Elliptic flow, non-flow and initial-state fluctuations at RHIC

Transcription

1 1 Elliptic flow, non-flow and initial-state fluctuations at RHIC Constantin Loizides (LBNL) INT-10-A: Opening workshop May 4-8, 010 New results from LHC and RHIC Seattle, 8 May 010

2 The PHOBOS collaboration Burak Alver, Birger Back, Mark Baker, Maarten Ballintijn, Donald Barton, Russell Betts, Richard Bindel, Wit Busza (Spokesperson), Vasundhara Chetluru, Edmundo García, Tomasz Gburek, Joshua Hamblen, Conor Henderson, David Hofman, Richard Hollis, Roman Hołyński, Burt Holzman, Aneta Iordanova, Chia Ming Kuo, Wei Li, Willis Lin, Constantin Loizides, Steven Manly, Alice Mignerey, Gerrit van Nieuwenhuizen, Rachid Nouicer, Andrzej Olszewski, Robert Pak, Corey Reed, Christof Roland, Gunther Roland, Joe Sagerer, Peter Steinberg, George Stephans, Andrei Sukhanov, Marguerite Belt Tonjes, Adam Trzupek, Sergei Vaurynovich, Robin Verdier, Gábor Veres, Peter Walters, Edward Wenger, Frank Wolfs, Barbara Wosiek, Krzysztof Woźniak, Bolek Wysłouch ARGONNE NATIONAL LABORATORY INSTITUTE OF NUCLEAR PHYSICS PAN, KRAKOW NATIONAL CENTRAL UNIVERSITY, TAIWAN UNIVERSITY OF MARYLAND BROOKHAVEN NATIONAL LABORATORY MASSACHUSETTS INSTITUTE OF TECHNOLOGY UNIVERSITY OF ILLINOIS AT CHICAGO UNIVERSITY OF ROCHESTER

3 Outline 3 1) System size, energy, pseudorapidity, and centrality dependence of elliptic flow, PHOBOS, PRL 98, 430, 007 (nucl-ex/ ) ) Cluster properties from two-particle angular correlations in p + p collisions at 00 and 410-GeV, PHOBOS, PRC 75, , 007 (arxiv: [nucl-ex]) 3) Importance of correlations and fluctuations on the initial source eccentricity in A+A collisions, PHOBOS + U.Heinz, PRC 77, , 008 (arxiv: [nucl-ex]) 4) System size dependence of cluster properties from two-particle angular correlations in Cu+Cu and Au+Au collisions at 00 GeV, PRC 81, 04904, 010 (arxiv: [nucl-ex]) 5) High transverse momentum triggered correlations over a large pseudorapidity acceptance in Au+Au collisions at 00 GeV, PRL 104, 06301, 010 (arxiv: [nucl-ex]) 6) Event-by-event fluctuations of azimuthal particle anisotropy in Au + Au Collisions at 00 GeV, PRL, 104:14301, 010 (nucl-ex/070036) 7) Non-flow correlations and elliptic flow fluctuations in gold-gold collisions at 00 GeV, PRC81, , 010 (arxiv: [nucl-ex]) 8) Collision geometry fluctuations and triangular flow in heavy-ion collisions, B. Alver, G. Roland, accepted in PRC, 010 (arxiv: [nucl-th])

4 Outline 4 1) System size, energy, pseudorapidity, and centrality dependence of elliptic flow, PHOBOS, PRL 98, 430, 007 (nucl-ex/ ) ) Cluster properties from two-particle angular correlations in p + p collisions at 00 and 410-GeV, PHOBOS, PRC 75, , 007 (arxiv: [nucl-ex]) 3) Importance of correlations and fluctuations on the initial source eccentricity in A+A collisions, PHOBOS + U.Heinz, PRC 77, , 008 (arxiv: [nucl-ex]) 4) System size dependence of cluster properties from two-particle angular correlations in Cu+Cu and Au+Au collisions at 00 GeV, PRC 81, 04904, 010 (arxiv: [nucl-ex]) 5) High transverse momentum triggered correlations over a large pseudorapidity acceptance in Au+Au collisions at 00 GeV, PRL 104, 06301, 010 (arxiv: [nucl-ex]) 6) Event-by-event fluctuations of azimuthal particle anisotropy in Au + Au Collisions at 00 GeV, PRL, 104:14301, 010 (nucl-ex/070036) 7) Non-flow correlations and elliptic flow fluctuations in gold-gold collisions at 00 GeV, PRC81, , 010 (arxiv: [nucl-ex]) Burak Alver 8) Collision geometry fluctuations and triangular flow in heavy-ion collisions, B. Alver, G. Roland, accepted in PRC, 010 (arxiv: [nucl-th]) next week

