Final Report Summary - LEPTON PAIRS (Determination of proton parton densities using early LHC data and of contrains on new physics beyond the Standard Model)
Project context and objectives
The new proton-proton collider Large Hadron Collider (LHC) at CERN (European Organisation for Nuclear Research) produced ~5 fb^-1 of experimental data by the end of 2011. This large dataset allows for measurement of a large range of production processes with small statistical uncertainty. The main goal of LHC experiments is to discover the hypothetical Higgs particle in order to answer the question of the origin of mass of elementary particles within the electroweak theory and complete the Standard Model of particle physics (SM) as well as to find new phenomena, such as supersymmetry, quantum gravity, etc. However, since the centre-of-mass energy at LHC (7-14 TeV) is far below the Planck scale (10^16 TeV), the physicists have to look at small deviations of experimental distributions of jets, muons, electrons, etc., with high transverse momentum from the SM predictions to find signals of new particles or interactions. Therefore, the understanding of both the theoretical and the experimental uncertainties is the key for the searches for new particles and interactions at LHC.
In collisions of high-energy hadrons, one can distinguish two main phases. In the first (perturbative) phase, the high transverse momentum partons (quarks and gluons) are produced. The probability to find a parton of a given flavour in the initial hadron with a fraction x of its momentum at the scale Q^2 is defined by parton density function (PDF).
In the second (non-perturbative) stage, these partons leave the interaction region and, due to the large strong coupling, start to emit additional gluons and to produce quark-antiquark pairs, leading to a bunch of highly collimated particles in the final state. This bunch of particles, the hadron jet, approximately carries the energy of the initial parton and follows its direction. The kinematic distributions of the final state jets are predicted by the quantum chromodynamics (QCD) theory describing the strong interaction. The jets are absorbed in the calorimeters and their properties are reconstructed by measuring the energy deposited by the jet constituents in the calorimeter cells. Establishing the jet energy scale and the evaluation of its uncertainty is the key point of all physics measurements with jets.
As a Marie Curie fellow working in the CERN ATLAS calorimeter team, I worked on the measurement of the energy response of the ATLAS calorimeter to single isolated charged hadrons in proton-proton collisions. I have measured the calorimeter response as a function of the particle momentum and rapidity for particle momenta between 500 MeV and 20 GeV and within the acceptance of the tracking detector by comparing the calorimeter energy depositions to the particle momentum measured in the inner detector. I demonstrated that the ATLAS detector simulation model describes the observed response to better then ~5 % accuracy in both square-root s = 900 GeV and square-root s = 7 TeV proton-proton collisions for all rapidity and to better than ~2 % in the central detector region. Using the results of this precise in-situ measurement of the calorimeter response to single isolated hadrons, the ATLAS collaboration has established the jet energy scale with uncertainty at the level of ~2 %.
The measurement of jet production at LHC and the comparison against the QCD predictions provides a powerful test of the theory calculations at a new centre of mass energy as well as of the detector performance. The ATLAS collaboration performed two measurements of inclusive jet cross section for jets within the ±4.4 rapidity range and for transverse momenta between 20 GeV and 1.5 TeV. I made substantial contribution to these measurements of the jet transverse momentum spectrum. I made the calculation of final report next-to-leading order quantum chromodynamics (NLO QCD) predictions of the inclusive jet cross-section and evaluated the theoretical uncertainties in the SM predictions. I made the data to theory comparisons using different PDF sets and have evaluated the chi-squared test for this comparison. A reasonable agreement between the measured data points and NLO QCD theory predictions within the uncertainties has been found. It appeared that the HERAPDF (London, UK) 1.5 had provided the best description of the ATLAS jet cross section measurement with chi-squared/Ndf = 15.7/16. Together with a PhD student, I calculated the correction (and its uncertainty) for non-perturbative effects in jet production. I also contributed to the measurement of the properties of underlying event using neutral and charged particles in square-root s = 900 GeV and square-root s = 7 TeV proton-proton collisions.
