new development in monte carlos for the lhc · new development in monte carlos for the lhc....

Introduction New tools Parton level Parton to hadron level Forthcoming attractions Hadron decays Conclusions

New development in Monte Carlos for the

LHC

Frank Krauss

IPPP Durham

RAL, 28.5.2008

F. Krauss IPPP

New development in Monte Carlos for the LHC


Outline

1 The next generation event generators

2 The new OO tools

3 Signals & backgrounds at the parton level

4 From parton level to exclusive studies at hadron level

5 Forthcoming attractions in Sherpa

6 Modelling hadron/tau decays

7 Conclusions

F. Krauss IPPP



Why simulate events?Many interesting signals at the LHC:Higgs (or alternative EWSB), SUSY, ED’s, . . .

But: Severe backgrounds in nearly all channels,(almost always with large influence of QCD)

=⇒ depend on detailed understanding of QCD.

Typically: Take backgrounds from data in some region,but extrapolate to signal region.

Examples:

Central jet-veto in VBF (Higgs)

Multi-jet backgrounds for SUSY (e.g. Z+jets)

Todays signals = tomorrows backgrounds.

F. Krauss IPPP



Why does anyone write a new event generator?New tools on the market: Pythia8, Herwig++, SherpaReflecting increased needs (precision, new physics, etc.):

getting rid of old errors (having new ones)

easier implementation of new physics models

incorporate new, better methods!

systematic inclusion of HO QCD

F. Krauss IPPP



The impact of HO QCD

Example: SUSY searches (4 jets + E/T ), observable: Meff

F. Krauss IPPP



Simulation’s paradigm

Basic strategyDivide event into stages,separated by different scales.

Signal/background:Exact matrix elements.

QCD-Bremsstrahlung:Parton showers (also in initial state).

Multiple interactions:Beyond factorization: Modeling.

Hadronization:

Non-perturbative QCD: Modeling.

Sketch of an event

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

F. Krauss IPPP



Introducing SHERPAT.Gleisberg, S.Hoche, F.K., A.Schalicke, S.Schumann and J.C.Winter, JHEP 0402 (2004) 056

New event generator, written from scratch in C++.

Fully automated matrix element generation,

Parton shower implementation (similar to PYTHIA),new improved parton shower formulations in preparation,

Unique feature: Multijet ME+PS merging,

Cluster hadronization model (still to be tuned to data),also interface to string fragmentation of Pythia,

Hadron and tau decays,

spin correlations,

Underlying event according to old Pythia model,new model based on BFKL evolution to be released.

F. Krauss IPPP



PYTHIA 8T.Sjostrand, S.Mrenna, and P.Skands, arXiv:0710.3820 [hep-ph]

Rewrite of Pythia 6.4 in C++.

2 → 2 matrix elements for some processes (limited w.r.t.old Fortran code), plus use of LHA input,

k⊥-ordered parton shower,

Lund-string fragmentation,

Hadron decays and link to external code,

interleafed underlying event model.

F. Krauss IPPP



Herwig++M. Bahr et al., arXiv:0804.3053 [hep-ph]

Rewrite and improvement of Herwig 6.5 in C++.

2 → 2 matrix elements for many processes (especiallyBSM), through a helicity-amplitude library, plus use ofLHA input,

new angle-ordered parton shower,

first processes through PowHEG method,

cluster fragmentation,

Hadron and τ decays,

spin correlations,

improved eikonal underlying event model.

F. Krauss IPPP



Parton level simulations

Stating the problem(s)

Multi-particle final states for signals & backgrounds.

Need to evaluate dσN :

∫

cuts

[

N∏

i=1

d3qi

(2π)32Ei

]

δ4

(

p1 + p2 −∑

i

qi

)

|Mp1p2→N |2 .

Problem 1: Factorial growth of number of amplitudes.

Problem 2: Complicated phase-space structure.

Solutions: Numerical methods.

F. Krauss IPPP




Factorial growth: e+e− → qq + ng

n #diags

0 11 22 83 484 384

1 2 3 4

Number of gluons 1

10

100

1000

Num

ber

of d

iagr

ams

F. Krauss IPPP




Taming the factorial growth in the helicity methodReusing pieces: Calculate only once!

