Set your browser to use document fonts to see all symbols correctly.

Ian Hinchliffe: "Reconstructing SUSY at the LHC"

  1. LHC discovery of Susy (not necessarily minimal SUGRA)

    • How confident are we that LHC can discover Susy?

      Very if it is light enough to be reached.

    • What is the model dependence in making this claim?

      You have to work hard to construct a scenario where you cannot do anything. Provided that the cross-sections are reasonable (masses in the accessible range) and there is reasonable energy release in the decays, there will be no problem. About the only scenario that is a problem would be one where ALL sparticles are almost degenerate. e.g everything is 1 TeV ± 20 GeV. Then the energy release is so small that nothing much is visible, except for initial state gluon radiation.

    • What is the integrated luminosity required to rule out squarks or gluinos below 2.5 TeV? Can LHC tell us that SUSY is not relevant for EWSB?

      It is 10 fb-1 at 2 TeV. I would guess that 2.5 TeV is about 30 fb-1.

    • For what regions of parameter space are slepton/sneutrino discoveries and measurements possible?

      This is discussed in my talk. In cases where they are produced in the decays of squarks/gluinos the reach is large. If you have to rely on direct production, you can get to 350 GeV or so.

    • Assess the probability for LHC to discover the whole Susy spectrum, leaving LC with only refined measurements of masses and properties to do.

      The chance of finding all of the sparticles is zero. As I discuss in my talk, the heavier gauginos and Higgs bosons will be hard and direct production of sleptons is a tough game if they are above 300 GeV. However, if you ask "Assess the probability that LHC will be able to determine the SUSY model and predict the masses of all the undiscovered sparticles," then I think that the chances are quite high unless the squark and gluino masses are above some value where the event rates become small (say 1.25 TeV)

  2. Sparticle properties:

    • What precision is expected for sparticle mass determinations, and to what extent are model dependent assumptions needed?

      As I discuss in my talk you can often determine masses without reference to an underlying model. In these cases the precision is likely to be controlled by systematic effects which we would expect to be at the percent level for measurements involving jets and better for leptonic and photon measurements. Once you are in a model, the precision goes up as many measurements contribute to the model parameters and hence to the predicted value (an error) on any particular mass.

    • What are the expectations for LHC to measure quantum numbers of sparticle states? branching ratios? coupling constants or coupling relations (e.g. ge,n,W = g[e\tilde],[n\tilde],[W\tilde]) ? Mixings (stau, stop, gauginos)?

      Not enough work has been done to make a definite statement here. We have seen a few cases where more than one decay channel is open and one can then see the relative branching ratios. Generally one measures sigma x B. If the mass is measured then sigma can often be predicted with accuracy if one assumes that squark and gluinos couple with alphas. (We did a reasonable job on the top cross section). Gaugino mixing will be hard to see if is small and the heavier states are higgsino-like. (this is discussed in my talk). If the lighter states are higgslike you can get a cascade of squarks and gluinos and its easier. The cases where the mixing is quite big also easier because the squarks and gluinos are likely to decay into all of the states. Squark and slepton mixing is discussed in the next answer.

    • How well can the higgs-related parameters (m, tan(b) be determined from the sparticle properties? The trilinear couplings?

      The trilinear couplings are not easy for several reasons. In SUGRA models the value at the GUT scale is an irrelevant parameter, nothing at low energy is sensitive to it. In the models we have looked at we assumed that the trilinear terms entered multiplied by Yukaka couplings. That means that they only have an effect for third generation where stop sbottom and stau mixing are important. (Remember that b-squarks are usually easier than light squarks as the jet combinatorial background is much less) The mixing can be inferred from the masses, if these can be reconstructed and from the fact that both eigenstates will decay to the same final states (Remember that qL and qR usually have different decay modes if they are lighter than a gluino)

    • How can CP-violating phases beyond CKM be measured?

      No idea. You might get big statistical samples so there is always hope. There was some work by Hall et al.

    • What differences in mass between chargino and Su(2) neutralino in anomaly-mediated susy breaking are accessible? How?