5 Correlation studies with PHOBOS Large coverage: η < Spectrometer Octagon Ringcounter counters Ring Hits (no pt information) Vertex detector Tracks (near mid-rapidity) (Dep. on vertex and trk quality: 0<η<)

6 Inclusive two particle correlations 6 Au+Au@00GeV 0-10% % Fn, R, = n 1 1 Bn, Similar structure in p+p, Cu+Cu and Au+Au collisions, a clear signal of elliptic flow is visible in A+A collisions Remember: No pt cutoff (i.e. pt>7 35 MeV/c) PHOBOS, PRC (007) PHOBOS, PRC (010)

7 Cluster model: G R = k eff 1 1 B Cluster model fit to correlation in Δη Effective cluster size 7 G exp 4 Cluster width Possible source for correlations: Production of intermediate objects which decay into particles p+p@00gev Project onto Δη axis and fit with a simple parametrization 6-6 PHOBOS p+p 00 GeV scale error NB: In A+A this removes elliptic flow PHOBOS, PRC (007) PHOBOS, PRC (010)

8 Cluster model fit results 8 Au+Au Multiplicity of the clusters is large (up to 6 charged particles - more than for known resonances) Cluster width exceeds that for isotropic decay at rest (~1) AMPT p+p Cu+Cu PHOBOS NB: Extrapolated to full phase space Cluster size Cluster width PHOBOS, PRC (007) PHOBOS, PRC (010)

9 Cluster model fit results 9 Multiplicity of the clusters is large (up to 6 charged particles - more than for known resonances) Cluster size Cluster width exceeds that for isotropic decay at rest (~1) p+p Cluster sizes and width very similar at the same centrality, defined as the same fraction of cross section. The shape of the interaction area seems to even determine the properties at hadronization? NB: Extrapolated to full phase space AMPT Cu+Cu Au+Au Cluster width PHOBOS PHOBOS, PRC (007) PHOBOS, PRC (010)

10 Elliptic flow measurement in PHOBOS 10 Reaction-plane / Sub-event technique Correlate reaction plane determined from azimuthal pattern of hits in one part of the detector with information from other parts a of the detector Subevent B 0.1 < η < 5.4 tan A = sin A cos A Subevent A -0.1 < η < -5.4 Separation of correlated subevents typically large in η v obs = cos A B obs v = v events cos A B events Poskanzer, Voloshin, nucl-ex/ Resolution correction

11 Elliptic flow and collision geometry η < 1 Cu+Cu Statistical errors Au+Au Geometry should cancel out in the v /ε ratio PHOBOS, Au+Au, 00,130, GeV: PRL (005) PHOBOS, Cu+Cu, GeV: PRL (007) PHOBOS, Cu+Cu,.4 GeV: prel. QM06 11 Ncoll/0.5Npart Overlap area S Eccentricity

12 Elliptic flow and collision geometry 1 Ncoll/0.5Npart Cu+Cu Overlap area S Au+Au Statistical errors No scaling between Cu+Cu and Au+Au using the standard eccentricity definition PHOBOS, Au+Au, 00,130, GeV: PRL (005) PHOBOS,Cu+Cu, GeV: PRL (007) PHOBOS, Cu+Cu,.4 GeV: prel. QM06 STAR+NA49+E877, PRC (00) (data taken with no adjustments) Eccentricity

13 Study non-flow contribution via eta gaps 13.0 < η < 5.4 Track-based method 0<η<1.5 Statistical errors only Spectrometer Cu+Cu, 00 GeV PHOBOS Octagon and Rings.0 < η < 3. Preliminary Octagon Only 3.0 < η < 5.4 Rings Only The large η separation between the subevents and the signal region suppresses the non-flow contribution in track-based (hit-based) method.

