of stars in the foreground lensing galaxy. They demonstrate (mathematically) that there is not enough of them to account for the caustics. So now they start sprinkling in distributions of planets among the stars. These are unbound planets, i.e. ones that are wandering between the stars. If they were gravitationally bound they would be too close to their parent stars to account for caustics. They vary the number and mass distribution of planets in their mathematical models and compare the resulting simulated caustics with the real ones. When they get a pattern that looks similar, they shout bingo! The matching pattern is caused by a distribution of planets with masses between that of the Moon and Jupiter, without about 2000 planets per star.
So on the one hand, this is a really interesting result from detailed and painstaking work. On the other hand, advertising it as an actual detection of planets -- as opposed to a statistical analysis of a pattern -- will probably sound a little bit disappointing to the layperson who delves into it.
Hope this all made a bit of sense! :pac:
More reading:- What is Gravitational Lensing? -- from the Canada France Hawaii lensing survey
- Gravitational lens and gravitational microlensing on Wikipedia
- Caustic (optics) on Wikipedia
- Gravitational lensing in astronomy -- a very good intro to the history and maths of lensing, but see especially the section on quasar microlensing.
- Microlensing of distant quasars -- a fairly technical overview presentation of observing techniques
- OU Astrophysicists Discover Extragalactic Planets for First Time -- the OP's link
- Probing Planets in Extragalactic Galaxies Using Quasar Microlensing -- source paper on arXiv
0
Advertisement