Factoring out: Reduce number of multiplications!

Implemented as a-posteriori manipulations of amplitudes.

; Ze+ ; Z

e+

e�

e+

e��+ ; Z

e+ ; Z

e+

e�

e+

e��+

F. Krauss IPPP




Recursion methods (off-shell)

Basic idea: Recursively build one-particle off-shell currents(various versions of this: Berends-Giele, Alpha etc.).

“Classical” example: n-gluon amplitudes:Start with two on-shell gluons, represented by their polarization vectors,hence the currents associated with them are Jν (k) = ε

ν (k).

Then the two-gluon current reads (no colors) Jµ(k = k1 + k2) =ig3

(k1+k2)2V µνρJν (k1)Jρ(k2).

From this, larger and larger currents can be built recursively.

For quarks, the currents are given by spinors, and similar reasoning applies for the construction of theone-particle off-shell currents.

Treatment of color: Color-ordering the amplitudes

=⇒ C(1, ..., n) = Tr [T a1 . . . T an ], where T a are color matrices in fundamental representation.

Problem: Need to sum over all allowed permutations.

F. Krauss IPPP




Integration methods: Multi-channelingBasic idea: Translate Feynman diagrams into channels

=⇒ decays, s- and t-channel props as building blocks.R.Kleiss and R.Pittau, Comput. Phys. Commun. 83 (1994) 141

Integration methods: “Democratic” methods

Rambo/Mambo: Flat & isotropicR.Kleiss, W.J.Stirling and S.D.Ellis, Comput. Phys. Commun. 40 (1986) 359;

HAAG: Follows QCD antenna patternA.van Hameren and C.G.Papadopoulos, Eur. Phys. J. C 25 (2002) 563.

F. Krauss IPPP



Automated cross section calculation AMEGIC++F.K., R.Kuhn, G.Soff, JHEP 0202 (2002) 044.

Uses helicity/recursion methods;

Helicity method supplemented with “factoring out”(taming the factorial growth)

Phase space integration through multi-channeling(i.e. one phasespace mapping/Feynman diagram)

Implemented & tested models: SM, SM+AGC, THDM,MSSM, ADD.

Tested in > 1000 SM & > 500 MSSM channels.

Still under development(towards higher multis, more models, . . . )

F. Krauss IPPP



Implementing CSW recursion relations: A snaphotF.Cachazo, P.Svrcek and E.Witten, JHEP 0409 (2004) 006

Obtained summing over colours and helicities,sampling much better

Jet cross sections @ LHC, kmin⊥

= 20GeV

Process helicity MHV where possible MHV only(≤ 2 quark lines)

jj → jj 745.85 µb±0.10% 745.85 µb±0.10%57 s 44 s

jj → jjj 81.274 µb±0.20% 81.274 µb±0.20%826 s 166 s

gg → gggg 10.112 µb±0.23% 10.145 µb±0.23%1.5 ks 0.6 ks

jj → jjjj 23.23 µb±0.27% 23.245 µb±0.26% 23.208 µb±0.26%35 ks 7.6 ks 5.8 ks

gg → ggggg 2.6592 µb±0.16% 2.6915 µb±0.15%131 ks 41 ks

jj → jjjjj not possible 7.3829 µb±0.25% 7.3294 µb±0.17%970 ks 295 ks

F. Krauss IPPP



COMIX - a new matrix element generator for SherpaT.Gleisberg & S.Hoeche, in preparation

Colour-dressed Berends-Giele amplitudes in the SM

Fully recursive phase space generation

Example results (cross sections):

gg → ng Cross section [pb]n 8 9 10 11 12√

s [GeV] 1500 2000 2500 3500 5000Comix 0.755(3) 0.305(2) 0.101(7) 0.057(5) 0.019(2)Maltoni (2002) 0.70(4) 0.30(2) 0.097(6)Alpgen 0.719(19)

σ [µb] Number of jets

bb + QCD jets 0 1 2 3 4 5 6Comix 470.8(5) 8.83(2) 1.826(8) 0.459(2) 0.1500(8) 0.0544(6) 0.023(2)ALPGEN 470.6(6) 8.83(1) 1.822(9) 0.459(2) 0.150(2) 0.053(1) 0.0215(8)AMEGIC++ 470.3(4) 8.84(2) 1.817(6)

