Saturday, 30 March 2013


The greatest hits of the passing month in HEP were the Higgs update from the LHC and the release cosmological results from Planck. But winter conferences is quite generally  the time for experiments to dump new results.  Here's a short summary of the recent important announcements that might have escaped you in the general turmoil.   

  •  CP violation in D meson decays from LHCb
    A year ago LHCb surprised everybody with an evidence for CP asymmetry in D meson decays. More precisely, they measured the asymmetry between the D0→π+π- and D0bar→π+π- decay widths,  and also between the D0→K+K- and D0bar→K+K- widths, finally quoting the difference ΔAcp between these two asymmetries so as to reduce systematic errors.  Although the standard model prediction for this observable cannot be reliably calculated with present techniques, more or less informed estimates locate the asymmetry in the 0.01-0.1% ballpark. The previous LHCb result was ΔAcp=(-0.82±0.21)%, which prompted some cautious talk of new physics.  The recent update  LHCb not only takes advantage of the increased statistics (from 0.6 to 1fb-1), but also adds a new category of events where the flavor of the initial-state D-meson is tagged using the muon from the semileptonic B decay that decayed into a D-meson (see the diagram on the right-hand side of the picture, only the prompt category with the pion tagging was used in the previous analysis).  The new LHCb result is a dramatic U-turn. The asymmetry in the prompt category went down to ΔAcp=(-0.34±0.18)%, while the asymmetry in the semileptonic category is measured with the opposite sign: ΔAcp=(0.49±0.33)%. The naive average of the 2 results yields  ΔAcp=(-0.15±0.16)%,   which means that not only the hint of new physics but also the evidence for CP violation in the D-meson sector have been completely wiped out.  For a detailed description of the analysis see here  and here.  
  • Dijet bump from CDF.
    In fact, in March we witnessed a real chainsaw massacre of hints of physics beyond the standard model: Neff in Planck, h→γγ in CMS, ΔAcp in LHCb, and finally the 145 GeV dijet bump  at the Tevatron have all gone away.  The last one is actually a pleasant surprise. Recall, that 2 years ago CDF found a dijet bump near 145 GeV in events featuring a W boson and 2 jets with the significance larger than 4 sigma. If confirmed, the bump would mean a new particle at the weak scale: maybe a hadrophilic Z' boson, maybe a technipion, etc. But D0 did not confirm the bump, and there was a danger that the disagreement between the Tevatron experiments would never be resolved. Moreover,  testing this anomaly at the LHC will probably never be possible due to an overwhelming standard model background of  W+jets events. It seemed that we would have to live with this ghost forever. Luckily, CDF went back to the analysis and found a bug. The main culprit seems to be a difference in the detector response to quark and gluon jets, not taken into account in the previous analysis. Also, CDF improved modeling of the  multijets background where one of the jets fakes a lepton from  W boson decay. Once these improvements are included, the 145 GeV bump vanishes completely. One less thing to worry about. More importantly, the fact that CDF now understands well the W+jets spectrum adds some more confidence to Tevatron Higgs searches which rely in part on similar analysis techniques.   
  • μ → e γ  from MEG. 
    On the precision front, there was an important update from the MEG experiment which searches μ → e γ decays.  The standard model predicts this decay should be too rare to be observable  (a tiny branching fraction is induced via the neutrino mixing). On the other hand, it is straightforward to produce a large branching fraction in models with new sources of lepton flavor violation, including supersymmetric and composite Higgs models. The latest MEG update sets the limit on the branching fraction at  5.7x10^­‐13 at 90% CL, which represents a factor of 4 improvement of the previous limit. In any case, precision measurements of rare SM processes seem to be the most promising way to go ahead for particle physics. Indeed, our options to increase collider energies are very limited, whereas in many areas precision can be improved by orders of magnitude at a much lower cost. The hope is that one of these days one of these experiments will observe a non-standard process which will give us a clue about physics beyond the standard model. 
  •  Stops from the LHC
    Searches for supersymmetry no longer provoke a nervous anticipation, rather a weary frustration at the more and more stringent limits. Two important updates using the full 8 TeV dataset came shortly before the Moriond conference. One is from ATLAS searching for direct production of stop pairs, where each stop  decays to a top quark and an invisible neutralino.  The limit on the stop mass is almost 700 GeV, except in the kinematic region where the sum of top and neutralino masses is very close to the stop mass.  Another important update is the CMS search for pair production of gluinos, where each gluino decays via an off-shell stop to 2 top quarks and an invisible neutralino. In this case, the limit on the gluino mass reaches 1.3 TeV. The limits may be somewhat improved in the future by including more channels, but they already give the feeling of the final reach of the 8 TeV LHC. The two searches mentioned above are important for theorists because they probe the presence of stops near the TeV scale, which is the necessary condition for supersymmetry to address the naturalness problem of electroweak symmetry breaking.  Taken together, the SUSY results gradually force us to accept that supersymmetry cannot completely explain naturalness; even if it's just behind the corner it is getting more and more in difficult in most realizations to avoid fine-tuning at the 1-10% level.  See here for more details of the CMS and ATLAS analyses.


