A Scintillating Description of Why Quasars Twinkle. Or, How We Tried to Break the Very Large Array.

One of the great things about being a radio astronomer is that if you have a good idea, and can convince a group of fellow astronomers that it really is a good idea, then you get to use an amazing telescope like the Very Large Array (VLA) to try it out. For free from anywhere in the world! This is a story about MASIV, a project we worked on that used the VLA in an unusual and previously untested way. Our idea? To try to understand why quasars twinkle.

If your eyes could work at radio wavelengths like the VLA, the sky would have a very different look. One thing you would notice straight away is the many, very bright points of light, similar to stars, but not in the constellations you are familiar with. In this radio view of the sky, the stars are not stars but quasars—young galaxies billions of light-years away.

Quasars are brilliant beacons of intense light from the centers of distant galaxies. They are powered by supermassive black holes enthusiastically feeding on the infalling debris. This feeding frenzy unleashes a torrent of radiation that can outshine the collective light of billions of stars in the host galaxy. Radio light is created as the material is ejected from the region around a supermassive black hole at the center of the quasar. The energies involved are so extreme that even though they are so far away, they dominate the sky.

Since the very early days of radio astronomy, observers noticed that quasars change brightness. It’s not surprising that the material falling into a central black hole—and then flung out—wouldn’t happen continuously. But astronomers noticed something mysterious: some quasars changed their brightness much too quickly: getting brighter, then darker, and brighter again over the course of a day or less. If this was happening near a black hole, the physics would have to be unbelievably extreme, and we’d have a very difficult time explaining it. But there’s another possible explanation for this mysterious behavior, something called Intra-Day Variability, or IDV, or, simply, twinkling.

Turbulence in our atmosphere that’s due to wind and weather causes the light from stars to appear to change very quickly. This is what is referred to as twinkling. And, it turns out, some quasars twinkle too. But the cause isn’t the Earth’s weather; it’s the weather in our Milky Way galaxy: movement of the ionized gas cruising between the stars. Astronomers call this phenomenon Interstellar Scintillation but really think of it as twinkling.

Back in the late 1990s and early 2000s, we really didn’t know if this variability in quasars was mainly due to twinkling, or extreme black hole physics, or a mixture of both. We knew that in a few special cases that it was twinkling, but if we were really going to get a handle on what was going on, we were going to have to look at hundreds of quasars and see if they varied on timescales of days or less. We were either going to need lots of time with a small array of telescopes or a shorter amount of time with a big array. So guess what we did? We went big!

We decided that the best way to do this was to choose about 700 bright quasars and observe each one every two hours for three days. The quasars we chose were nice and bright so you didn’t need to spend very much time looking at them to get a good measurement. In fact, we didn’t even really need all of the telescopes at the VLA; just five or six of them would be enough. We decided to break up the VLA into five smaller arrays (or sub-arrays) and program each one to observe a different bunch of quasars. That way we could observe each one regularly enough and still cover all 700. Most astronomers want to throw the full power of the VLA in one direction to get a really good image of their celestial target; that’s why you usually see pictures of the VLA with all the telescopes looking in the same direction. We had other plans.

When we mentioned this idea to our colleagues at the VLA in Socorro, we got some interesting responses:
NRAO: You know that no one has actually tried this before, don’t you?
Us: Um, OK, but it says here that you can do five sub-arrays.
NRAO: Yes, but no one’s ever been crazy enough to do it!
Us: We’re willing to risk it. Please, can we? Please?
NRAO: You’re going to have to come over from Australia, get here ahead of time, and help make it work. Sorry.
Us: Are you kidding? Try and stop us!

In January 2002, Hayley and I got on a plane and traveled to Socorro. We’d learnt a lot over the previous few months about how the VLA works, how to schedule it in sub-array mode, and wrote a bunch of software to tell each sub-array what to do. We also knew that we might have to make changes on the fly once the observations started if it turned out things didn’t work.

When you’re doing a project as big as ours, you’ve gotta have a good name for it; so, Hayley came up with MASIV, which sounded great. She, then, came up with the necessary abbreviations from “Micro-Arcsecond Scintillation Induced Variability Survey,” which fit perfectly!

When we got to Socorro, we met up with Ken Sowinski, who was our local contact and VLA expert. It turned out that we’d already developed a bit of a reputation. He introduced us to everyone else: “These are the Australians who are doing that infamous experiment.”

We spent the next few days getting over our jet lag, spending time checking and double-checking our schedule files, and getting ready to process the data in real-time so we could make sure it was all working. On the day of the first observations, we drove out to the site to meet the telescope operators and get ready to work on the data once it started coming in. I’ve had the privilege of using the VLA a few times now but am still blown away by the sight of all those telescopes!

Well, the time finally arrived and the observations started. There were problems initially, and I seem to remember having to make some rushed last-minute changes. But thanks to the VLA staff, it didn’t take long to get things fixed up and we were soon getting good data. Over the next year, we came back for more observations every four months and then back again in 2009 for some more follow-up observing. Each time, one or two of us were there to work with the locals and help make sure things ran smoothly.

The VLA has four main configurations called A, B, C and D-array, with A-array being the configuration where the telescopes are spread out as much as possible and D-array where they’re very close together. Our configuration became known as the “the VLA in Dis-Array.”

So, what did we discover from all of this? It turns out that the light from quasars that we see through the plane of the Milky Way show much more variability than those that are seen in other directions. This tells us that twinkling is the main cause, not extreme physics. But, now that we know it’s twinkling, we can say some interesting things about what’s going on near the black hole. For a quasar to twinkle, all the radio waves have to come from a very small region. If the quasar doesn’t twinkle, it means that the radio waves are coming from a wide area and the twinkling sort of averages out and you don’t see very much change. Now we know that just by checking to see if a quasar is varying (which you can do with even a single medium-sized radio telescope), you can tell if the black hole is going bananas or just having a lazy day off.

It’s been a great privilege to use the VLA and to have worked with people at the NRAO who have been so knowledgeable, helpful, and tolerant of astronomers doing their best to break their telescope. All of us in the MASIV group wish the VLA a very happy 40th birthday and look forward to seeing lots more amazing science from it in the future!

Sadly, since writing this article, our friend and MASIV colleague J-P Macquart passed away. J-P played a key role in the MASIV project, bringing his in-depth knowledge of scintillation theory and his assistance in data analysis and interpretation. We miss his insights, his sense of humour and his friendship. J-P’s work on MASIV led to his ground-breaking work on using Fast Radio Bursts to identify missing matter in the Universe.

About the Author:
Jim Lovell Jim has a PhD in physics from the University of Tasmania and has worked primarily in the fields of radio astronomy and geodesy. Jim has worked as an astronomer in Japan at the Institute of Space and Astronautical Science on the VLBI Space Observatory Programme mission (which involved the first orbiting radio telescope), then in Canberra conducting research in astronomy and supporting observations at NASA’s Deep Space Station at Tidbinbilla for CSIRO. He served as Project Manager of the AuScope VLBI project at the University of Tasmania, which involved the construction and operation of an array of three radio telescopes in Australia (Hobart (Tas), Katherine (NT) and Yarragadee (WA)) designed specifically for geodesy. Jim is currently working in several consultancy-based roles with an emphasis on applying his skills and experience obtained in astrophysics to environmental applications.
View All Blog Posts