14 Participant eccentricity Participant Eccentricity y' y 14 PHOBOS Glauber MC b x' Ψ0 x b Ψ0 Participants Au+Au Cu+Cu b The spatial distribution of the interaction points of participating nucleons for the same b varies from event-to-event. Thus, define part = y x 4 y x xy Co-variance term not in standard definition 0 part 1 Introduced at QM05, PHOBOS, PRL (007)

15 Elliptic flow and collision geometry 15 Ncoll/0.5Npart Hydro limit Au+Au Cu+Cu Statistical errors Scaling between Cu+Cu and Au+Au using participant eccentricity definition PHOBOS, Au+Au, 00,130, GeV: PRL (005) PHOBOS,Cu+Cu, GeV: PRL (007) PHOBOS, Cu+Cu,.4 GeV: prel. QM06 STAR+NA49+E877, PRC (00) (data taken with no adjustments) Overlap area S Participant eccentricity

16 Eccentricity scaling is global Statistical errors only 16 Statistical errors only Unity of geometry, system, energy, transverse momentum and pseudorapidity for the same Npart (~area density) PHOBOS, JPG (007)

17 Robustness of eccentricity definition Skin depth Nuclear radius 17 σ NN inel Inter-nucleon seperation Baseline parameters: Nucleon-nucleon cross section: σnn=4mb Skin depth: a=0.535fm Wood-saxon radius: RA=6.38fm Inter-nucleon separation distance: d=0.4fm Robust definition wrt variation of Glauber parameters and to varying assumptions about matter production (the latter not shown here, see extra slides) PHOBOS+Heinz, PRC (008)

18 Expected relative flow fluctuations part part Uncertainty from variations of Glauber MC parameters Baseline parameters: Nucleon-nucleon cross section: σnn=4mb Skin depth: a=0.535fm Wood-saxon radius: RA=6.38fm Inter-nucleon separation distance: d=0.4fm 18 Participant eccentricity model Analytic (b=0fm) Baseline 90% C.L Broniowski et al., PRC (007) 00 GeV Au+Au PHOBOS Glauber MC Number of participants If initial state fluctuations are present, expect large relative flow fluctuations: v v ~ part part PHOBOS, nucl-ex/060805

19 Measuring elliptic flow fluctuations Observed v distribution 19 Parametrized v distribution g(vobs) f(v) σ v Max-Likelihood fit to determine: vobs Kernel Detector and acceptance effects Finite-number fluctuations Multiplicity fluctuations v <v> obs g v <v> and σ v 1 = 0 K v obs, v f v dv Kernel K(vobs,v) v vobs Detector response PHOBOS, PRL (010)

20 Relative fluctuations Measured relative total fluctuations PHOBOS 0 η <1 Au+Au, 00 GeV CGC-MC (fkln) Drescher, Nara, PRC 76 (007) Data (flow + non-flow) Participant eccentricity (Glauber) Number of participants Shown at QM06 as flow fluctuations, however non-flow contribution (included in sys.error) was underestimated. Now interprete as total v fluctuations. PHOBOS, PRL (010)

21 Measure non-flow contribution 1 PHOBOS uses data driven analysis to measure the contribution of non-flow Flow is a function of η and correlates particles at all Δη Non-flow (δ) is dominated by short range correlations (small Δη) Study correlations at different Δη v 1, cos 1, =v 1 v 1, v 1, Assume non-flow to be zero for Δη> fit fit Fit v 1, =v 1 v, 1 Subtract fit results at all (η1,η) Integrate over particle pairs to obtain / v Numerically relate / v and v / v to obtain flow / v vfit 1 vfit 1, Non-flow PHOBOS, PRC (010)

22 Variation of the fit region 40-45% 35-40% 30-35% 0-5% 15-0% 10-15% 5-30% v v 6-10% Non-flow ratio as a function of Δη cut used to define the fit region. Red-point is baseline for analysis, while black points are used for systematic error Saturation is very encouraging, however does not rule out contributions with very little Δη dependence. PHOBOS, PRC (010)

23 Variation of non-flow strength in fit region 3 Non-flow in p+p, 00 GeV / v for assumptions in Au+Au, 00 GeV All non-flow m=10 Measure non-flow in p+p data, and compare to MC generators m=3 m=0(1) Assume non-flow in fit region to be m times non-flow in p+p (rather than 0) fit fit 1 v 1, m HIJING =v v MC 1 PHOBOS, PRC (010)

24 For m= % of v / v Relative fluctuations Measured relative flow fluctuations m=0(1) m=3 4 CGC-MC (fkln) Drescher, Nara, PRC 76 (007) m=10 Initial state fluctuations if indeed present seem not to be significantly enhanced in later stages of the collision. Short-range (Δη<) non-flow contribution are removed PHOBOS, PRC (010)

