The Higgs boson Part 2: Discovery of the Higgs
On July 4th 2012, the ATLAS and CMS collaborations at the Large Hadron Collider (LHC) at CERN, Geneva, independently announced the discovery of a new particle with a mass over 130 times greater than a proton. This was in the range that had been predicted for the elusive Higgs boson, and it had been verified to extremely high certainty.
But what do we mean by ‘high certainty’? In Particle Physics everyone agrees on how sure you need to be to declare something as a discovery, and it’s quite stringent. Firstly, we assume that there is no new particle and calculate the probability that the signal we measure is consistent with there not being a particle there. If that probability is less than 0.32 than we say we are 1 sigma certain, 0.046 = 2 sigma (this is usually good enough for most science), 0.0027 = 3 sigma… We could stop at 3 sigma, as 0.27% chance of being wrong seems awfully small, but there have been a number of possible particle discoveries at 3 sigma level that later turned out just to be statistical anomalies. So just to be absolutely sure, the usual threshold to claim a discovery is 5 sigma (0.0000003 chance of observed signal being consistent with no new particle hypothesis).
Right, so we’re now pretty sure there is a new particle (that is a boson). But is it the Higgs boson? Luckily, by March 2013, evidence surfaced that the new particle is what we call ‘spinless’ (meaning it looks the same from every direction) and that its interactions match the theoretical predictions of the famed Higgs boson. Further testing will take place from 2015 onwards, as the LHC ramps up towards its maximum energy of 14 TeV for the first time (double the energy it was running at in 2011 – more on particle energies in a future blog post).
(Next time: Exotic Higgs decays)
Can you tell an exact date of the restarting of the LHC?
I don’t think there’s an exact date at the moment, just ‘Spring 2015’.
The Atlas front page says April 2015. Note that is the start of the Physics run. The process of restarting has already begun. Restarting something as big and complex as the LHC is not just a matter of turning a key in the ignition. More details here if you’re interested: http://home.web.cern.ch/about/updates/2014/06/cern-announces-lhc-restart-schedule
Another really good blog post, Ryan; well done!
The way you’ve written it, it seems that the key PDFs must be Gaussian; is that so? And if so, why?
I mean, with such an extraordinarily complicated experiment as the LHC, with surely a bazillion systematics to deal with, why would the PDF end up being Gaussian?
The Gaussian is used to define what we mean by the statistical certainty required for something to be 5 sigma (i.e. it’s a theoretical construct). You are right that the pdf of real experiments can be more complicated, but quite often we can just use the central limit theorem for large N to reduce every pdf to a Gaussian.