      In the range expected by current models, it is very hard. If the mass is outside this range then two signals open up. If the mass difference is very small, a stiff track that goes some distance and then appears to stop (as the resulting charged particle is too slow) should be easy to see (but not to trigger on, however the rest of the event provides an adequate trigger). If it makes it to the muon system, time of flight an be used to measure the mass very accurately. As the mass difference increases, one can hope to see the decay products. This anwser should not be interpereted as a statement that these models are hard to investigate; only that the mass differance between these two states is difficult to measure.

  3. LHC-specific detection issues

    • What will be the impact of multiple event pileup on discovery or mass measurements?

      None on discovery as this comes from events with energetic jets, leptons and missing ET. Precision mass measurements that involve jets will be affected a little. The hardest is the class of models where hadronic t-id is needed for relatively low pt taus. Note that unless the masses are large, the event rates at "low luminosity" are more than adequate for detailed studies and measurements. Direct production of sleptons and gauginos needs a jet veto to remove the SUSY background from squarks and gluinos. We expect that this jet veto will have a higher threshold at high luminosity due to pile up effects.

    • What precision for jet energy scale is possible in LHC detectors; how will it be established, and what is the impact on sparticle mass measurements?

      A percent or so. I refer you to Chapter 12 of the ATLAS physics TDR for a detailed discussion of jet energy (and other) calibration issues. It is likely to be the dominant source of systematic errors on the squark and gluino masses.

    • Could a LC make measurements that enable subsequent reanalysis of LHC data to make new advances? (or vice versa) We are trying here to establish the potential synergy of having both types of machines. (any examples of such complementarity?)

      If this question is a polite way of asking whether LHC will have thrown out a signal because of a trigger selection, the answer is no. The very detailed studies have not revealed any signal which would have escaped a preexisting trigger. The precise knowledge of the LSP mass would aid in model independent mass reconstructions. Our current thinking is that we can not do better than 10% on the LSP mass unless a model is used. New things seen at an NLC would certainly cause one to take another look (just as the e-jet events at HERA caused CDF and D0 to look carefully) and would almost certainly suggest new analyses. Here is an extreme example where NLC could be invaluable. Suppose that the squark can decay into all four neutralinos with comparable rate and that the heavier gauginos cascade to the lighter ones. If I knew what the mass of the next to lightest one was and its nature, both of which I should be able to get from its production in e+e-), then my LHC analysis would be much simplified. I believe that the biggest complementarity arises from the LHC's ability to produce squarks and gluinos and from the NLC's ability, assuming that its energy is big enough, to produce those states that are most difficult at LHC such as the Higgs's (all of them) and the heavier gauginos.

  4. Understanding the Susy Model

    • Can measurements distinguish between mSUGRA, GMSB, Anomaly mediated supersymmetry (etc.) ? For what regions (what fractions?) of parameter space? What luminosity is needed?

      As I have already indicated luminosity is not a big issue, how much you need is controlled by the mass scale. Just to remind you a 1 TeV gluino/quark is about 1 event every 20 mins at 1033 cm-2 sec-1. I will now stick my neck out and make a possibly inflamatory statement. If either squarks or gluinos are below 1 TeV, the models will be distinguished. The most difficult case is where the signals are qualitatively similar; for example GMSB with a long lived bino and SUGRA, where one must make mass measurements to distinguish them.

    • How well, and with what model assumptions, can unification at SGUT scale be demonstrated?

      Gaugino mass unification is the easiest. You will see the gluino and some gauginos. Their decays and production will constrain their identities. Model dependence is small here. Scalar mass unification requires measuring the different squark masses. Except for stop and sbottom, there is no way to measure the flavor. smoun/slectron can easily be separated. A model of the production would help but it is probably not essential. You might have to use a model to infer at least some of the squark masses. While the stau can probably be identified, measuring the mass in a model independent way is hard.

    • What is the set of measurements needed to fully explore R-parity violating SUSY and how well will the measurement of the Yukawa couplings LLL, LQQ, QQQ types be made?

      My talk discussed some R-parity stuff. If the R-parity violation is small an absolute measurement of the coupling is impossible (unless you can the lifetime of the state that decays via the coupling). The relative magnitudes can be measured by for example comparing the LSP rates reconstructed into light quark jets vs that into b-jets. Statistical samples are huge.