jump to navigation

The inside story of the potential Higgs signal January 27, 2007

Posted by dorigo in Blogroll, internet, news, physics, science.

Offering simple reading material about experimental results in high energy physics, and commenting on their importance and implications, is a good idea and one of the motivations for running this blog, but simple explanations are not always enough to make a story entertaining. A story told from the eyes of the person in front of which events develop is always a tad more fascinating.  

Now, all that to say that the potential signal of a MSSM Higgs boson I discussed a week ago in this post  has been explained in detail at the cosmic variance blog  by one of the makers of the analysis.

John Conway (pictured on the left next to our detector), a colleague in CDF and a very smart experimental physicist who has been working on this very issue for twenty years now, tells the story from the inside. Quite a nice post, and highly advisable.

It remains to say that the Higgs boson signal CDF sees is most likely a fluctuation. Which is unfortunate for two reasons: one, because what could be a Nobel-prize discovery is not there. Two, because the fact that CDF has an excess of events at around 160 GeV of reconstructed Higgs mass means that the 95% C.L. limit the experiment is able to set on the production of that particle is much weaker than it could have been.

In these days -well, thirty years now- of failed searches for physics beyond the Standard Model, the minimal result one usually can still publish and be proud about is a so-called exclusion plot, where points of the parameter space for the studied new theory are excluded (at 95% confidence level, for instance) by the fact that the new particles predicted by the theory are NOT found.

Of course, as one reckons with the lack of a signal in the data and switches to setting a limit, one would then like to set the most stringent possible limit. When the data show no bump, the limit is better than when the data show upward fluctuations…

The plot on the left shows the plane of two parameters: the unknown value of the hypothesized MSSM Higgs boson mass, and the value of the parameter called “tangent beta”, that is the tangent of the ratio between the expectation values of the two higgs fields in the model. Huh, too technical ? Ok, think of tangent beta as a dial that defines how much the higgs bosons like to couple to (ok, say “attract”) particles of a certain kind, “down-type quarks and leptons”.

LEP 2 already excluded all the blue region of this plot. CDF has different characteristics, due to the different production mechanisms of the particle in proton-antiproton collisions (LEP collided electrons and positrons). The light purple region is the one CDF expected to exclude with the analyzed data. The dark purple region is instead what CDF can exclude: a much smaller region! So either the Higgs is there after all (at, say, tan(beta)=50 and Mh=160 GeV), or CDF has been very unlucky!


1. Tony Smith - January 28, 2007

Tommaso, you say “… either the Higgs is there after all (at, say, tan(beta)=50 and Mh=160 GeV), or CDF has been very unlucky! …
“tangent beta” … is the tangent of the ratio between the expectation values of the two higgs fields in the model. …”.

What you say seems to be set out in a bit more detail in the PDG review at http://pdg.lbl.gov/2006/reviews/susy1_s046.pdf entitled SUPERSYMMETRY, PART I (THEORY) Revised April 2006 by Howard E. Haber, which says in part:
“… vd [vu] is the vacuum expectation value of the neutral component of the Higgs field … that couples exclusively to down-type (up-type) quarks and leptons … the ratio
tan(beta) = vu/vd … is a free parameter of the model …”.

Since tan(pi/4) = tan(45) = 1 in which case vu = vd and the neutral Higgs field coupling is the same for all types of quarks and leptons,
since 45 is pretty close to 50 in terms of the plot shown on this blog entry,
does that mean that
if CDF has not been “very unlucky”,
not only
there is a Higgs bump around 160 GeV,
but also
the neutral Higgs does not distinguish between up and down types of fermions ?

If so,
could that mean that the Higgs might be a non-supersymmetric Standard Model Higgs,
that the supersymmetric stuff in the analysis was not necessary, but still was very interesting in that it indicates that nature favors SM over MSSM ?

My apologies if these questions are so naive that I may not be seeing some obvious stuff requiring supersymmetry, but I ask anyway because if I don’t ask, I don’t learn.

