Colliding Black Holes in Pulsar Data
Ben - Pulsars are the remnants of stars which rapidly rotate and produce distinctive pulses of radio emissions. Researchers are looking for these telltale repeating patterns in huge swathes of data and sometimes, they encounter something unexpected. I spoke to Evan Keane from the Max Planck Institute in Bonn, in Germany.
Evan - Well, I've been involved in surveys to search for pulsars in order to answer a number of questions. One just quite important is how many pulsars there are in the galaxy? For astronomers, it used to be stars. It's quite important for us to know how many there are and pulsars are very extreme stars.
Ben - What are you looking for? What are the signatures?
Evan - We use very big telescopes like the Lovell telescope at Jodrell Bank. We collect large volumes of data by regular human standards. For instance, surveys I've searched which would be considered small, would be several terabytes of data you have to search for, and now, we're talking about petabytes so that's a million gigabytes. I mean, we're looking for needles in a very large haystack basically. And a lot of work goes into a very clever analysis of this data, trying to think up new tricks.
Ben - So how many pulsars do we know of?
Evan - Currently known, there's about 2,000 pulsars give or take. There should be a lot more than that. Usually, they're described as being clock-like. So, we're looking for a signal which repeats quite regularly and there are quite a lot of data analysis techniques we can use to tune into these, anything that's periodic like that, we can find that. There's a lot of research using these guys as very accurate clocks and "super clocks in space" are ideal tools we can use for a lot of experiments, and that's one thing I'm interested in. One thing I've been working on recently is those pulsars that are not so reliable, so we get a pulse and then they're quiet - they're off for say, 100 pulses. So they are mostly off, but sometimes when they're on, they're quite bright.
I look through a pulsar survey, looking for these kind of objects because they don't come up in the standard analyses and there's lots and lots of pulsar data that hasn't been searched for these kind of signals. In this search, I found one signal of particular interest. This was a single burst that lasted just 7 milliseconds, so not very long, and its inferred distance was quite far. In fact, it's outside the galaxy which is the interesting thing about this.
How do I know that when I hear one pulse? Well, when something, a pulsar or some source emits some light and it comes towards us, it travels at the speed of light, but it actually is subject to delay. So the light hits against stuff. Stuff between us and the source and the will slow it down. And as it turns out, a high frequency emission arrives before the low frequency emission and if you measured this delay, you can work out how far it's come. And if you work it out for this particular source I found, you'll find that the delay is much longer than the maximum delay that could've happened from it coming through the entire galaxy. So, the rest of the delay has to come from something beyond our galaxy. There are many theoretical papers of sources that should do this. Annihilating black holes should give a burst like this. There's a beautiful paper from Martin Reese from 1977 where he predicts what an annihilating black hole will look like and it looks just like this!
Ben - And what physically do you think that signal is coming from? Is it an annihilating black hole or is it a pulsar that's only sent very strong pulse so far?
Evan - This depends crucially on how far away it is. We want to absolutely nail down the distance of the source. If we were to get up close to this source, we would realise it was really, really bright. In fact, much too bright to be a pulsar. So if it's far away as I'm telling you, it's not a pulsar. So what I've been trying to do is break the standard models of how far this is away. We think we understand how much stuff there is in our galaxy and I've just been thinking, "Well, suppose we're wrong - we're totally wrong." And let's say it's a pulsar at the edge of our galaxy, this would mean it's a pulsar like the Crab pulsar which occasionally gives very loud, very strong bursts. Now if this source were a pulsar, we would see many pulses from this. This is why I was in Australia last week to use a telescope there at the Parkes Radio Telescope and I observed this guy for 10 hours, looking for some pulses. I haven't seen any which is good. Either result would've been good because seeing some pulses will tell us a model which we thought we understood to be correct is wrong and that there's a weird pulsar at the edge of the galaxy, that's one scenario which is interesting. Second scenario is one model that we thought was correct is correct which is always nice, and that we've detected this extra galactic burst of quite a significant interest.
Ben - So what's your next step? What are you going to do to try and follow up that event?
Evan - Well, there's nothing I can do. This signal happened 10 years ago. Pulsar people, they collect so much data that they can't analyse it because they want to plan ahead. That means that there's lag between when this actually happened and when it was noticed in the data. In some senses it's inconclusive because I can only present there was this very strong burst, it seems to come from far away for all of these reasons. I can suggest that there's a high probability that it is something like this. But for the future, these events should be much more common. If we can monitor the whole sky, we should be seeing this every day. But hopefully now, in the current or in the next generation of telescopes, we can be more real-time. As much as it's possible, we want to detect - people do it with all the other wavelengths in x-ray and gamma ray telescopes. They send alerts to everyone on earth saying, "Something just went bang. Everybody look there" and we want to do that.
Ben - Evan Keane from the Max Planck Institute explaining why quick action is needed to study very brief radio signals and possibly to identify colliding black holes.