F. Krauss IPPP



COMIX - a new matrix element generator for SherpaT.Gleisberg & S.Hoeche, in preparation

Colour-dressed Berends-Giele amplitudes in the SM

Fully recursive phase space generation

Example results (phase space performance):

F. Krauss IPPP



Survey of public parton-level tools

Comparison of tree-level toolsTools:

Models 2 → n Ampl. Integ. public? lang.Alpgen SM n = 8 rec. Multi yes FortranAmegic SM,MSSM,ADD n = 6 hel. Multi yes C++CompHep SM,MSSM n = 4 trace 1Channel yes CHELAC SM n = 8 rec. Multi yes FortranMadEvent SM,MSSM,UED n = 6 hel. Multi yes FortranO’Mega SM,MSSM,LH n = 8 rec. Multi yes O’Caml

Typically fed into shower MC’s through LHA, methodthen MLM.

F. Krauss IPPP



Nomenclature

Specifying higher-order corrections: γ∗ → hadrons

In general: NnLO ↔ O(αns )

But: only for inclusive quantities(e.g.: total xsecs like γ

∗ →hadrons).

Counter-example: thrust distribution

In general, distributions are HO.

Distinguish real & virtual emissions:Real emissions → mainly distributions,virtual emissions → mainly normalization.

F. Krauss IPPP



Nomenclature

Anatomy of HO calculations: Virtual and realcorrections

NLO corrections: O(αs)Virtual corrections = extra loopsReal corrections = extra legs

UV-divergences in virtual graphs → renormalization

But also: IR-divergences in real & virtual contributionsMust cancel each other, non-trivial to see:N vs. N + 1 particle FS, divergence in PS vs. loop

F. Krauss IPPP



Orders in ME and PS

ME vs. PSMatrix elements good for:hard, large-angle emissions;take care of interferences.

Parton shower good for:soft, collinear emissions;resums large logarithms.

Want to combine both!Avoid double-counting.

αs vs. Log

resummed in PS

exact ME

LO 5jet, but also

NLO 4jet

L

α n

m

NLLexact ME

LO 4jet

4

4

4

4

4

5

5

5

5

5

5

5

F. Krauss IPPP



Correcting the parton shower

Example: e+e− → qqg

ME :

∣

∣

∣

∣

∣

+

∣

∣

∣

∣

∣

2

PS :

∣

∣

∣

∣

∣

∣

∣

∣

∣

∣

2

+

∣

∣

∣

∣

∣

∣

∣

∣

∣

∣

2

0

0.2

0.4

0.6

0.8

1

1x0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

2x

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1ME over PS

F. Krauss IPPP




Practicalities of ME-correctionsObviously, ME < PS is not always fulfilled.

Could enhance PS expression by a (large) factor.Question: Efficiency of the approach?

Therefore: realized in few processes only:Best-known: ee → qq, qq → V , t → bW

F. Krauss IPPP




Power showerCan use ME corrections for “power shower”:

This is the evil empire of MC event generators!

In qq → V , start parton shower @ spp.

Re-weight first emissions on both legs with ME.

Effect: More hard radiation through showering.

F. Krauss IPPP



MC@NLO

S.Frixione, B.R.Webber, JHEP 0206 (2002) 029

S.Frixione, P.Nason, B.R.Webber, JHEP 0308 (2003) 007

Basic principlesWant:

NLO-Normalization and first (hard) emission correct,Soft emissions correctly resummed in PS.

Method:Modify subtraction terms for real infrared divergences,use first order parton shower-expression,this is process-dependent!

In practise much more complicated.

Implemented for DY, W -pairs, gg → H , Q-pairs.

F. Krauss IPPP



MC@NLO

Example results: W -pairs @ Tevatron

F. Krauss IPPP



Combining MEs & PS: LO-MergingS.Catani, F.K., R.Kuhn and B.R.Webber, JHEP 0111 (2001) 063

F.K., JHEP 0208 (2002) 015

Want:All jet emissions correct at tree level + LL,Soft emissions correctly resummed in PS

Method:Separate Jet-production/evolution by Qjet (k⊥ algorithm).Produce jets according to LO matrix elementsre-weight with Sudakov form factor + running αs weights,veto jet production in parton shower.

