Gravitational wave detection and electromagnetic counterpart Background Referring to this article on Fermi EM signal, 0.4 s after the gravitational wave detection by LIGO, FERMI detected an electromagnetic signal (poorly localized) with a false alarm probability of 0.0022.
Question I wanted to ask what the status of this discovery is currently. I heard some theorists have started debating. There seems to be a small consensus that the EM signal is not a counterpart for the gravitational waves? Or is it simply too early to start making conclusions, since we do not know if the EM radiation is from the same source or not?
My confusion This is somewhat hard for me to understand though. Since sub-parsec scales are not really resolved anyway, why would there be no possibility of there being matter circling the two black holes, which upon black hole merging annihilates away as a result of high energy particle collisions?
Edit: Additional question: Did IceCube detect any neutrinos at around the same time? EEdit: No they did not
 A: One computes probabilities for a coincidence , as the 0.0022 you quote, to give framework on the credibility. There is a chance of 1/455 to be a coincidence. So it is really necessary to wait for more gravitational detectors, synchronized so as to get accurate locations and not depend just on a time coincidence, as this, though not very probable, is also not at the five sigma level which is the establishement level in particle physics for new results, the Higgs for example.
A number of papers appear in arxiv.org on the subject and there will be no problem to associate gammas with gravitational waves from mergers in the future. For now only modelling can be done assuming correlation .
A: As @AnnaV says, there's no way to know for sure if this is a true association or not.  People are definitely skeptical of it... There will be many more similar GW detections, and that should tell us more conclusively.
In terms of surrounding material: it's not just an issue of having some gas around.  The real crux of the issue is that (we believe) you would need very strong magnetic fields to produce the type of emission possibly-observed, and the only way to keep strong magnetic fields on BHs is with lots of surrounding material.  In particular, the BHs would need to be accreting at near Eddington values (the highest rate possible), which shouldn't be possible for coalescing BHs.
The reason this is (believed to be impossible) is because the gravitational-wave emission makes the binaries come together very very quickly, and the mechanisms which dissipate energy in accretion disks act much slower.  While the BHs are farther apart, they can each (or both together) have a gaseous accretion disk feeding them material --- at which time they could have strong magnetic fields, and produce strong emission (believed to be via the Blandford-Znajek mechanism).  But as they come closer, the gaseous disk will eventually 'decouple' (it wont be able to keep up --- see image) and the magnetic fields should weaken --- and any emission along with it.

Simulation of a gaseous disk around a binary BH merger (Farris et al. 2012).
All that being said, this is an entirely new window into the universe... it's very possible there will be surprises that change our understanding!
A: As others have said, the association is tenuous. A follow up observation in, say, the radio band would help with localization, but it is expected that the EM signal will be too faint. Having said that, models have been proposed to explain how you might get EM radiation from a BH-BH merger. 
The two that caught my eye were:


*

*Perna et al. 2016. The idea here is that you start with two massive stars of a low metallicity. These stars undergo supernova explosions, but for one of the stars, the outer layers of its envelope remain bound at large radii. This forms an accretion disk. At high temperatures the gas in the accretion disk is partially ionized and accretion can proceed. As the temperature drops, accretion is choked and the disk simply persists as 'long-lived dead'. During the inspiral phase, tidal torques heat up the disk from the outside in. Inner regions are cooler and so no accretion can take place. Consequently the material at the larger radii piles up at the outer edge of the dead zone. When the tidal heating reaches the inner part of the disk, as the BHs merge, all the built up material can rapidly accrete resulting in a short gamma ray burst.

*Zhang 2016 Here you start with two BH, one of which has a small amount of change. During the inspiral phase this results in a magnetic dipole normal to the orbital plane, which in turn creates a magnetosphere. During the final merger phase, dissipation of the Poynting flux into the outflow results in a short GRB. Since this is dissipated at a large radius, the EM signal should be slightly delayed with respect to the GW signal, which is what's observed. Curvature radiation - radiation resulting from a charged particle moving along a magnetic field line could also result in a Fast Radio Burst, coincident with the GW signal. This model can be constrained through simultaneous GW, GRB and FRB observations.
Now as others have said, we obviously can't draw any conclusions from one data point. However I think it is interesting that such a phenomenon could occur.
It also slightly raises our hopes at the simultaneous detection of an EM signal with a GW signal from a NS-BH or NS-NS merger which is currently the established model for how we get short GRBs.
