Event Shape Update. S. Hanlon I. Skillicorn. A. Everett A. Savin. Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

Transcription

1 Event Shape Update A. Everett A. Savin S. Hanlon I. Skillicorn Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

2 Outline Event Shape Motivation and the Power Correction Method Out-of-Plane Momentum Event Selection Kinematic Checks Comparison of two analyses NLO Calculations Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

3 Approach to Non-Perturbative Calculations pqcd prediction measured distribution Correction factors for non-perturbative (soft) QCD effects Theory reduces corrections for any infrared safe event shape variable, F: Used to determine the hadronization corrections F = F + perturbati ve F power correction F pow = a F 16 µ I P Q β0 Q K ln α ( µ ) ( Q) (ln + + 1) 2 I αs αs ( 3π Q µ I 2π µ I β0 0 Q ) Power Correction α 0 independent of any fragmentation assumptions = non-perturbative parameter (Dokshitzer, Webber, Phys. Lett. B 352(1995)451) Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

4 Extraction of Parameters 0 α B T 2 M 1-T T ZEUS 1 s.d. errors 95% confidence region (stat. + exp. sys. errors) ZEUS (prel.) α ZEUS T γ 1 s.d. errors 95% confidence region (stat. + exp. sys. errors) ZEUS (prel.) C T T 2 M B γ 0.4 B γ Means T γ 0.35 C Differential Distributions α s (M Z ) α s (M Z ) Test of Power Correction for Means and Differential Distributions already made preliminary Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

5 Energy Flow and Dijets Instead of inclusive events, we use dijets in the current region of the Breit frame. Dijets: pqcd part of <F> calculation well understood Event topology well understood New Event Shape Variables: K out, Azimuthal Correlation Must define an event plane in the Breit frame Use Thrust to define the event plane Transverse Energy Flow * γ axis T k = max n ˆ k i r pi nˆ r p i i k T γ axis T T axis T M axis T m axis Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

6 Out-of-plane Momentum Energy flow out of event plane defined by proton direction and thrust major axis Sensitive to perturbative & non-perturbative contributions Dijet event: Perturbative physics takes place in the plane Non-perturbative physics give rise to out-of-plane momentum ' K = out p h h out Event Plane Photon Thrust Major Axis Quark Outgoing Particle Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

7 Event Selection Selection cuts 100 GeV2 Q 2 DA y JB > 0.04 y el < 0.95 Vertex with z < 40 cm 38 < E-p Z < 65 GeV Good positron Sinistra Probability > 0.9 E e,da > 10 GeV ZUFO selection η lab < 2.2 η Breit < 3.0 good acceptance region P T > 0.5 GeV Breit frame jet cuts At least 2 jets in Breit frame: E 1,T > 6 GeV E 2,T > 5 GeV y 2 > 0.1 Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

8 Kinematic Reconstruction E e,da Events 2 QDA Events Ariadne Data GeV Ariadne Data GeV x DA Events θ SiCal Events x DA Ariadne Data Ariadne Data radians Choose Ariadne to calculate detector corrections Ariadne is area normalized to the data within the cut boundaries Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

9 Hadronization Effect Shift in peaks illustrates the hadronization effect Use power correction to calculate this shift. Events PAR K OUT /Q Hadron Level Ariadne Parton Level Ariadne /Q K OUT Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

10 zufo K out /Q y 2 K/Q Comparison of Ariadne and Corrected Data > 0.1 Corrected Data Hadron Level Ariadne Measure K/Q because powers of ln(k/q) are used in the resummation. Use ZUFO to calculate the momentum out of the event plane. Ariadne describes the data well K OUT /Q Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

11 ??? Should I Combine the Previous 2 Slides to 1 Using This Plot??? K OUT /Q Events Ariadne Corrected Data 10 Ariadne Hadron Level Parton Level Ariadne K/Q Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

12 K/Q Analyses Comparison zufo K out /Q y 2 > 0.1 zufo K out /Q y 2 > 0.1 Events Glasgow Raw Data Wisconsin Raw Data K OUT /Q Events Glasgow Ariadne Corrected Data Wisconsin Ariadne Corrected Data K OUT /Q Some of the difference due to sensitivity in the Thrust calculation for determining the event plane. A.E. and I.S. in good agreement. Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

13 K/Q NLO Prediction *Res = Resummation * K/Q Resummation + LO + NP line should be a good fit to our data. LO + Resummation necessary for power correction calculation. However, the theorists won t give us the program for calculating the resummation. Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

14 NLO Calculation We continue to work while waiting for the theorists to give us the requested curves. Our NLO calculation matches theorist NLO within 1%. Just waiting for resummation... DISENT Calculation Theorist Calculation Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

15 Additional NLO DISENT calculates to lowest order in K/Q Also use NLOJET++ to calculate perturbative K/Q NLOJET can calculate to the next hgher order in K/Q Lowest Order K/Q NLOJET DISENT K/Q Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,

16 Conclusion Plan Summary Dijet event shapes sensitive to non-perturbative physics Good agreement between first and second analyses, and previous and current measurements. Currently working on systematics. Plan to have measurement ready for ICHEP Event Shapes, A. Everett, U. Wisconsin ZEUS Meeting, June 24,