Process-independent implementation.

F. Krauss IPPP



Combining MEs & PS

n-jet rates @ NLLS.Catani et al. Phys. Lett. B269 (1991) 432

At NLL-Accuracy

R2(Qjet) =ˆ

∆q(Ec.m., Qjet)˜2

R3(Qjet) = 2∆q(Ec.m., Qjet)

·Z

dq

"

αs (q)Γq(Ec.m., q)∆q(Ec.m., Qjet)

∆q(q, Qjet)

∆q(q, Qjet)∆g (q, Qjet)

#

Sudakov weightsExample: γ∗ → qqg

WSud =αs (q)

αs (Qjet)· ∆q(Ec.m., Qjet)

∆q(Ec.m., Qjet)

∆q(q, Qjet)∆q(q, Qjet)∆g (q, Qjet)

F. Krauss IPPP



Combining MEs & PS

Algorithm as scale-setting prescriptionExample: p⊥ distribution of jets @ Tevatron

Consider exclusive W + 1- and W + 2-jet productionComparison with MCFM; J.Campbell and R.K.Ellis, Phys. Rev. D 65 (2002) 113007

in : F.K., A.Schalicke, S.Schumann and G.Soff, Phys. Rev. D 70 (2004) 114009

20 40 60 80 100 120 140 160 180p

T (jet) [GeV]

10-5

10-4

10-3

10-2

10-1

1/σ

dσ/d

p T [

1/G

eV]

MCFM NLOSherpaLO

Wj @ Tevatron

PDF: cteq6lCuts: p

Tlep> 20 GeV, |ηlep

|<1

pT

jet> 15 GeV, |ηjet|<2

pT

miss> 20 GeV

∆Rjj> 1.0

20 40 60 80 100 120 140 160 180p

T (first jet) [GeV]

10-4

10-3

10-2

10-1

1/σ

dσ/d

p T [

1/G

eV]

MCFM NLOSherpaLO

Wjj @ Tevatron

20 40 60 80 100p

T (second jet) [GeV]

10-4

10-3

10-2

10-1

1/σ

dσ/

dpT [

1/G

eV]

PDF: cteq6l

pT

jet> 15 GeV, |ηjet|<2

Cuts: pT

lep> 20 GeV, |ηlep|<1

pT

miss> 20 GeV

∆Rjj> 1.0

Sherpa = tree-level matrix elements with αs scales and Sudakov form factors.

F. Krauss IPPP



Combining MEs & PS

Vetoing the shower

WVeto =

(

1 +

Z

Ec.m.

Qjet

dq Γq(Ec.m., q) +

Z

Ec.m.

Qjet

dq Γq(Ec.m., q)

Z

q

Qjet

dq′Γq(Ec.m.q

′) + · · ·

)2

=

(

exp

Z

Ec.m.

Qjet

dq Γq(Ec.m., q)

!)2

= ∆−2q (Ec.m., Qjet)

=⇒ Cancels dependence on Qjet.

F. Krauss IPPP



Combining MEs & PS: Independence on Qjet

F.K., A.Schalicke, S.Schumann and G.Soff, Phys. Rev. D 70 (2004) 114009

Example: p⊥ of W in pp → W + X @ Tevatron

Qjet = 10 GeV Qjet = 30 GeV Qjet = 50 GeV

/ GeV Wp20 40 60 80 100 120 140 160 180

[ p

b/G

eV ]

W

/dp

σd

-210

-110

1

10

210

SHERPA

W + XW + 0jetW + 1jetW + 2jetsW + 3jets

/ GeV Wp20 40 60 80 100 120 140 160 180

[ p

b/G

eV ]

W

/dp

σd

-210

-110

1

10

210

SHERPA


/ GeV Wp20 40 60 80 100 120 140 160 180

[ p

b/G

eV ]

W

/dp

σd

-210

-110

1

10

210

SHERPA


F. Krauss IPPP



Data: Azimuthal decorrelations of jets at the

Tevatron

IdeaCheck QCD radiation pattern

∆Φ12 @ Run II (D0)