Tony Smith

2. dorigo - January 28, 2007

Hi Tony,

it is tan(beta) that is equal to 50 in my example above, not beta. For that choice of parameters, the Higgs fields have quite different vacuum expectation values on up-type and down-type quarks. tan(beta)=50 means beta very close to pi/4…

In any case, CDF data, even with a much more significant bump of the type discussed, would not be able to determine the value of tan(beta), since cross section uncertainties would be still large. For that you would have better study other decays as well.


3. Tony Smith - January 28, 2007

Sorry for the muddled stuff in my previous comment.
I guess it shows how wishful thinking,
in this case my preference for SM over MSSM,
can distort understanding a plot by naive people like me.

I should have said:

“Since for tan(beta) = 50 you have vu = 50 vd and
the neutral Higgs field coupling is very different for u and d fermions,
does that mean that,
if there really is a Higgs bump at 160 GeV and at tan(beta) = 50,
the SM (minimal Standard Model) is wrong,
the MSSM may be correct,
so that it would not only be finding the Higgs
but also finding the first clear evidence of supersymmetry ?”.

A related quesion might be:

How would the same data be plotted by a non-supersymmetric Standard Model analysis?

about the bump around 160 GeV in an H to tau tau CDF plot,
Tommaso said in his entry on this blog entitiled
“A 2.1 sigma eccess of MSSM Higgs! January 19, 2007”:
“… the bump is most likely a fluctuation. 
Unconfirmed voices claim that D0 sees a deficit of events
where CDF sees an excess. …”.

Over at Cosmic Variance,
John in a 26 January 2007 blog entry “Bump Hunting Part 2”
elaborated further, saying:
“… At the evening session … at Aspen [January 2007] … a colleague
from our competitor experiment Dzero at the Tevatron, Greg Landsberg,
a professor at Brown, …[said]…
Dzero had a deficit where we had an excess!
… Their method was quite similar to ours,
so it was hard to escape the conclusion that if they really had a deficit,
our excess is more likely to be a statistical fluctuation.
We’re still waiting to see the answer from them,
which should happen soon. …”.

John over at CV says that
the D0 method of data analysis was “quite similar” to that of CDF.
the MSSM is a lot more complicated than the SM and (as you can
tell by the naivite’ of my questions) I don’t have a much of an
intuitive understanding of the MSSM.

Could subtle differences between the two “quite similar” methods
lead to different background calculations around 160 GeV ?

Since 160 GeV is around the region of W pairs, could the
“quite similar” methods handle W pair stuff differently ?

Tony Smith

4. dorigo - January 28, 2007

Hmmm, if the H->tau tau signal is genuine, then it hardly fits in the SM, because of the too high xs*B(tau tau). Therefore yes, it would be definitely physics outside the SM. Not sure of whether it would fit the MSSM better than anything else, though.

A non-MSSM analysis would still have to cope with similar backgrounds in this particular final state. I would venture to say that one would not bother looking for direct H->tau tau production in the SM, due to the expected ridiculously small S/N ratio. So one would rather search for that signature together with a W or Z decay to leptons. This has not been done so far, as far as I know.

D0 have not blessed a new result on the same search, so I can only guess that they are indeed doing a very similar thing to what we have done in CDF: If they see no excess, well, John said it: it pretty much kills the CDF signal, no less than Delphi, Opal and L3 killed the Aleph bump back in 1999.


5. Anthony Thornton - January 31, 2007

” I have an profound interest in the Higgs field due to my studies
of many ancient and classical studies of math expression that
were for unknown reasons passed down through the great
arts, expecially those of the Symbolist period. I will come back
to this HF subject which is also identified in ancient learning
as a body/field known only as the NOTHING BODY SPACE
(NBS), being that the energy is so great that absolutely nothing
can be or exist next to it. Again, I will return..”

Sorry comments are closed for this entry

%d bloggers like this: