Science —

Possible Higgs boson signals, but we won’t know for sure until next year

If the Higgs exists at anywhere near the energies that we think it must, then …

A collision that produced four high energy electrons, a possible sign of the Higgs boson.
A collision that produced four high energy electrons, a possible sign of the Higgs boson.

This morning, the spokespeople for the two main detectors at the Large Hadron Collider, ATLAS and CMS, gave talks on their teams' latest results in the search for the Higgs boson. As expected, the results were a bit ambiguous, as the signals that are consistent with the presence of the Higgs didn't rise much above two standard deviations away from background noise. But the details were even more confusing. Although both teams see signals in roughly the same area, CMS sees two of them, and appears to exclude the area where ATLAS' signal peaks.

The Higgs boson is predicted to be the last undetected particle of the Standard Model. It's a necessary outcome of the Higgs field, which provides the other particles mass. There are other ways of producing a Higgs field (and other ways of having a Higgs-like particle), but all of these require extensive modification of the Standard Model. Since the Standard Model works extremely well, researchers have been searching for the expected Higgs particle; the LHC was designed in part to be able to find it.

But doing so may require several years worth of collisions (and if you don't understand why, see yesterday's article). The person who described the CMS results, Guido Tonelli, started his talk with that caution: based on the amount of data we have, we'd only expect any signal to, at best, reach 2-3 sigma from background noise. And both speakers (ATLAS was represented by Fabiola Gianotti) described the challenges involved in finding any signal at all.

The Higgs can be produced by a variety of mechanisms (my personal favorite: "gluon fusion"), and falls apart by many more. Both detectors can search for the products of different forms of decay, which are called channels. Some of those are relatively easy to identify, such as decays that produce four high-energy leptons (electrons, muons, or taus) or two high-energy gamma rays. Others are a bit more challenging; a decay pathway that involves W bosons ends up producing neutrinos, which will pass through the hardware undetected and don't have a well defined mass. Different channels are more or less effective at different energies; for energies below 140GeV, the three channels I just described are the most significant.

Known, non-Higgs processes described by the Standard Model also generate similar-looking events, so the key is to be able to identify a Higgs signal in these channels that stands out above this background.

For most of the area that the detectors have looked at, there's nothing that does stand out. ATLAS has elimated the entire region from 130-476 GeV with a 95 percent confidence. CMS has gone even further, eliminating up to 600GeV. On the low end, the CMS team has found that the Higgs must be lighter than 127GeV.

And that's where things start to get odd. For ATLAS, the area with the largest peak above background is centered right at 127GeV. It's fairly broad, though, and extends a bit below that. CMS, in contrast, has its largest peak at 124GeV—close by, but clearly at a point where ATLAS' peak has started to tail off. Even more confusing, CMS sees a weaker but still distinct peak at about 119GeV.

How significant are these results? Locally, the ATLAS peak is quite large, reaching 3.6 sigma. But as part of a global Higgs search, the Look Elsewhere effect drops the significance down to 2.4 sigma. (The Look Elsewhere effect accounts for the fact that if you examine a large range of energies, the chance of seeing a peak like this at random goes up.) For CMS, the global significance is even weaker, dropping the peak to 1.9 sigma. For physicists, it takes a three sigma signal to even claim evidence for seeing a Higgs, so we're not even at the stage where we have evidence yet.

But it's clear that, if the Standard Model Higgs exists, it's going to be right around 125GeV. And, if the LHC performs up to expectations, we should have a much better idea of whether it is actually there by the end of next year's run.

So far, the performance has been excellent. About 98 percent of the electronics within each detector are operating as they should, and both are capturing over 90 percent of the collisions provided by the LHC. ATLAS' Gianotti said that, for measurements of some standard model particles, the LHC has already reached precisions that it took the Tevatron 20 years to generate. She also showed a spectacular image of a proton bunch crossing that produced 20 collisions nearly simultaneously, and said that they were averaging a dozen collisions for every bunch crossing before the machine was shut down for the winter. So there's every reason to think that next year's data will be decisive.

What would a Higgs in this mass range mean? At a press conference afterwards, Tonelli, the CMS spokesman, suggested that it would require a heavy particle to act as a sort of "bodyguard," for reasons he didn't explain. But he said it could be very heavy, possibly well beyond the range of masses that the LHC can produce. If we get an actual Higgs mass to work with, it's possible that the theorists can now narrow this range down a bit—if they can stop themselves from coming up with extensions to the Standard Model that call for the particle's existence, at least.

One small aside that may interest only me. If the Higgs mass turns out to be 125GeV, that would mean it's lighter than the top quark, and thus well within the range of the Tevatron, energetically. Unfortunately, the huge amount of noise at these energies kept the Tevatron from ever spotting it, which reinforces the importance of the statistical processes used to identify particles.

Listing image by Photograph by CMS/CERN

Channel Ars Technica