1. QCD Physics with ATLAS Mike Seymour University of Manchester/CERN PH-TH ATLAS seminar January 25 th / February 22 nd 2005

2. QCD Physics with ATLASMike SeymourATLAS seminar – 2 Tools of the Trade Fixed-order perturbation theory All-order perturbation theory ‘Best of both’ Non-perturbative fits and models Analysis tools

3. QCD Physics with ATLASMike SeymourATLAS seminar – 3 Tools of the Trade Fixed-order perturbation theory –LO – state of the art for high parton multiplicity – highly automated –NLO – up to five partons – partly automated –NNLO – two partons (  (e + e – ), DIS, Drell–Yan) –Valid for infrared safe observables All-order perturbation theory ‘Best of both’ Non-perturbative fits and models Analysis tools

4. QCD Physics with ATLASMike SeymourATLAS seminar – 4 Tools of the Trade Fixed-order perturbation theory All-order perturbation theory –Identification of large logarithmic terms at every order –Evolution (Altarelli–Parisi, BFKL) –Exponentiation and resummation (Sudakov) –Only possible for well-defined observables –Numerical resummation (parton showers) ‘Best of both’ Non-perturbative fits and models Analysis tools

5. QCD Physics with ATLASMike SeymourATLAS seminar – 5 Tools of the Trade Fixed-order perturbation theory All-order perturbation theory ‘Best of both’ Non-perturbative fits and models –Parton distribution functions –Fragmentation functions –Hadronization models –Underlying event models Analysis tools –Jet algorithms –Event shapes –‘Event definitions’ ‘gaps between jets’ cross sections Dijet azimuthal correlation (DØ)

6. QCD Physics with ATLASMike SeymourATLAS seminar – 6 What do we mean by the Underlying Event? “Everything except the hard process” but… initial state radiation factorization scale parton distribution functions parton evolution  underlying event model integral part of event model

7. QCD Physics with ATLASMike SeymourATLAS seminar – 7 Why should we be interested? 1.QCD Connection with: diffraction saturation confinement total cross section Can we predict/understand the properties of hadrons? 2.Experiments Occupancy Pile-up Backgrounds

8. QCD Physics with ATLASMike SeymourATLAS seminar – 8 Why should we be interested? 2.Experiments Occupancy Pile-up Backgrounds 3.Physics Jet cross sections Mass reconstruction Rapidity gaps/jet vetoes E miss reconstruction Photon/lepton isolation  “Don’t worry, we will measure and subtract it” But… fluctuations and correlations crucial

9. QCD Physics with ATLASMike SeymourATLAS seminar – 9 Fluctuations and correlations log  ptpt Steep distribution ) small sideways shift = large vertical Rare fluctuations can have a huge influence ) corrections depend on physics process

10. QCD Physics with ATLASMike SeymourATLAS seminar – 10 How do we Measure Underlying Event? http://www.phys.ufl.edu/~rfield/cdf/chgjet/chgjet_intro.html “Transverse” region of two-jet events

11. QCD Physics with ATLASMike SeymourATLAS seminar – 11 HERWIG’s Soft Underlying Event model Compare underlying event with ‘minimum bias’ collision Parameterization of (UA5) data + model of energy-dependence G.Marchesini & B.R.Webber, PRD38(1988)3419

12. QCD Physics with ATLASMike SeymourATLAS seminar – 12 Multiparton Interactions Assume p–p collision ¼ local instantaneous sampling of two disks of partons  Can have several perturbative scatters within one event PYTHIA and Jimmy (add-on for HERWIG) Low p t scatters ! underlying event

13. QCD Physics with ATLASMike SeymourATLAS seminar – 13 Proton Radius parameter within Jimmy Increasing  2 to 2 GeV 2 (i.e. decreasing proton radius by 40%) with ptmin=3 GeV gives ~perfect description of Tevatron data… I.Borozan, PhD thesis, unpublished Transverse

14. QCD Physics with ATLASMike SeymourATLAS seminar – 14 Tools of the Trade Fixed-order perturbation theory All-order perturbation theory ‘Best of both’ Non-perturbative fits and models –Parton distribution functions –Fragmentation functions –Hadronization models –Underlying event models Analysis tools –Jet algorithms –Event shapes –‘Event definitions’ ‘gaps between jets’ cross sections Dijet azimuthal correlation (DØ)

15. QCD Physics with ATLASMike SeymourATLAS seminar – 15 Leading order: a parton is a jet But… Need an algorithmic jet definition Jet Algorithms one jet or two?

16. QCD Physics with ATLASMike SeymourATLAS seminar – 16 Jet Algorithms Cone algorithmCluster algorithm Depends on energy flow relative to cone radius R Depends on separation of subjets relative to cutoff R

17. QCD Physics with ATLASMike SeymourATLAS seminar – 17 Jet Algorithms Cone vs cluster algorithm MHS, Z.Phys.C62(1994)127 Top mass reconstruction as a test case Update in progress (Chris Tevlin, ATLAS student) Cluster Cone

18. QCD Physics with ATLASMike SeymourATLAS seminar – 18 Joey Huston: SM Physics at ATLAS

19. QCD Physics with ATLASMike SeymourATLAS seminar – 19 Joey Huston: SM Physics at ATLAS

20. QCD Physics with ATLASMike SeymourATLAS seminar – 20 Dijet Azimuthal Correlations Recent DØ analysis. Motivation: find an observable –to tune event generators on, that separates out hard, soft and non-perturbative physics –that is as insensitive as possible to poorly-understood jet energy response callibration ! DØ slides… –Markus Wobisch’s talk at http://conferences.fnal.gov/tev4lhc/talks/040916_TeV4LHC_QCD/ http://conferences.fnal.gov/tev4lhc/talks/040916_TeV4LHC_QCD/ Full machinery of QCD resummation should be applicable: calculation in preparation…

21. QCD Physics with ATLASMike SeymourATLAS seminar – 21 Summary Every measurement in a hadron collider involves doing QCD phenomenology! Most involve a long chain of corrections –higher orders –parton distribution functions –jet algorithm –hadronization –underlying event to connect measurement to physics Always ask how well these corrections are known: –how much am I relying on QCD phenomenology? –how well understood is it?