What's in a Pixel?
- phys1500.astronomy
- Aug 31, 2018
- 3 min read
by Isabel Colman
When you think of pictures in astronomy, you might think of colourful images of space, high-resolution and glimmering with millions of points of starlight. Maybe you’re imagining something like this:
This is a picture of the sky near the bright belt of the Milky Way. That bright star over to the right-hand side is Vega, and if you’re into constellations, you might be able to pick out Cygnus and Lyra.
But we’re not going to talk about pictures like that one. We’re going to talk about pictures like this:
This is a picture of a star, in false colour to indicate relative brightness. Each square in this image is a pixel on a CCD with a width of four arcseconds. (For reference, the moon is roughly thirty arcminutes wide in the night sky.) If you were to map this image onto the first one, it would take up only a tiny fraction of it. In fact, we can do exactly that, because this small image, which we call a “postage stamp” is somewhere in that bigger one.
The postage stamp above is just one of many to come from NASA’s Kepler space telescope. The Kepler mission looked at the same patch of sky for four years straight, and every thirty minutes it downloaded a picture of its entire field of view (shown below), which is made up of 94,617,600 pixels in total. Contained in each of these pixels is a measurement of brightness! The big picture is cut up into postage stamps, targeting around over 150,000 stars within the field. From a series of postage stamps over time, we can measure changes in the brightness of stars.
Kepler was designed to look for a type of change in a star’s brightness that indicates it has a planet orbiting it, and it’s been hugely successful—over the course of the mission, astronomers have discovered over 2,000 new exoplanets. However, the way a star’s brightness changes over time can also indicate other things, such as inherent stellar variability, or “stellar oscillations.”
Not all the stars we study have names like Vega. The one I showed above is called KIC 7461601, where KIC stands for Kepler Input Catalog. In the image below, the top panel is showing some of the brightness measurements over time for KIC 7461601, called a “light curve,” and the bottom panel is showing a Fourier transform of that light curve. I won’t get into the maths of the Fourier transform, but basically what it does is convert a series of measurements over time into a series of measurements of frequency. Put simply, this means we can use changes in brightness to see the frequencies and patterns of those changes. (The study of stellar oscillations is called asteroseismology, if you’re interested in learning more.)
To break this down, that comb-like pattern of oscillations to the left of the bottom panel indicates that this star is a red giant. The high spike suggests some kind of binary behaviour—however, we can do the maths and work out that, if this were a red giant with a companion star orbiting it, then the companion would literally be inside the red giant.
So, what’s going on? Well, the light curve shown in the top panel is created by taking a composite of the light from several pixels in each postage stamp. The most likely explanation is that, during this process, we’ve captured light from more than one object: a red giant and a binary star system. The good news is that we can check if this is true by looking at our postage stamps. Because we’ve measured the change in brightness over time of each pixel, we can also perform a Fourier transform on each pixel. That process gives us an image like this:
Each pixel displays a measure of oscillation, the patterns of how the light measured by this pixel changes in brightness over time. Some pixels have no oscillations, just noise. But the exciting part is that we can very clearly see that the red giant oscillations and the binary peak are coming from different pixels. Remember, we can map this to the sky, and in doing so, we’re able to confirm that we’re looking at two different objects.
We know of about 180 objects like KIC 7461601. Roughly half of them have so far proved to be chance alignments, as above, but we think that the other half are red giants orbited by a compact binary system—that is, triple stellar systems! You can read more about all of this in a paper I published last year: https://arxiv.org/abs/1705.00621.
So, while it’s always nice to look at pretty, high-resolution pictures of space, don’t forget that you can also find a wealth of information contained in just a handful of pixels!
コメント