I am one of the authors of the recently developed APPLGRID software package, which provides a coherent framework for the efficient extraction of PDFs and the strong coupling from the experimental data in NLO accuracy. The main idea of the method is to split the time-consuming evaluation of the NLO QCD cross-section into two parts. In the first part, the perturbative coefficients from a NLO QCD calculation are stored in look-up tables binned in parton momentum fractions and momentum transfer. In the second part, these coefficients are convoluted with any PDF in a fraction of second. This method is based on the Lagrange interpolation of PDFs and the use of symmetry in hard scattering matrix elements allowing to group weights into sub-process contributions. It offers a robust, transparent and uniform way to study the theoretical uncertainties for various QCD and electroweak processes via the interfaces to NLO cross section calculators, like NLOJET++ and MCFM (Monte Carlo for FeMtobarn processes). The package is used by the ATLAS collaboration for the interpretation of the jet cross-section measurement and by the HERAfitter project for the fast calculation of NLO QCD cross-sections in PDF fits.
The uncertainty in the theory calculations of scattering processes in proton collisions is due to the limited knowledge of PDFs, which is of the order of 5 %. There is also an additional dependence on the choice of PDF set (CTEQ, MSTW, NNPDF, etc.). I have studied this dependence of the theory predictions as well as the theory uncertainty due to the PDFs uncertainties, the uncertainty in the strong coupling constant measurement and the residual dependence of the observable upon the scale choice on the example of the inclusive jet and dijet mass cross-sections and using the APPLGRID technique.
The high precision measurements of LHC data will require the increasing knowledge of the proton's PDFs. Currently, the PDFs are best determined by simultaneous fits to datasets measured in various deep-inelastic scattering (DIS) and proton-(anti)proton collisions. The experimental data to be used in the PDF fits should be presented with both statistical and systematic correlated and uncorrelated uncertainties, such that the result of PDF fit provides meaningful uncertainties. The recent ATLAS collaboration measurement of the inclusive jet and dijet production cross-section provides the full and consistent treatment of the experimental uncertainties. The information on these uncertainties is based on my results of the single hadron response measurements.
The new proton-proton collider Large Hadron Collider (LHC) at CERN (European Organisation for Nuclear Research) produced ~5 fb^-1 of experimental data by the end of 2011. This large dataset allows for measurement of a large range of production processes with small statistical uncertainty. The main goal of LHC experiments is to discover the hypothetical Higgs particle in order to answer the question of the origin of mass of elementary particles within the electroweak theory and complete the Standard Model of particle physics (SM) as well as to find new phenomena, such as supersymmetry, quantum gravity, etc. However, since the centre-of-mass energy at LHC (7-14 TeV) is far below the Planck scale (10^16 TeV), the physicists have to look at small deviations of experimental distributions of jets, muons, electrons, etc., with high transverse momentum from the SM predictions to find signals of new particles or interactions. Therefore, the understanding of both the theoretical and the experimental uncertainties is the key for the searches for new particles and interactions at LHC.
In collisions of high-energy hadrons, one can distinguish two main phases. In the first (perturbative) phase, the high transverse momentum partons (quarks and gluons) are produced. The probability to find a parton of a given flavour in the initial hadron with a fraction x of its momentum at the scale Q^2 is defined by parton density function (PDF).
In the second (non-perturbative) stage, these partons leave the interaction region and, due to the large strong coupling, start to emit additional gluons and to produce quark-antiquark pairs, leading to a bunch of highly collimated particles in the final state. This bunch of particles, the hadron jet, approximately carries the energy of the initial parton and follows its direction. The kinematic distributions of the final state jets are predicted by the quantum chromodynamics (QCD) theory describing the strong interaction. The jets are absorbed in the calorimeters and their properties are reconstructed by measuring the energy deposited by the jet constituents in the calorimeter cells. Establishing the jet energy scale and the evaluation of its uncertainty is the key point of all physics measurements with jets.
