Understanding the SM background to Supersymmetry

22 September 2009


Searching for Supersymmetry (SUSY) is one of the major aspects of the LHC physics program. SUSY is a theory of symmetry between matter and forces through a new discrete symmetry called ‘R-parity’ which depends on the lepton, baryon and spin quantum numbers of the final state particles of a physics process. R-parity conservation leads to a stable neutral particle called neutralino known as the Lightest SUSY Particle or the LSP.  The LSP escapes detection thus causing a sizeable amount of missing energy in the events. In SUSY signal processes, massive colored particles like squarks and gluinos are produced and eventually decays into lighter fundamental particles like leptons, and quarks and gluons, detectable as jets. Therefore searches for events with leptons, jets and missing energy are the most direct way to test for the existence of R-parity conserving SUSY.

For any new physics at the LHC, understanding the Standard Model (SM) processes is of utmost importance. Many of these SM processes will mimic the SUSY final state, so it is essential to understand and estimate the contribution from these processes to the SUSY search.  Final states with two leptons of same electric charge (same-sign leptons for short) have the advantage that contributions from SM processes are very small. Regardless of the size, we have to estimate the SM contribution to claim anything new. Therefore, the main work done for this analysis was to demonstrate the validity of a method to estimate the SM background rate for events with same-sign leptons, jets and missing energy.

The main background sources for this channel are top-quark pairs, W boson production with light or heavy flavor jets and single top production. We use two separate 2-dimensional distributions to evaluate the background based on four weakly correlated variables. The four variables are the pT of the second leading lepton, pT of the second leading jet, total missing energy, and the transverse mass of the leading lepton and missing energy in an event as shown on the figure below.


Four variables subdivided into four regions in the top anti-top production processes. This is the dominant background process to the SUSY search with two same-sign leptons. The background events are mostly accumulated in the regions denoted SB (side-band region in the left plot) and region B in the right plot. The signal on the other end is found mostly in the signal-like region denoted SR in the left plot and A in the right plot.



We use these four variables to define four regions which we label side-band (SB), signal-like region (SR) for the first distribution, and region A and region B in the second distribution. Region B and SB are the background enriched regions whereas region A and SR are the signal enriched regions. Here, we are interested in estimating the SM event rate in the region ASR (signal region: the events that are both in the region A and SR) by using regions ASB, BSB and BSR, those acts as the control regions.

The whole method works because these four variables have the same distribution in all four regions considered. Hence, if we can measure the ratio of background events in two of these regions, we can infer the amount of background events in the fourth region if we know how much there is the third region.

Using the above technique, we could reproduce the true SM event rate in the ASR region to within statistical errors. This was checked for two scenarios, first, in the absence of any SUSY events, that is, when only Standard Model processes were present in the four regions, and second, for the case where SM and SUSY events are present in real data. In the latter, the four phase-space regions are contaminated by other SUSY events and that could affect the method to estimate the true SM event rate in the signal region. It would also imply a bias on the extraction of the number of signal events in this region. Therefore, we have checked this thoroughly by considering contamination of our SUSY signal events from different minimal supergravity (mSUGRA) SUSY processes. We found that even with the contamination from these SUSY events in the control regions, we could still accurately estimate the true SM contribution.

The whole analysis was based on an integrated luminosity of 200 pb-1 and a center-of-mass energy of 10 TeV. 


Left: Effective mass distribution in the signal region. Discovery reach as function of mSUGRA parameters m0 and m1/2 (middle) and as a function of squark and gluino masses (right) in different integrated luminosity scenarios.



After the estimation of the SM background in the presence of different R-parity SUSY signal processes, we calculated the signal significance for different integrated luminosity scenarios. We found that we will be able to discover certain SUSY signals and we will be able to find evidence of gluinos and squark masses up to 500 GeV with a 3σ significance with as little as 200 pb-1 of integrated luminosity.

Though the analysis with two same-sign leptons may not be the day one analysis to discover Supersymmetry, we believe that owing to the nature of this analysis, it can be an effective signature to contribute to early SUSY searches in ATLAS.

 

 

 

Tapas Sarangi

University of Wisconsin-Madison