Thursday, 21 March 2013

The universe after Planck

Here's the 2013 rendition of the cosmological Mona Lisa, 

or, in our jargon, the multipole expansion of the Cosmic Microwave Background  (CMB) power spectrum. Compared to previous experiments, what distinguishes Planck is that it measures the power spectrum all the way from the largest angular scales down to less than 0.1 degrees.  Before, to cover the entire range, one had to combine several different CMB experiments: WMAP, SPT, ACT, which was more vulnerable to  unknown systematic errors (indeed, the results from SPT and ACT were not completely consistent). Thanks to Planck, the errors are reduced, especially at large multipoles,  and the confidence in the results is strengthened.

For the general public, the most palatable piece of information is about the  composition of the Universe. The dough is made of 69.2±1.0% dark energy, 25.8±0.4% dark matter, and 4.82±0.05% baryons, after combining Planck with other cosmological measurements. As you can  see, the errors are of order 1 in 100, which represents a factor of 2 increase in precision compared to WMAP-9. The central values have shifted a bit: there's additional 2% of dark matter at the expense of dark energy, but the change is consistent within 2 sigma with previously quoted errors.

From a particle physicist's point of view the single most interesting observable  from Planck is the notorious Neff. This observable measures the effective number of degrees of freedom with sub-eV mass that coexisted with the photons in the plasma at the time when the CMB was formed (see e.g. my older post for more explanations). The standard model predicts  Neff≈3, corresponding to the 3 active neutrinos. Some models beyond the standard model featuring sterile neutrinos, dark photons, or axions could lead to Neff > 3, not necessarily an integer.  For a long time various experimental groups have claimed Neff much larger than 3, but with an error too large to blow the trumpets. Planck was supposed to sweep the floor and it did.  They find Neff=3.3±0.5, that is no hint of anything interesting going on. The gurgling sound you hear behind the wall is probably your colleague working on sterile neutrinos committing a ritual suicide.  
Another number of interest for particle theorists is the sum of neutrino masses. Recall that oscillation experiments tell us only about the mass differences, whereas the absolute neutrino mass scale is still unknown.  Neutrino masses larger than 0.1 eV would produce an observable imprint into the CMB. In fact, the SPT experiment recently made a  claim that  the sum of neutrino masses is 0.32±0.11 eV, a 3 sigma evidence for a non-zero value.  That would be huge, if confirmed. Well, no such luck:
Planck sees no hint of neutrino masses and puts the 95% CL limit at 0.23 eV. (Check out the comment section for a more informed statement).

