Two of a Kind
It's extremely rare for the two sister collaborations to disagree in their findings, although it's not unprecedented. According to Fermilab Today, over the last decade CDF and DZero have agreed on their results more than 99% of the time, spanning over 500 different ultra-precise subatomic measurements. They are specifically designed to to check each other's results.
Both DZero and CDF are part of Fermilab's Tevatron collider, but the experiments are located at different points on the accelerator ring, and track different collisions, although the same physical processes occur at both locations.
The detectors are somewhat different in design, and in the selection criteria used to analyze events. So there will be differences in how they respond to a specific particle, like an electron or muon. But both collaborations look at the same types of collisions and measure the same quantities, and both are looking for the elusive Higgs boson.
"They gather their data and do their analyses completely independently, and make slightly different choices about how to select which data to record and analyze," Ann Nelson, a particle physics theorist at the University of Washington, told Discovery News. That said, "The data are from the same underlying microphysics and from collisions of the same beams at the same energy, so the conclusions drawn should be the same if both are doing things correctly."
It is certainly possible for CDF to see something that DZero has missed, according to Gordon Watts, a physicist with DZero. "When we first heard about [the CDF result], one worry was that our different approach had led us to miss something that CDF was seeing," he told Discovery News. "This turned out not to be the case, but sometimes a different way of looking at the data can lead to a discovery your competitors failed to make."
Where's the Error?
So why aren't DZero and CDF agreeing in this case? The most obvious explanation is that the CDF physicists made a subtle error in their modeling of the background events. Nelson emphasizes that scientists on both teams are being as meticulous as possible in their analyses. "However, the 'bump' [in question] is a small excess on top of a large background, so a slight mistake in the background measurement or modeling could be to blame."
Watts agrees with Nelson that the discrepancy is likely to be found in the background modeling. "Since we aren't looking for a particular signal, it all comes down to how you model the detector and the backgrounds and so forth," he says, acknowledging that it's possible the error lies not with CDF but with DZero. "In DZero, perhaps we grossly over-estimated our errors and that has hidden the signal. However, there are other physics results that we have published that make that all but impossible."
Really, the only certainty at this point is that one of the two collaborations will eventually prove to be wrong. "Here are two results, both carefully vetted by teams of more than 500 physicists, and there is an obvious disagreement," says Watts. "In six months to a year, either CDF or DZero will be wearing some egg on their face."
Fermilab will set up a task force to perform a comparative analysis of the results from these two experiments in agonizing detail, plot by plot, in hopes of understanding where the discrepancies lie. Fermilab Director Pier Oddone announced that this task force will include scientists from both CDF and DZero, as well as two Fermilab theorists. While the Tevatron will shut down operations in September, there is more than sufficient data to resolve the issue, he said.
As for the LHC...
Meanwhile, physicists at the Large Hadron Collider in Switzerland are analyzing their own data for signs of this mysterious potential particle. Thus far, there is no signal, although Guido Tonelli, a physicist with CERN's Compact Muon Solenoid (CMS) collaboration, told Nature that there is still too little data available for analysis to make a definitive conclusion. Early results from the LHC will be reported at the end of July during a high energy physics conference in Grenoble, France.
"Trust me, almost everyone, at some level, hopes that this is real," says Watts. "But we have to let the data speak, and so far, it is far from conclusive."
And what if the LHC sees the same signal as CDF? "We'll have a huge party!" Watts jokes. "Seriously, if they do see it, then all the papers proposing different models will be scoured for their distinguishing features, and all of us experimenters will run off to try to compare them with data. Very little sleep will be had. It will be a hell of a lot of fun."