What happened in M31 on March 1st?

Avid followers of space science may remember the incident of May 2014 in which rumours spread like wildfire that Swift had detected a Gamma Ray Burst in the Andromeda Galaxy (also known as M31). In that case, it was all a twitter-storm in a tea cup. Well now (on March 1st 2015) Swift has, perhaps, detected something from M31 again. This time there is no huge Twitter reaction; in fact the little Twitter traffic I've seen on the subject is being very restrained and cautious, which is nice to see. But, this trigger does look like being something rather interesting, but let me start with a cautionary summary: this is almost definitely not a GRB in M31.

I'll start with a brief narrative of what happened, and then have a stab at interpreting it, but please be aware: this is an evolving situation. What I write now is subject to change, as more data become available.

What we do know — the sequence of events

M31 in X-rays

Image: ESO/DSS

At 06:20:54 UT, the Burst Alert Telescope (BAT) on Swift triggered on something. This means it believed it had detected a new gamma-ray source. The region those gamma-rays came from is shown as the green circle in the image to the right, which is superimposed on the DSS image of M31. As you can see, the location is consistent with the outskirts of that galaxy. Swift immediately swung round to point at this location and start collecting data with the X-ray and UV/optical telescopes onboard, but no X-ray source was found, and only already-known optical objects were there. The Swift team announced that the event was probably not a GRB, and could way just be noise, that is, nothing real at all. However, this was still preliminary. When a GRB goes off we can only get limited data products telemetered to the ground. We needed to wait for the full dataset to analyse this event properly.

Later on that day, around 18:15 UT, while Swift was responding to another GRB trigger (GRB 150301B; there had already been an earlier GRB before the M31 trigger) the data arrived on the ground, and a BAT team member, Dr. Jay Cummings, determined that the BAT data were definitely from a real event, and not simply noise in the detector. This is where the fun started. The X-ray Burst Scientist on duty, Dr. Alessandro Maselli, found evidence for a very faint, tentative X-ray source in the BAT error circle (that is, within the area BAT said the trigger had occured). This was a weak, and by no means certain detection: it might just be noise. Also, if it was real, it was not a new source: it was at the same place as an object detected in X-rays back in 2005, by the XMM-Newton observatory. So, the team issued a new announcement to astronomers confirming that the event was real, although its origin still unknown. It was not clear whether this faint X-ray object was related to the BAT trigger at all, and so more observations were planned.

By midday on Monday March 2nd, we had all the data and, it being a working day, many more people were online to discuss it. And as we did so, our interest spiked. Last year I produced a catalogue of all the X-ray sources Swift had detected up to October 2012. As part of this I had developed some new software to find faint sources; so I ran this software on the observations from March 1st, and found that our tentative X-ray detection became a solid, real detection of a X-ray emitting object. Also, my catalogue showed that we'd looked at the position of this object before but hadn't detected it. From this we worked out that when we detected the object on March 1 it was at least 3 times brighter than it had been last time we looked at that part of M31 (in September 2011). We also worked out, with the new data taken on the evening of March 1 and on March 2, that this newly detected object had already faded away and was no longer bright enough for Swift to detect. Finally, we checked how bright the object was in 2005 when XMM-Newton had detected it, and we found that when Swift detected the object on March 1 2015 was about 35 times brighter than when XMM-Newton had detected it.

So, all of this strongly suggested that, whatever the X-ray source was, it had undergone some kind of transient outburst. That is, it had briefly got brighter, and then faded away again. Given this, it is pretty likely that this is the event that triggered the BAT. So we issued this result both as circular and an Astronomer's Telegram, to let the professional community know about it.

But what is it?

The above is everything we can say for certain. Now we have to enter the realm of scientific deduction to work out what this object is. This is ongoing work, and anything I say here is subject to change, but here are a few thoughts:

It's not a GRB (in M31)
This is very, very unlikely to be a GRB in M31. It's simply far too faint. Given its brightness in the BAT, and the distance to M31, if this object was in M31 then it is about 10,000,000,000 times fainter than a normal GRB. I guess someone may try to invoke some complicated model, such as having the pencil-like ‘jets’ of the GRB pointed away from Earth; but that's going to be pretty hard to reconcile with the data: such an event should not (as far as I am aware) have triggered BAT at all.
It is probably not a GRB at all
I'm going to stick my neck out here and say, I don't think this is even a GRB. M31 is relatively large, so it's not impossible for a GRB to go off billions of light years away in a galaxy which, from our standpoint, is ‘behind’ M31. It's not impossible that this is what we've seen. But there are two arguments against this: first, the X-ray counterpart was so faint and short lived, it doesn't look at all like a normal GRB afterglow. This is not a fatal blow, we have seen a few long GRBs in the past with pretty much no afterglow. These so-called ‘naked’ bursts are believed to exist in regions with hardly any surrounding gas (it is this gas which, interacting with the GRB, gives the afterglow). But the fact that the X-ray source has been detected in the past by XMM-Newton makes it, I think, very unlikely that we're seeing a GRB. If the progenitor is bright enough (i.e. near enough) to be detected before exploding, the explosion should have been much brighter when it happened. Of course, there are possibilities which could rescue the GRB interpretation: maybe the previously-detected source was an Active Galactic Nucleus (AGN: a super-massive black hole at the centre of a distant galaxy), and the GRB went off in the galaxy around the AGN. Maybe the object Swift detected in X-rays is not the same as the object XMM-Newton saw, they just happen to lie close to the same line of sight (with the GRB being much further away): this is not as unlikely as you may think. But for now, I don't think this is a GRB.
Is it some other phenomenon in M31?
So if it's not a GRB, is it something else in M31? Well, maybe, but again, this is problematic. While the object is far too faint to be a GRB, it's also far too bright to be most other things, if it's in M31. There is a limit to how bright an object can be, depending on its mass. This is called the Eddington Limit, and based on this, if the object that triggered BAT is in M31 and is shining its light in all directions, it would have to be a black hole with a mass of about 30,000 times that of the Sun. This is pretty unlikely for an object towards the edge of M31. Of course, it could be a fainter object which is not shining its light equally in all directions: that light could be concentrated in narrow jets of material, like a GRB produces (as do many other phenomena). But even then, it's hard to work out what the object is. The brightest stellar objects we know of are called ULXes (Ultra-luminous X-ray sources), but the brightest ULXes are still about ten times fainter than this event was, more consistent with the brightness this object was when XMM-Newton saw it. Alternatively, the short-lived outburst could be some kind of explosion, which is then not limited by the Eddington Luminosity. For example Type-I X-ray bursts are thermonuclear exposions on the surface of a neutron star. But again, these are simply not bright enough to account for what Swift detected — if the object was in M31.

So, what is it?

Well, that's the question, and that's why this is so exciting! This is why I love the sort of science Swift does, because it is our job and mission, to discover strange new phenomena. It will take time, effort, and more data to work out what this is. Are my statements above wrong? Is it some exciting new class of object? Is it something routine like an AGN, that just happens to be lying behind M31, and had a small outburst (although if so, why did it trigger BAT?) Is it something like a type-I X-ray burst in our Galaxy, that just happens to be on the line of sight towards M31? The answer is, right now we don't know. But watch this space, and when we learn more, I'll let you know.

Disclaimer: All opinions expressed in here are my own and don't necessarily reflect the views of the University of Leicester or the Swift team. Any mistakes in here are also my own. As are any amazing Nobel-worthy insights.

Further information about GRBs: From swift.ac.uk | From NASA.