SHERPASHERPASHERPASHERPASHERPASHERPASHERPASHERPA

Sherpa

/2π π3/4 π/2π π3/4 π/2π π3/4 π/2π π3/4 π/2π π3/4 π/2π π3/4 π/2π π3/4 π/2π π3/4 π

> 180 GeV (x8000)maxTp

< 180 GeV (x400)maxT130 < p

< 130 GeV (x20)maxT100 < p

< 100 GeVmaxT75 < p

dije

tφ∆

/dd

ijet

σ d

dije

tσ

1/

-310

-210

-110

1

10

210

310

410

510

dijetφ∆

F. Krauss IPPP



Comparison with data from Tevatron

p⊥ of Z -bosons

/ GeV Z P0 20 40 60 80 100 120 140 160 180 200

10-3

10-2

10-1

1

10pt Z

Z + 0 jet

Z + 1 jet

Z + 2 jet

CDF

GeVpb

/

dPσ

d

/ GeV Z P0 5 10 15 20 25 30 35 40 45 50

G

eVpb

/

dPσ

d

1

10

pt Z

Z + 0 jet

Z + 1 jet

Z + 2 jet

CDF

F. Krauss IPPP



Comparison with RunII Z + X data: Jet multis(D0-Note 5066)

0 1 2 3 4 5 6

1

10

210

310

410

0 1 2 3 4 5 6

1

10

210

310

410

data w/stat errordata w/stat & sys errorPythia range statPythia range stat & sys

D0 RunII Preliminary

Jet Multiplicity

Nr.

of

Eve

nts

0 1 2 3 4 5 60.2

1

234

Jet Multiplicity

Dat

a / P

YT

HIA

0 1 2 3 4 5 6

1

10

210

310

410

0 1 2 3 4 5 6

1

10

210

310

410

data w/stat errordata w/stat & sys errorSherpa range statSherpa range stat & sys


Jet Multiplicity

Nr.

of

Eve

nts

0 1 2 3 4 5 60.2

1

234

Jet Multiplicity

Dat

a / S

HE

RP

A

F. Krauss IPPP



Combining MEs & PS

Comparison with data from Tevatronp⊥ of Z -bosons in pp → Z + X

Data from CDF, Phys. Rev. Lett. 84 (2000) 845

/ GeV Z P0 20 40 60 80 100 120 140 160 180 200

10-3

10-2

10-1

1

10pt Z

Z + 0 jet

Z + 1 jet

Z + 2 jet

CDF

GeVpb

/

dPσ

d

/ GeV Z P0 5 10 15 20 25 30 35 40 45 50

G

eVpb

/

dPσ

d1

10

pt Z

Z + 0 jet

Z + 1 jet

Z + 2 jet

CDF

F. Krauss IPPP



Combining MEs & PS

Comparison with data from TevatronJet rates in pp → Z + X (D0-Note 5066)

0 1 2 3 4 5 6

1

10

210

310

410

0 1 2 3 4 5 6

1

10

210

310

410



Jet Multiplicity

Nr.

of

Eve

nts

0 1 2 3 4 5 60.2

1

234

Jet Multiplicity

Dat

a / P

YT

HIA

0 1 2 3 4 5 6

1

10

210

310

410

0 1 2 3 4 5 6

1

10

210

310

410



Jet Multiplicity

Nr.

of

Eve

nts

0 1 2 3 4 5 60.2

1

234

Jet Multiplicity

Dat

a / S

HE

RP

A

F. Krauss IPPP



Combining MEs & PS


Jet spectra (1st jet) in pp → Z + X (D0-Note 5066)

50 100 150 200 250 300 350

1

10

210

310

50 100 150 200 250 300 350

1

10

210

310



jet [GeV]st 1T

p

Nr.

of

Eve

nts

50 100 150 200 250 300 3500.2

1

234

jet [GeV]st 1T

p

Dat

a / P

YT

HIA

50 100 150 200 250 300 350

1

10

210

310

50 100 150 200 250 300 350

1

10

210

310



jet [GeV]st 1p

Nr.

of

Eve

nts

50 100 150 200 250 300 3500.2

1

234

jet [GeV]st 1T

p

Dat

a / S

HE

RP

A

F. Krauss IPPP



Combining MEs & PS


Jet spectra (2nd jet) in pp → Z + X (D0-Note 5066)