Literally,  the most valuable Planck result is the measurement of the spectral index ns, as it may tip the scale for the  Nobel committee to finally hand out the prize for inflation.  Simplest models of inflation (e.g., a scalar field φ with a φ^n potential slowly changing it vacuum expectation value) predicts the spectrum of primordial density fluctuations that is adiabatic (the same in all components) and Gaussian (full information is contained in the 2-point correlation function). Much as previous CMB experiments, Planck does not see any departures from that hypothesis. A more quantitative  prediction of simple inflationary models is that the primordial spectrum of fluctuations is almost but not exactly scale-invariant. More precisely, the spectrum is of the form P∼(k/k0)^(ns-1), with ns close to but typically slightly smaller than 1, the size of ns-1 being dependent on how quickly (i.e. how slowly) the inflaton field rolls down its potential. The previous result from WMAP-9  ns=0.972±0.013 (ns=0.9608±0.0080 after combining with other cosmological observables) was already a strong hint of a red-tilted spectrum. The Planck result ns=0.9603±0.0073 (ns=0.9608±0.0054 after combination)  pushes the departure of ns-1 from zero past the magic 5 sigma significance. This number can of course also be fitted in more complicated models or in alternatives to inflation, but it is nevertheless a strong support for the most trivial  version of inflation.  

I was a bit surprised by how much emphasis in today's press conferences was put on the small glitches at low multipoles. It seems that Planck people are also a bit frustrated by the fact that their results are nothing but a triumphant confirmation  of old paradigms. Even at the LHC nobody would make a big deal of a 2.5 sigma anomaly, and in the present case we're in the area of astrophysics where error bars are treated more loosely ;-)  Moreover, according to Planck, the quadrupole mode in  the fluctuation spectrum is aligned with the ecliptic plane, which suggests some unknown background or pesky systematics at large angular scales. Of course, many a theorist will come up with a beautiful explanation  of  the low multipole anomaly. But not because  it's convincing, but because there's nothing else to ponder on...

In summary, the cosmological results from  Planck are really impressive. We're looking into a pretty wide range of  complex physical phenomena occurring billions of years ago. And, at the end of the day, we're getting a perfect description with a fairly simple model. If this is not a moment to exclaim "science works bitches", nothing is.  Particle physicists, however, can find little inspiration in the Planck results. For us, what Planck has observed is by no means an almost perfect universe... it's the perfectly boring universe.

Mind that this post is an outsider's perspective from the angle of particle physics. For a better insight into cosmological aspects of the Planck results see for example here. Note that Planck dumped 28 full-grown papers today, so browsing through it will take some time, and there may be some hidden treasures at the bottom of the chest..  

Tuesday, 19 March 2013

Higgs: more of the same

Everything's clear now.  The most important Higgs search channels have been updated by both ATLAS and CMS (the only interesting channel that has not is the decay to b-quarks).  We're not going to learn much new for the next 2 years.  The CERN director, heeding to Résonaances' callannounced Habemus Higgsam.

So, what's new since last week?