25 Which moment of v is measured? By now α is known: root-meansquared = 4i1 / i 0 i 1 Define 1/ v v 5 Ollitrault et. al., PRC (009) mean Event Plane Resolution, R PHOBOS R: For PHOBOS standard event-plane method v {EP }= v (For the observed fluctuations this implies up to about 10% difference) PHOBOS+Heinz, PRC (008)

26 Correction for non-flow and fluctuations 6 Published STAR results Derive analytic correction for non-flow and fluctuations in leading order of and v v { } = v v v {4 } = v v v {subep} = v 1 f R v 1 f R 0 f R 0. Differences between methods proportional to % Most Central tot= v Need additional assumption or information to separate between non-flow and fluctuations Ollitrault, Poskanzer, Voloshin PRC (009)

27 Correction for non-flow and fluctuations 7 Corrected mean results Model assuming: v = v part part Results for Glauber eccentricity = N part pp with pp = Glauber eccentricity tot = v % Most Central Corrected mean values agree in participant frame. Reduces errors on v measurements by about 0%. Eccentricity values are calculated for standard Glauber and a mix of 30:70 CGC (not shown) Ollitrault, Poskanzer, Voloshin PRC (009)

28 Relative fluctuations Measured relative flow fluctuations 8 Model corrections: Glauber CGC (30:70) m=0(1) m=3 Ollitrault et. al., PRC (009) m=10 Results based on analytic model tuned to STAR data applied to total measured fluctuations are consistent with PHOBOS data.

29 Correlations wrt trigger particle p+p (PYTHIA) pttrig >.5 GeV/c, 0<ηtrig<1.5 PTassoc > 7(η=3), >35(η=0) MeV/c = { Normalization NB: PYTHIA agrees with STAR at mid-rapidity for a similar set of pt cuts 9 Au+Au 0-30% (PHOBOS) - a(δη)[ ]} Scale factor Elliptic flow Raw correlation PHOBOS, PRL (010)

30 Ridge extent in Δη (near-side) 30 Near-side correlation for Δφ <1 Long range ridge yield In 0-30% most central Au+Au, there is a relatively flat correlated yield of about 0.5 particles per trigger particle (per unit η) PHOBOS, PRL (010)

31 Projected correlation along Δφ 0-10% % Au+Au PYTHIA Short-range Δη <1 PHOBOS Long-range -4 < Δη < - PHOBOS, PRL (010)

32 Integrated ridge yield (away side) 0-10% Short-range Δη <1 Au+Au PYTHIA 3 NEAR side short-range minus PYTHIA long-range Away side times 0.5 short, long minus PYTHIA short, long range PYTHIA PHOBOS arxiv: Long-range -4 < Δη <- PHOBOS PHOBOS # of participant nucleons (Npart)

33 Summary Strong short-range correlations are observed in A+A collisions. Large sizes of clusters cannot be attributed to low mass resonances. 33 Cluster size and width scales with the fractional cross section Understanding of flow, non-flow correlations, flow and eccentricity fluctuations has converged Short-range non-flow correlations contribute about ~10% (absolute) to v fluctuations measurement See next slides Remaining caveat is role of long-range non-flow contribution Initial state eccentricity fluctuations (if present) are consistent with the data, but do not leave much room for increase of fluctuations in later stages of the collision evolution Near-side correlations of associated particles with high pt>.5 GeV/c particle extend 4 units in pseudo-rapidity and diminish for a system with Npart~80.

34 Correlations at large η STAR inclusive 1.< η< PHOBOS inclusive PHOBOS pttrig>gev < η<4 < η<4 Long range correlations are well described by 3 Fourier components Alver, Roland, arxiv: [nucl-th]

35 Flow vs non-flow Standard picture: Flow = global ( collective ) Why is Second Fourier special? It is a large effect Second Fourier coefficient It is present at large Δη Non-flow = local ( clusters ) It is a function of η All Fourier coefficients 35 v/ε, v(pt), v(rp), v{4}, fluctuations, etc., make sense

36 Closer look at non-flow Remove first and second Fourier contribution and suppress short-range peak ( Δη <1) 36 Is Third Fourier special? It is a large effect It is there at large Δη Can it be linked to initial state? Is it a function of η (?) Measure centrality + pt dependence, 3 particle correlations, non-flow, etc. Ridge and broad away side: Even without trigger particle and at large Δη

37 Participant triangularity r cos r sin part = r Generalize from participant eccentricity to participant triangularity Alver, Roland, arxiv: [nucl-th] 37 r cos 3 r sin 3 3 = r Transform into center of participants and use polar coordinates y x r φ

