c_gamma=c_0 + a ( E_gamma / m_p) + b ( E_gamma / m_p)^2 + ...,
with c_gamma the speed of a photon of energy E_gamma, c_0 the speed of light ( for low energy photons), m_p the Planck mass and a,b are some parameters determined by your model. This style of papers then sets limits on a,b. And the general expectation of some space time foam model is, that a,b should be roughly 1.
From this the way they can probe the Planck scale is then simply a huge baseline, in this case 7e9 light years, compared to a rather short duration of the event ( ~ days).
> And the general expectation of some space time foam model is, that a,b should be roughly 1.
OK. Any idea how generic this expectation really is?
> compared to a rather short duration of the event ( ~ days).
What's this timescale have to do with it? They must be assuming they can identify photons emitted at the same time since they're measuring the arrival time difference in milliseconds.
[Omitting a longish rant, that once we give up Lorence invaraiance we do no longer know anything.] AFAIK it is essentially the fine tuning argument. So not really generic but in the absence of a good argument a reasonable default value.
> What's this timescale have to do with it?
The timescale was essentially meant illustrative, so a short everyday timescale compared to gigayears. ( And it is conservative, in the sense that I do not need to remember the exact details of a GRB spectra to use this one ;)
> OK. Any idea how generic this expectation really is?
I can't comment on this case specifically, but the entire notion of the "Planck energy" is based on that idea: that once you've found the combination of fundamental constants with the right units, the numerical coefficient that multiplies it will wind up being within one or two orders of magnitude of 1. By some miracle, the vast majority of systems in physics seem to match such expectations.
Here are a few relevant publications from Fermi:
https://www.sciencemag.org/content/323/5922/1688.full
http://www.nature.com/nature/journal/v462/n7271/pdf/nature08...
There's a difference in method between the Fermi and Nemiroff methods. The Nemiroff method is novel, but with lower statistics. Nemiroff himself has stated the results are not yet statistically significant. More short, bright GRBs need to occur.
I don't understand how photons with wavelength ~ 10^-17 meters are supposed to probe the planck scale, p_l ~ 10^-35 meter. I've skimmed the paper, but it seems most of that explanation is buried in references 11, 14, and 15. (Especially 15.) From their language, my hunch is that this is rather model dependent, i.e. the observations only bounds specific models of spacetime foam which happen to have the nice property of amplifying the dispersive effects.
(I'm a physicist, but this is way out of my area of expertise.)