20 40 60 80 100 120 140 160 180

1

10

210

310

20 40 60 80 100 120 140 160 180

1

10

210

310



jet [GeV]nd 2T

p

Nr.

of

Eve

nts

20 40 60 80 100 120 140 160 1800.2

1

234

jet [GeV]nd 2T

p

Dat

a / P

YT

HIA

20 40 60 80 100 120 140 160 180

1

10

210

310

20 40 60 80 100 120 140 160 180

1

10

210

310data w/stat errordata w/stat & sys errorSherpa range statSherpa range stat & sys


jet [GeV]nd 2T

p

Nr.

of

Eve

nts

20 40 60 80 100 120 140 160 1800.2

1

234

jet [GeV]nd 2T

p

Dat

a / S

HE

RP

A

F. Krauss IPPP



Comparison with RunII Z + X data: pj3⊥

(D0-Note 5066)

20 40 60 80 100 120

1

10

210

20 40 60 80 100 120

1

10

210



jet [GeV]rd 3T

p

Nr.

of

Eve

nts

20 40 60 80 100 1200.2

1

234

jet [GeV]rd 3T

p

Dat

a / P

YT

HIA

20 40 60 80 100 120

1

10

210

20 40 60 80 100 120

1

10

210



jet [GeV]rd 3T

p

Nr.

of

Eve

nts

20 40 60 80 100 1200.2

1

2345

jet [GeV]rd 3T

p

Dat

a / S

HE

RP

A

F. Krauss IPPP



Combining MEs & PS


Azimuthal correlation (∠1.jet,2.jet) in pp → Z + X (D0-Note 5066)

0 0.5 1 1.5 2 2.5 30

50

100

150

200

250

0 0.5 1 1.5 2 2.5 30

50

100

150

200

250 data w/stat errordata w/stat & sys errorPythia range statPythia range stat & sys


(jet,jet)φ ∆

Nr.

of

Eve

nts

0 0.5 1 1.5 2 2.5 30.2

1

234

(jet,jet)φ ∆

Dat

a / P

YT

HIA

0 0.5 1 1.5 2 2.5 30

50

100

150

200

250

0 0.5 1 1.5 2 2.5 30

50

100

150

200

250data w/stat errordata w/stat & sys errorSherpa range statSherpa range stat & sys


(jet,jet)φ ∆

Nr.

of

Eve

nts

0 0.5 1 1.5 2 2.5 30.2

1

234

(jet,jet)φ ∆

Dat

a / S

HE

RP

A

F. Krauss IPPP



A new feature: Merging in decay chains

Example: top-pair production @ LHC

0-0-0 1-0-0 0-1/0-0/10-1/0-0/1 1-1/0-0/11-1/0-0/1 0-1-10-0-0 1-0-0 0-1/0-0/10-1/0-0/1 1-1/0-0/11-1/0-0/1 0-1-10-0-0

= 80/20 GeVcutQ = 20/20 GeVcutQ

Apacic++ 2.0

pb

/GeV

W-,

W+,

b-j

et 1

,b-j

et 2

/dp

σ dσ

1/

-510

-410

-310

-210

-110

Set / Mean - 1 (CKKW)Set / Mean - 1 (CKKW)

-0.4-0.2

00.20.4

GeV W-,W+,b-jet 1,b-jet 2

p0 50 100 150 200 250 300 350 400 450 500

0-0-0 1-0-0 0-1/0-0/10-1/0-0/1 1-1/0-0/11-1/0-0/1 0-1-10-0-0 1-0-0 0-1/0-0/10-1/0-0/1 1-1/0-0/11-1/0-0/1 0-1-10-0-0

= 80/20 GeVcutQ = 20/20 GeVcutQ

Apacic++ 2.0

pb

/GeV

no

n-b

-jet

1/d

pσ

dσ1/

-310

-210

-110

Set / Mean - 1 (CKKW)Set / Mean - 1 (CKKW)

-0.4-0.2

00.20.4

GeV non-b-jet 1

p0 20 40 60 80 100 120 140 160 180 200

F. Krauss IPPP



Combining MEs & PS

Comparison with other merging algorithms: MLMp⊥ of jets in inclusive W+jets at Tevatron

F. Krauss IPPP



Combining MEs & PS

Comparison with other merging algorithms: MLMp⊥ of jets in inclusive W+jets at LHC

F. Krauss IPPP



Dipole showerImplemented in Ariadne ( L.Lonnblad, Comput. Phys. Commun. 71, 15 (1992)).