  • Finally, CMS came out with their long awaited update of the Higgs→γγ search. It is clear why they have been so shy: new calibration of the electromagnetic calorimeter completed after last summer forced them to revise the  significance of the Higgs signal in the diphoton channel. The current significance is merely 3.2 sigma, and from that it's easy to deduce that they wouldn't have been able to claim a formal 5 sigma discovery on the 4th  July had they done their sums back then. But that's of course irrelevant now, as there's no shadow of a doubt the Higgs is there at 125 GeV. The real game changer is the (related) fact the  signal strength relative to the SM one measured by CMS is μ=0.8±0.3 and, unlike the one in the ATLAS and previous CMS measurements, does not show any excess over the standard model prediction. Naively combining the signal strength measured by ATLAS and CMS one gets the disappointing μ=1.2±0.2. Move on folks, nothing to see here.  See Matt's blog for a more in-depth discussion of the CMS diphoton update. 
  • One visible consequence of the CMS updated is that  the preference for negative Yukawa couplings, displayed by the Higgs data before Moriond, completely vanished.  In the plot, borrowed from here,  is a section of the Higgs coupling space where one freely varies the coupling to W and Z bosons (denoted as "a"), and, independently, the coupling to the standard model fermions (denoted as "c"). Previously, the Higgs data somewhat  preferred  the bizarre region near a=1, c=-1  over the standard-model-like region near a=c=1, as the former was leading to an enhancement of the  Higgs→γγ rate. This is no more, only the standard-model-like island remains.   
  • Another important update last week was from ATLAS in the WW→2l2ν channel.  Again no surprises here: the signal significance almost 4 sigma, the rate μ=1.0±0.3.
  • To say that the Higgs is standard-model-like is an understatement.  This bastard screams and spits standard model.  After the Moriond updates the standard model gives an absolutely perfect fit to the combined data (previously it was disfavored at 80% confidence level, mostly due to the late diphoton excess).  Not even a single cliffhanger to makes us wait for  the next episode.....  If there's anything non-standard about the Higgs couplings to matter it is hiding very well and will be tricky to  uncover at the LHC, even after the energy upgrade. 
  • It's worth stressing again that is has been firmly established that the Higgs couples to mass of  W and Z bosons, the statement which can be reinterpreted that the Higgs gives mass to W and Z bosons.  After Moriond the evidence for that is strengthened even further.  The LHC Higgs data alone (blue in the plot) imply this coupling should be within ~30% of the standard-model value at 95% confidence level. If one also takes into account precision observables from LEP  (red in the plot) the allowed range shrinks to ~10% (although the latter number is a bit less robust, as fine-tuned new physics effects could alter the conclusions). Although announcing the Higgs is a Higgs at this particular moment of history was to a large extent arbitrary,  it was by no means premature.   
  • We see the Higgs, of which the corollary is that we don't see the invisible Higgs. The invisible decays could arise for example if the Higgs couples to dark matter, which is the case in a large class of well-motivated models. In the simplest situation, when the Higgs couplings to known particles take the standard model values, an invisible width would lead to a universal reduction of the rates in all decay channels. That is severely constrained by global fits, as seen in the plot (solid blue): the invisible branching fraction more than about 15-20% is strongly disfavored. Note that for the sake of this analysis "invisible" is "anything not measured", thus any non-standard decay channel is constrained this way. But that doesn't mean there's no point looking for exotic Higgs decays. In the end, we've already observed much more rare decays: the branching fraction to 2 photons is around 0.2%, and to 4 leptons  - around 0.01%. So 10% of invisible or some other freak decays seems like a goal well within reach of experimentalists.       
  • A fun fact for dessert. In CMS the measured central values of the Higgs rates in the 3 most sensitive channel all fall slightly below 1, while in ATLAS they are all  slightly above one.  The new motto for ATLAS: we try harder ;-)

Saturday, 9 March 2013

Higgs: what have we learned

Last Wednesday at the Moriond Conference the LHC dumped several new Higgs results based on the entire collected dataset of about 25 fb-1.  So, what do we know that we didn't know before?    

