This is just a short update on the saga of the anomalous excess of W-boson-pair production that the ATLAS and CMS collaborations have reported in their 7-TeV and 8-TeV proton-proton collision data. A small bit of information which I was unaware of, and which can be added to the picture.

As you might known, recently the interest for that mild departure of WW production rate from standard model production has been boosted by the publication of a couple of theoretical studies, which pointed out how the kind of excess observed could well be explained by the presence of a concurrent signal of supersymmetric particles in the sample. A not-too-weird realization of Supersymmetry would indeed contribute to the data used by ATLAS and CMS in their WW cross section measurements. I wrote about this here three weeks ago.

Following those studies, another group of phenomenologists pointed out how there was a more natural explanation for at least a part of the experimental excess: namely, the selection of events with no energetic hadronic jets that both experiments operate to enhance the fraction of WW production in their datasets could lead to a disagreement from model predictions, as the model predictions are computed with a technology that might be ill-suited to describe properly that corner of phase space. You may find more information on this in another recent post.

Now the latest bit of information concerns the CMS measurement of the 8-TeV WW boson cross section, which was published in January 2013 using only a part of the 8-TeV dataset.

I should first of all explain that the WW final state arises by standard model processes that involve the annihilation of a quark-antiquark pair, but it also sometimes arises thanks to the production of a Higgs boson. The Higgs boson may decay to two W bosons, creating the same final state. Of course, one can to some extent disentangle the two processes; but in a inclusive cross section measurement one might choose to not bother doing it. Further, as the CMS publication is dated January 2013, it is easy to imagine that the bulk of the work was done in the first part of 2012, when the Higgs boson signal was not yet an established one; hence the analyzers had to decide whether to "subtract off" a Higgs contribution or not. Subtracting a not-yet-confirmed signal would have been questionable...

So it was recently brought to my attention that the 8 TeV CMS measurement, found in excess of standard model production by roughly 20%, does include a contamination of Higgs boson decays of about 5%, which the theoretical prediction did not include. Hence, the disagreement is smaller than one could be led to infer from the measurement.

To be precise, the CMS measurement of WW production at 8 TeV reported in the paper is

σ(WW)_meas=69.9 +- 2.8 +- 5.6 +- 3.1 fb

(uncertainties are respectively statistical, systematic, and luminosity-driven ones), whereas the theoretical prediction for WW production alone is

σ(WW)_th=57.3 +2.3 -1.6 fb

but the paper adamantly says, just after quoting these numbers:


Additional processes may increase the production yield in the WW final state by as much as 5% for the event selection used in this analysis. Higgs boson production would give an additional contribution of about 4% of the cross section given above, based on next-to-next-to-leading-order cross section calculations for the H ->WW process [43] under the assumption that the newly discovered resonance [44, 45] is a SM-like Higgs boson with a mass of 125 GeV. Contributions from diffractive production [46], double parton scattering, and QED exclusive production [47]are also considered.

The quoted one is a small correction, but it is interesting to observe that the theoretical papers describing how a SUSY contribution could explain away the ATLAS and CMS excesses did not notice the fact. So if anything this new datum goes in the direction of further dismantling any hope that Supersymmetric decays could be hiding in the WW datasets.

UPDATE: in fact it is pointed out below that the paper by Rolbiecki et al. discussing the WW excesses do include the correction, unlike what I mistakenly wrote above - my justification for not noticing it is that I am presently on vacation on a greek island, with little internet connection, and I did not have the ability to double-check the paper... However, many -including myself- who went back to the experimental measurements and compared to the theoretical prediction directly, were deceived by the absence of a subtraction of the Higgs contribution in the 8 TeV CMS measurement. So I think it remains worthwhile to broadcast the correction...