UpshotExpansion around soft logs, particles always on-shell

dσ = σ0CF αs(k2

⊥)

2π

dk2⊥

k2⊥

dy .

Always color-connected partners (recoil of emission)=⇒ emission: 1 dipole → 2 dipoles.

Quantum coherence on similar grounds for angular andkT -ordering.

F. Krauss IPPP



Dipole shower

Implemented in Ariadne ( L.Lonnblad, Comput. Phys. Commun. 71, 15 (1992)).

Radiation pattern IS RadiationThere is none!Treat radiation in DIS as FS radiation betweenremnant & quark

Thus, no real Dipole Shower for pp collisions.

Cut FS phase space ofremnants:

F. Krauss IPPP



Dipole shower: New developments

J.Winter & F.K., arXiv:0712.3913 [hep-ph]

Initial state dipole showers

Complete perturbative formulation.

Dipoles and their radiationassociated to IS-IS, IS-FS andFS-FS colour lines.

Beam remnants kept outsideevolution.

Onshell kinematics, evolution in k⊥.

F. Krauss IPPP



Dipole shower: New developments

J.Winter & F.K., arXiv:0712.3913 [hep-ph]

Initial state dipole showers

Testbed: DY production.

PT spectrum of Z 0 boson.

Mainly recoils vs. 1stemission:by construction:ME-corrected.

)-e+ (eTp

Tevatron Run 1SHERPA

CDF data (2000)

)-e+ (eTp


)M, & Py6.2 hadr.2=(1+ ,ini

Dipole shower, p

)-e+ (eTp


, & Py6.2 hadr. ,max=p ,ini

Dipole shower, p [p

b/G

eV]

T/d

pσd

-310

-210

-110

1

10

[GeV]Tp(M

C-d

ata)

/ d

ata

-0.5

0

0.5

[GeV]Tp0 20 40 60 80 100 120 140 160 180 200

)-e+ (eTp


CDF data (2000)

)-e+ (eTp


)M, & Py6.2 hadr.2=(1+ ,ini

Dipole shower, p

)-e+ (eTp


, & Py6.2 hadr. ,max=p ,ini

Dipole shower, p [p

b/G

eV]

T/d

pσd

5

10

15

20

25

30

[GeV]Tp(M

C-d

ata)

/ d

ata

-0.2-0.1

00.10.2

[GeV]Tp0 2 4 6 8 10 12 14 16 18 20

F. Krauss IPPP



A new parton shower approach

Using Catani-Seymour splitting kernelsFirst discussed in: Z.Nagy and D.E.Soper, JHEP 0510 (2005) 024;

Implemented by M.Dinsdale, M.Ternick, S.Weinzierl Phys.Rev.D76 (2007) 094003,

and S.Schumann& F.K., JHEP 0803 (2008) 038.

Catani-Seymour dipole subtraction terms as universalframework for QCD NLO calculations.

Factorization formulae for real emission process:

Full phase space coverage & good approx. to ME.

Example: final-state final-state dipoles

splitting: pij + pk → pi + pj + pk

variables: yij,k =pi pj

pi pj +pi pk +pj pk, zi =

pi pkpi pk +pj pk

consider qij → qigj : 〈Vqigj ,k(zi , yij,k )〉 = CF

21−zi +zi yij,k

− (1 + zi )

ff

F. Krauss IPPP



A new parton shower approach

Using Catani-Seymour splitting kernelsFirst discussed in: Z.Nagy and D.E.Soper, JHEP 0510 (2005) 024;

Implemented by M.Dinsdale, M.Ternick, S.Weinzierl Phys.Rev.D76 (2007) 094003,

and S.Schumann& F.K., JHEP 0803 (2008) 038.