38 Triangular flow ψ 38 ψ3 dn ~1 v n cos n n R d v = cos R dn ~1 v n cos n n n d v = cos Alver, Roland, arxiv: [nucl-th] v 3 =0 v 3 = cos (alternatively keep the sin terms)

39 Triangular flow in AMPT 39 v = cos ψ v = cos ψ3 Participant triangularity leads to triangular flow in AMPT Alver, Roland, arxiv: [nucl-th]

40 Triangular flow in data PHOBOS 40 STAR The ratio of triangular to elliptic flow qualitatively agrees in data and AMPT STAR arxiv: PHOBOS PRC 81, (010) PHOBOS PRL 104, 0630 (010) Alver, Roland, arxiv: [nucl-th]

41 Summary/Outlook Triangular flow (caused by initial state fluctuations) is a natural explanation for the ridge effects. 41 Need to (experimentally) characterize its properties Centrality and η (and pt) dependence Understand non-flow contribution If collective property, then it must be treated as background to all our two (and three) particle correlation measurements, ie taken out as we do for second Fourier component What about higher moments? Yes, all moments are there in the initial state, but the question is: Are (or how often are) they dominating the initial pressure gradients Theory side is already working on establishing the connection between hydro and triangularity

42 Extra 4

43 Triangular flow in hydro 43 Triangular flow may be a new handle on the initial geometry and the hydrodynamic expansion of the medium Luzum, Ollitrault (work in progress)

44 Generalized eccentricity 44 Some work in progress figures

45 Example for Pentagrularity 45 Ecc5 > 0.6 (but note also ecc~0.5 and ecc3~0.4) Some work in progress figures

46 Eccentricity and triangularity 46 Some work in progress figures

47 47 Possible source for correlations: Production of intermediate objects which decay into particles Cluster model Cluster model

48 Centrality determination Makeup of nuclei Made up of nucleons drawn from Wood-Saxon distribution Separate by b (with dn/db~b) Collision of nuclei 48 Assume: Nucleons travel along z on straight-line paths and interact when their centers are within NN inel / #Participants is number of nucleons that interact at least once (Npart~A) #NN-collisions is total number of collisions (Ncoll~A4/3) Nucleus 1 Relate to data via Glauber MC based detector simulations Nucleus x Participants PHOBOS, NPA (005) y b Impact parameter

49 Assumptions of particle production 49 Model two component scenario Matter production via participants and binary collisions dnaa dnpp 1 x = Npart x Ncoll d d Mixture with x=0.13 describes mid-rapidity dn/dη quite well 10% increase in eccentricity for central Au+Au Include thermalization time by smearing the matter around the original production point Hard-sphere and Gaussian For chosen set of parameters only a very small effect NB: More generalized studies also done, see Broniowski et al., PRC 76 (007) PHOBOS+Heinz, PRC 77 (008)

50 Flow methods Two-particle cumulant Measures: v {} = v v v 1/ M v {}= cos 1 Four-particle cumulant Measures: 1/ 4 v {4 }= cos 1 cos v {4 } = v v v 1/M 3/ 4 cos A v {subep}= R R= cos A B Measures: v {subep} = v 1 f R v NB: For simplicity, n (as index and in cos terms) dropped 1 f R Ollitrault, Poskanzer, Voloshin PRC (009)

51 Challenges of event-by-event v obs 51 PHOBOS Multiplicity Array -5.4<η<5.4 coverage Holes and granularity differences Usage of all available information in event to determine event-by-event a single value for vobs Azimuthal angle Hit Distribution ~11 units in η dn/dη HIJING + Geant 15-0% central Primary particles Hits on detector Pseudo-rapidity Pseudo-rapidity

52 obs Event-by-event measurement of v Event-by-event measurement of vobs Deal with acceptance effects 5 Probability distribution function Use all available hit information Probability distribution function for hit positions: φ η P, ; v obs, 0 =p [1 v cos 0 ] Normalization incl. acceptance Probability of hit in (φ,η) Maximize the likelihood function to obtain vobs and φ0 (event plane angle) obs L v n, 0 = i=1 P i, i ; v obs, 0 PHOBOS, PRL (010)

53 obs Event-by-event measurement of v vobs Au+Au, 00 GeV vobs Au+Au, 00 GeV Triangular v(η) Trapezoidal v(η) PHOBOS, PRC 7, (005) PHOBOS, PRC 7, (005) v(η) =Pseudo-rapidity triangular 53 Pseudo-rapidity P, ; v obs, 0 =p [1 v cos 0 ] Use known, measured shape Analysis is run on triangular and trapezoidal shape. Results are averaged at the end. PHOBOS, PRL (010)