As a Marie Curie fellow working in the CERN ATLAS calorimeter team, I worked on the measurement of the energy response of the ATLAS calorimeter to single isolated charged hadrons in proton-proton collisions. I have measured the calorimeter response as a function of the particle momentum and rapidity for particle momenta between 500 MeV and 20 GeV and within the acceptance of the tracking detector by comparing the calorimeter energy depositions to the particle momentum measured in the inner detector. I demonstrated that the ATLAS detector simulation model describes the observed response to better then ~5 % accuracy in both square-root s = 900 GeV and square-root s = 7 TeV proton-proton collisions for all rapidity and to better than ~2 % in the central detector region. Using the results of this precise in-situ measurement of the calorimeter response to single isolated hadrons, the ATLAS collaboration has established the jet energy scale with uncertainty at the level of ~2 %.
The measurement of jet production at LHC and the comparison against the QCD predictions provides a powerful test of the theory calculations at a new centre of mass energy as well as of the detector performance. The ATLAS collaboration performed two measurements of inclusive jet cross section for jets within the ±4.4 rapidity range and for transverse momenta between 20 GeV and 1.5 TeV. I made substantial contribution to these measurements of the jet transverse momentum spectrum. I made the calculation of final report next-to-leading order quantum chromodynamics (NLO QCD) predictions of the inclusive jet cross-section and evaluated the theoretical uncertainties in the SM predictions. I made the data to theory comparisons using different PDF sets and have evaluated the chi-squared test for this comparison. A reasonable agreement between the measured data points and NLO QCD theory predictions within the uncertainties has been found. It appeared that the HERAPDF (London, UK) 1.5 had provided the best description of the ATLAS jet cross section measurement with chi-squared/Ndf = 15.7/16. Together with a PhD student, I calculated the correction (and its uncertainty) for non-perturbative effects in jet production. I also contributed to the measurement of the properties of underlying event using neutral and charged particles in square-root s = 900 GeV and square-root s = 7 TeV proton-proton collisions.
I am one of the authors of the recently developed APPLGRID software package, which provides a coherent framework for the efficient extraction of PDFs and the strong coupling from the experimental data in NLO accuracy. The main idea of the method is to split the time-consuming evaluation of the NLO QCD cross-section into two parts. In the first part, the perturbative coefficients from a NLO QCD calculation are stored in look-up tables binned in parton momentum fractions and momentum transfer. In the second part, these coefficients are convoluted with any PDF in a fraction of second. This method is based on the Lagrange interpolation of PDFs and the use of symmetry in hard scattering matrix elements allowing to group weights into sub-process contributions. It offers a robust, transparent and uniform way to study the theoretical uncertainties for various QCD and electroweak processes via the interfaces to NLO cross section calculators, like NLOJET++ and MCFM (Monte Carlo for FeMtobarn processes). The package is used by the ATLAS collaboration for the interpretation of the jet cross-section measurement and by the HERAfitter project for the fast calculation of NLO QCD cross-sections in PDF fits.
The uncertainty in the theory calculations of scattering processes in proton collisions is due to the limited knowledge of PDFs, which is of the order of 5 %. There is also an additional dependence on the choice of PDF set (CTEQ, MSTW, NNPDF, etc.). I have studied this dependence of the theory predictions as well as the theory uncertainty due to the PDFs uncertainties, the uncertainty in the strong coupling constant measurement and the residual dependence of the observable upon the scale choice on the example of the inclusive jet and dijet mass cross-sections and using the APPLGRID technique.
The high precision measurements of LHC data will require the increasing knowledge of the proton's PDFs. Currently, the PDFs are best determined by simultaneous fits to datasets measured in various deep-inelastic scattering (DIS) and proton-(anti)proton collisions. The experimental data to be used in the PDF fits should be presented with both statistical and systematic correlated and uncorrelated uncertainties, such that the result of PDF fit provides meaningful uncertainties. The recent ATLAS collaboration measurement of the inclusive jet and dijet production cross-section provides the full and consistent treatment of the experimental uncertainties. The information on these uncertainties is based on my results of the single hadron response measurements.