  • One thing I learned is that Higgs remains exciting enough to make me wake up at 8am. But barely so. Indeed, the time of the revolution is over, we are now entering a 20-year long period of incremental progress and painful squeezing of experimental errors. The best evidence of declining enthusiasm is that it took me 3 days to scribble a summary for Résonaances.  Also the experimenters seem to be less excited, as they didn't bother to update all the important channels in time for Moriond. Soon the Higgs will become just an annoying background to more interesting searches ;-)
  •   The more we look the more standard-model-like it seems.  Early on in the game there were several glitches, but with time all the Higgs observables are converging toward the standard model prediction. The only worth-mentioning discrepancy that remains is the h→γγ rate measured by ATLAS. But even that is getting less compelling. The diphoton rate normalized to the standard model one drifted from 1.80-0.36+0.42 last December to 1.65-0.30+0.34 in Moriond. That means the significance of the excess didn't grow after adding more data, which is usually a bad sign. Besides, ATLAS still struggles with the twin peaks -- the 2.5 GeV discrepancy between Higgs masses measured in the γγ and ZZ→4l channels -- which suggests that an unknown systematic error or at least an unlucky fluctuation affects one or both of these measurements. Anyway, next week the air will be clearer when CMS says what they have to say. Except for ATLAS γγ, the rates in the remaining channels updated for Moriond (ATLAS ZZ, CMS WW ττ) are within 1 sigma uncertainty from the standard model prediction. 
  •  The single most interesting piece of news was the CMS update in the h→ττ decay channel. Almost 3 sigma evidence for a signal brings up to 4 the number of channels where we clearly observe the Higgs signals (I'm not counting  the bb channel, as the Tevatron no longer claims an evidence there). It's also the first direct evidence that the Higgs couples to fermions, although indirectly we knew that before (the rate of Higgs production via gluon-gluon fusion is roughly consistent with the process being  mediated by a top quark loop). It's also an evidence that Higgs couples to down-type fermions, which excludes a chunk of the  parameters o 2-Higgs doublet model    
  • I'm starting to get really annoyed by the endless talk of  "Higgs spin determination".  As if there was any serious issue at stake.  I've said it in the previous post, and I'll say it again: there's  absolutely no way the 126 GeV particle has any other spin than zero.  We don't even have a working model for a light spin-2 resonance that would be theoretically sound and not excluded experimentally. At some level, too much  caution may become unscientific.   Well, in the end we can't be 100% sure whether Newton's law is responsible for the motion of the planets in the solar system; it might be 7 dwarfs pulling the string in exactly the way as to mimic the inverse square law....  The question is a bit better posed for "Higgs parity determination", as in that case we at least have a sensible model for a pseudoscalar. But even then, a pseudoscalar  simultaneously reproducing the observed rates  in the γγ, ZZ, WW, and ττ channels is absolutely improbable. Of course, the above doesn't mean that the measurements involved in "spin and parity determination" are useless.  On the contrary,  it is perfectly conceivable that the Higgs has some (subleading) non-standard interactions with matter and those measurements are in a position to reveal them.  However,  they should not be viewed as measuring the spin of the Higgs  but as testing whether differential distributions for Higgs production and decay agree with the standard model.  A better way to present results is the one in the CMS plot here that shows the limits on the variable fa3 -- the relative contribution of the pseudoscalar interactions to the h→ZZ decay amplitude squared. From that we read off that fa3 larger than 58% is excluded at 95% confidence level, already a useful piece of information for constraining Higgs interactions.
  •  It's also worth pausing for a moment on Higgs searches in channels where we should not see a signal, because the standard model either forbids that particular decay or predicts too small a rate to be detectable with the current amount of data. In this category are the h→Zγ, h→μμ, and h→invisible searches  presented in Moriond. Quite generally, this is a very  promising direction for new physics searches, probably more promising than measuring the Higgs couplings. Indeed, the precision with  which we can measure the Higgs couplings is limited not only by statistics and experimental uncertainty but also by theory. Predicting the Higgs rates in a hadronic collider with a precision better than 10% may be tough, and we're already not too far from that level of experimental precision.  On the other hand, it's entirely possible that Higgs has some non-standard interactions that open a new decay channel or lead to an order-of-magnitude enhancement of the existing one. Observing such a rare decay process would be a clear smoking gun of new physics. Obviously, no luck so far, but I personally consider searching for rare non-standard Higgs decays our best chance for finding new physics at the LHC. 
  • The idea  of the sausage game is take a movie title and replace one word with sausage, for example  Harry Potter and the Deathly Sausage, Star Wars: The Sausage Strikes Back, Saving Private Sausage, etc.  The organizers of Moriond propose a similar parlour game with the Higgs.  Indeed, the Higgs jumps on us everyday from  the arXiv, from newspapers and TV, sometimes I open the fridge and see the Higgs. That is very tiring in the long run, especially if you have to listen to a string of Higgs talks after a day of skiing. So, to make the talks more attractive to listeners,  last year the speakers were asked to replace Higgs with the BEH sound; this year the word was SMS which, I must say, is less captivating. The rumor is that next year Higgs will be replaced by "Little Bunny".  The talk about  "Combined searches for Little Bunny in ATLAS and CMS" will definitely attract some a lot of attention :-)