54 Determining the kernel Measure and record the vobs distribution in bins of v and multiplicity (n) from large MC samples 54 K(vobs,v, fixed n) HIJING events Modified φ to include triangular or trapezoidal flow PHOBOS MC v Fit response function (ideal case) K v obs, v,n = v obs e v obs v I 0 vobs obs v v (J.-Y.Ollitrault, PRD (199) 46, 6) Changed to account for detector effects v An B v (suppression) C = D n (finite resolution) PHOBOS, PRL (010)

55 Extracting dynamical fluctuations obs g v 55 1 = 0 K v obs, v f v dv Measured Constructed from MC Gaussian Ansatz: f v =exp Comparison with data g(vobs) 15-0%,Au+Au, 00 GeV g(vobs) g1 g [ v v v ] Different trials for Ansatz f(v) Use kernel + integrate f1 f Fit prob.: 0.94 (0.006) vobs v Compare expected g(vobs) for trials with data: Maximum-Likelihood fit <v> and σ v PHOBOS, PRL (010)

56 Elliptic flow fluctuations: <v> and σv Dynamical flow fluctuations Mean elliptic flow η <1 η <1 PHOBOS preliminary <v> PHOBOS preliminary σv (90% C.L.) v v Au+Au 00 GeV Number of participants 56 (90% C.L.) Au+Au 00 GeV Number of participants Systematic errors: Variation in η-shape Variation of f(v ) MC response Vertex binning Ф binning 0 Scaling errors cancel in the ratio: relative fluctuations, σv/<v> PHOBOS, PRL (010)

57 Event-by-event v vs published results Standard methods 57 <v> Averaged over events to measure the mean Hit- and track-based Use reaction plane subevent technique η <1 PHOBOS Au+Au, 00 GeV PRC 7, (005) Number of participants Very good agreement of the event-by-event measured mean v with the hit- and tracked-based, event averaged, published results PHOBOS, PRL (010)

58 Numerical subtraction 58 Lookup table Keep results as lookup table Results slightly depend on σn Use σn = 0.4, 0.6 and 0.8 PHOBOS, PRC (010)

59 Construction of correlated yield = { 59 - a(δη)[ ]} Raw correlation: ratio of per-trigger same event pairs to mixed event pairs Elliptic flow: V(Δη) = <vtrig><vassoc> PHOBOS Phys. Rev. C 7, (R) (005) a(δη) Scale factor: accounts for small multiplicity difference between signal and mixed events B(Δη) Normalization term: relates flow-subtracted correlation to correlated yield PHOBOS, PRL (010)

60 Subtraction of elliptic flow s(δφ,δη) b(δφ,δη) PHOBOS Long Range -4 < Δη < - 60 Short Range -1 < Δη < Elliptic Flow a(δη) [ 0 Δφ 100 [deg] Δφ 100 ] PHOBOS, PRL (010)

61 ZYAM implementation ZYAM factors from d-fit in Δη and Npart a(δη) PHOBOS preliminary Δη % 40-45% 35-40% 30-35% 5-30% 0-5% 15-0% 10-15% 6-10% 3-6% 0-3% Constant term: bias of the pt-triggered signal distribution to higher multiplicity Gaussian term: Δη correlation structure underneath v-subtracted Δφ correlations. Width/amplitude/Npart-dependence same as inclusive correlations PHOBOS, PRL (010)

62 STAR vs PYTHIA 6 η < 1 4 < pttrig < 6 GeV/c 0.15 < ptassoc < 4 GeV/c STAR, PRL 95, (005) PHOBOS is limited by staticstics in p+p, therefore take PYTHIA as a reference, which matches the STAR measurement well.

63 Integrated ridge yield (near side) 0-10% Short-range Δη <1 Au+Au PYTHIA 63 NEAR side short-range minus PYTHIA long-range PHOBOS arxiv: Long-range -4 < Δη < - PHOBOS PHOBOS # of participant nucleons (Npart) PHOBOS, PRL (010)

64 AMPT Model 64 AMPT model: Glauber initial conditions, collective flow Elliptic flow subtracted correlations Correlations R R AMPT Au+Au 0-0% AMPT produces similar structures correlation structures Lin et. al. PRC7, (005) Ma et. Al. PLB (006)