A Higgs Hunter's story...

[Note from Shaun: Here is Higgs hunter Mikko Voutilainen's account of the recent search for the Higgs. You can find the teaser to this post here. And my own, partially cynical, but ultimately upbeat, account of Higgs-things, here.]

Here it is, finally

[I assume the readers of this blog are somewhat familiar with the Higgs boson; if not, there's a nice summary on the CMS pages here]

So, this is the follow-up to the teaser I wrote a week ago. Now that everybody knows we found a Higgs boson at \(125.3\pm 0.6\) GeV, I'm free to talk about our finding, what it means and how we got there. Note the intentional use of 'a' Higgs there: although we, beyond reasonable doubt (less than one in a million chance of an error, to be precise), found a new particle, it's not 100% sure yet if it's *the* Higgs boson predicted by the standard model, or one of its many twins predicted by the hundreds of theories out there. There's even a tiny chance of it being an altogether different particle yet.

We actually already know a fair deal about this new particle besides the rather impressively precise estimate of its mass: it seems to be produced at a rate that matches the standard model prediction within about 20% uncertainty, it decays into bosons (W, Z and photon) and fermions (b-quarks and tau-leptons) roughly in the ratios predicted by the standard model, and in particular it decays into W and Z bosons in the ratio predicted by the standard model. The last point is rather important, because the Higgs mechanism, and the Higgs boson along with it, was invented to give mass to the W and Z bosons, and leave the photon massless. This also fixes the ratio of the decay rates to W and Z. If the new particle didn't decay into Z's and W's in just the right ratio, it couldn't be the Higgs boson we predicted.

We've also had a stab at determining the more abstract properties of the particle such as a quantum number called parity, but the statistics are low and the results still inconclusive. Predictions say we should be able to tell by the end of the current run, when we've collected 2--3 times the amount of data we have now. At this point we should also have more precise determination of the particle's decay rates in all the different channels, in order to gain more confidence in calling the particle a Higgs boson or something else.

So, is this the end, or the beginning of something new? I'm really hoping for the latter. If the new particle turns out to be 'just' the standard model Higgs boson and there's nothing new to be found, that would be fairly boring. If instead it's a Higgs twin, we may have just opened a window into a new landscape of particles.

At the moment it's too early to tell for sure, but there are a few interesting features to the way the new particle decays. It seems to decay into photons more often than expected, and to tau-leptons less often than expected. Taking all the decays to fermions together, they only seem to add up to about half of the rate predicted by the standard model, albeit with an error of about 50% as well. That coincidence is causing a bit of excitement nevertheless.

It might not be too bad for the standard model, though, it could just indicate that it's 'non-minimal'. While the Higgs coupling to W and Z is pretty tightly constrained, all the other particle masses are more of an ad-hoc addition to the theory, and there's some freedom to adjust how these particles couple to the Higgs boson without breaking everything else. Another good example of something that would require a 'non-minimal' standard model are the neutrino masses, which in the simplest expectation are exactly zero. We now know they are not zero, although we've still to nail down exactly how much they weigh (it's very very little in any case).

What for me was most interesting in this was to see first-hand how things have evolved towards a big discovery. Things started rolling about six months ago, when the first results from LHC Higgs boson searches were presented last December. Back then both ATLAS and CMS saw a hint of a Higgs at 125 GeV, with about 2-2.5 sigma statistical confidence. If you were a Higgs-believer, you could have given the signal more than 95% chance of being true.

After December it was decided that we wouldn't look at the 2012 data in the signal region before we had enough to confirm or refute the hint seen in 2011. This process is called blinding, and its important for making sure the analyzers are not unconciously affected by their prior expectations. Blinding is also one of the reasons we've tried to keep a lid on the results until today's seminar so that the experiments would not affect each other's findings between opening their signal box opening and presenting the final results. I think we were fairly successful in the end, although rumors started circulating on the blogs within days, and by yesterday almost every major newspaper (including Nature) had run a story on Higgs.

Between opening the signal box and seeing the first evidence of a new particle there was a whole lot of work going on for 2--3 weeks to prepare for ICHEP. The analyses added around 50% more data, the particle properties were studied in more detail, the CMS management had regular meetings with both ATLAS and CERN directors, people were working day and night to scrutinize the results, prepare documentation, etc. The final days were spent polishing plots, rehearsing presentations and fine-tuning press releases. Although I didn't happen to be at CERN during that period (I did attend the signal box opening in the beginning, though), I could at least participate through the almost daily video meetings and by keeping my own small piece of CMS running (I'm responsible for a team calibrating jets).

Just two days prior to the seminar there was also a presentation of the Tevatron Higgs results at Fermilab. The Tevatron people had done a superb job in squeezing every last bit of sensitivity out of their data and fell just a hair's width short of claiming evidence for the Higgs (they got 2.94 sigma by the most optimistic count, and needed 3.0). The Tevatron experiments collected data for ten years before shutting down last summer, and have the same amount of data (10 fb-1) available for analysis as the LHC experiments now. The lower collision energy of the Tevatron, 2 TeV versus 8 TeV at LHC, means roughly ten times less Higgs bosons are produced, but they still have better sensitivity in one single channel, the Higgs decaying into two b-quarks. I was watching that live on video, too, cheering for my old colleagues (I did my PhD on D0, one of the two experiments at the Tevatron).

And then, finally, today we had a chance to see how our colleagues and rivals at ATLAS were doing with their Higgs search. According to blog rumors, newspaper leaks and sensitivity estimate just a tad behind CMS, but never far. As it turned out, both CMS and ATLAS came up with the same significance in the end, within 0.1 sigma precision. Both experiments have now just made it to the 5-sigma milestone, and it's pretty clear that the signal has been effectively confirmed by at least three experiments (counting D0 and CDF together as a single Tevatron experiment).

P.S. I wrote a lengthy story about the box opening the same evening when I was at CERN, and stored it on a time capsule on my e-mail account. I'm not sure if it's interesting anymore, but at least I shouldn't be breaking any confidentiality rules by releasing it. [Shaun speaking: I now have this item in my possession, so if anyone wants to see it please let me know and I will upload it in a few days.]