Banner
    DZERO: 23 MeV Uncertainty On W Boson Mass
    By Tommaso Dorigo | March 2nd 2012 05:58 AM | 15 comments | Print | E-mail | Track Comments
    About Tommaso

    I am an experimental particle physicist working with the CMS experiment at CERN. In my spare time I play chess, abuse the piano, and aim my dobson...

    View Tommaso's Profile
    Just a few days ago the CDF collaboration announced their new measurement of the W boson mass, with a considerable improvement over the precision of the current world average for that quantity. Now DZERO, the competitor experiment, has also published their own new measurement, which is based on a statistics of 4.3 inverse femtobarns - twice as many data as the ones used by CDF.

    The new DZERO measurement is quite compatible with the CDF result: DZERO finds M(W)=80367+-26 MeV. The uncertainty is larger, mainly because of their systematic uncertainty on the electron energy calibration. Note that this measurement only exploits the W -> eν final state, while CDF uses both electron and muon final states.

    The figure below shows the fits to two of the three kinematical quantities sensitive to the W boson mass: the transverse mass of the electron-neutrino system (left) and the electron transverse momentum distribution (center). The red points are experimental data, the blue curve is the fit result, and the black histogram represent the estimated background component. At the bottom there are the residuals of the fit, showing that the modeling of the observed distributions is generally very good.



    DZERO combines the new result with its own earlier determinations of the W mass, obtaining the result

    M(W)= 80375 +- 23 MeV           (DZERO combined)

    Combined with the CDF result and the other past measurements, this should end up allowing us to know the W mass to a precision of 13 MeV or so, i.e. less than two parts in ten thousand. I am eager to see the new world average and the resulting global Standard Model fits... Stay tuned, they can't take long!

    Comments

    Does a higher precision W mass mean the possibility of a standard model theoretical discrepancy with the possible Higgs signal?

    dorigo
    Hi Anon,

    while in principle it could, we already know that it doesn't. Things are very consistent, and the global fits will not move much any more, unless some new result with significantly better precision on MW, or Mtop, or sin2thetaW casts a monkey wrench in. I expect none of the above of course.

    Cheers,
    T.
    Of course!
    You probably already know how that this result should be improving the consistency of the SM.

    Hi Tommaso, nice post.

    I suggest you to change the name to '23 MeV uncertainty ...':
    just reading the title I was expeting a 'bias' of 23 MeV somewhere,
    and maybe I'm not alone.

    I too read it that way. I was worried to find a post full of educated speculation on which experiment group is more likely to be correct. I wasn't in the mood to read that so I almost didn't click on it. Glad to see there was no real problem!

    Halliday

    Yes, when I read the title, I was expecting to see that the DZERO collaboration was correcting a previous error of 23 MeV.

    Perhaps something more on the line of 'DZERO:  Error on W Boson Mass Decreased to 23 MeV'.

    David

    dorigo
    Ok, it's your native language after all... I think we should educate the public on the fact that every measurement comes with an error, but I acknowledge that there is a consensus here. So I changed the title.
    Cheers,
    T.
    Halliday

    Tommaso:

    You are absolutely correct that "we should educate the public on the fact that every measurement comes with an error" (and/or uncertainty).

    Another, highly related issue that "we should educate the public on" is "the fact that every" modeling result "comes with an error" and/or uncertainty.  Unfortunately, many models express only a set of values with no indication of error/uncertainty.

    Even worse, perhaps, on both counts (measurements and modeling results), most decision-makers/management-types don't even want to know that there are any such errors/uncertainties:  They want definite, black and white answers.  They're like lawyers, they don't like all the caveats and other conditional statements of scientists.

    David

    By the way, also the "style guidelines" of the CMS and ALEPH experiments (and probably many other HEP collaborations, but I've only been a member of these two) insist that one should always talk about "uncertainties" and not about "errors", condemning the latter as jargon.

    dorigo
    Interesting - I was not aware of this.
    Cheers,
    T.
    Is this measurement sensitive enough to measure the Pion contribution to the W mass? It's on the same order of magnitude as the uncertainty here, right?

    dorigo
    Measure the pion contribution ? What exactly do you mean ? Any reference ?
    T.
    Well, the pions are pseudo goldstone bosons of quiral symmetry breaking, so they get absorbed by the W and Z. It's like technicolor, only at the MeV scale. Admitedly i'm not sure how this goes on in the case where the pion has a non zero mass, but I suppose you get some linear combination that's the physical pion triplet and another that gets absorbed.
    Here's a ref.: http://www.isv.uu.se/graduate/fpp.pdf

    Wow, those are clean samples. No QCD at low m_t -- yet the article does not mention how they remove it?

    dorigo
    Hi John,

    indeed, clean samples. Electrons are isolated tracks with a corresponding matching electromagnetic energy cluster, with little or no leak in the hadronic compartment. Once one selects that, and asks for significant missing transverse energy, a jet veto, and other generic cleanups, the data really is mostly W->ev events.

    This is not the cleanest sample at a hadron collider however. Z->μμ samples are among the cleanest signals you can get - at the peak the S/N is of the order of 1000:1.

    Cheers,
    T.