F. Krauss IPPP



First steps towards NLO

Automated Catani-Seymour subtractionT.Gleisberg& F.K., eprint 0709.2881

-0,34

-0,32

-0,3

-0,28

-0,26

σ real

,sub

trac

ted [

pb]

-11-10-9-8-7-6-5-4-3-2-1

log(αcut

)

-0,4%

-0,2%

0

+0,2%

+0,4%

(a)

-1,5

-1

-0,5

0

σ real

,sub

trac

ted [

pb]

-11-10-9-8-7-6-5-4-3-2-1log(α

cut)

-4%

-2%

0

+2%

+4%

(b)

(a) e+e− → 2jets

(b) e+e− → 3jets

220

240

260

280

300

320

σ real

,sub

trac

ted [

pb]

-11-10-9-8-7-6-5-4-3-2-1log(α

cut)

-1%

-0,5%

0

+0,5%

+1%

(c)

-10

-8

-6

-4

-2

0

σ real

,sub

trac

ted [

106 p

b]

-11-10-9-8-7-6-5-4-3-2-1log(α

cut)

-4%

-2%

0

+2%

+4%

(d)

(c) ep → e + 1jet

(d) pp → 2jets

F. Krauss IPPP





-30

-20

-10

0

10

20

30

σ [p

b]

realvirtualsum

-2-10log(α)

-4%

-2%

0

+2%

+4%

(a)-10

0

10

20

σ [p

b]

realvirtualsum

-2-10log(α)

-2%

-1%

0

+1%

+2%

(b)

(a) e+e− → 2jets

(b) e+e− → 3jets

-5

0

5

σ [n

b]

realvirtualsum

-2-10log(α)

-6%-4%-2%

0+2%+4%+6%

(c)-500

0

500

σ [p

b]

realvirtualsum

-2-10log(α)

-2%

-1%

0

+1%

+2%

(d)

(c) ep → e + 1jet

(d) pp → W

F. Krauss IPPP





0 0,1 0,2 0,3 0,4 0,51-Thrust

10-3

10-2

10-1

100

101

1/N

dσ/

d(1-

Thr

ust)

DELPHILONLO

0 0,1 0,2 0,3 0,4 0,5 0,6Major

10-2

10-1

100

101

1/N

dσ/

d(M

ajor

)

DELPHILONLO

10-3

10-2

10-1

100

101

102

103

dσ/d

p T [

pb]

LONLO

0 20 40 60 80 100 120 140 160p

T(first jet) [GeV]

0,81

1,2

1,4

NL

O/L

O 10-1

100

101

102

103

dσ/d

η [p

b]

LONLO

-4 -3 -2 -1 0 1 2 3 4η(first jet)

0,51

1,52

NL

O/L

O

F. Krauss IPPP



Soft physics simulation in SherpaImplemented a new version of the cluster fragmentationmodel;

a new module for the simulation of hadron and τ decays(special emphasis on τ , B and D decays, includingmixing, CP-violation etc.);

a new module for the simulation of photon radiation inhadron decays based on the YFS approach;

all in current, new release, Sherpa 1.1.

F. Krauss IPPP



Example (1): Spin correlations in H → τ+τ−

Angle of planes of decay products (piν) in c.m.s

0

0.1

0.2

0.3

0.4

0.5

0.6

0 0.5 1 1.5 2 2.5 3

no spin correlationsspin correlations with scalar Higgs

spin correlations with pseudoscalar Higgs

F. Krauss IPPP



Example (2): Form factors in decays

B → D∗ℓν τ → ππν

F. Krauss IPPP



Example (3): BB-mixing and decay into J/ΨKS

-10 -5 0 5 10

Eve

nts

0

10000

20000

30000

40000

50000

60000

tag candidates0B

tag candidates0

B

t(ps)∆-10 -5 0 5 10

Can

did

ate

Asy

mm

etry

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

SherpaTheory C=0 S=0.725

F. Krauss IPPP



Example (4): Photon radiation in W → ℓν

[GeV]γE0 10 20 30 40 50

]-1

[G

eVγ

dEΓd

tot

Γ1

-610

-510

-410

-310

-210

WINDEC

PHOTONS++

AMEGIC++

total photon energy in decay frame

F. Krauss IPPP



Example (5): Photon radiation in J/ψ → ℓℓ

[GeV]γE0.5 1 1.5 2 2.5 3

]-1

[G

eVγ

dEΓd

tot

Γ1

-710

-610

-510

-410

-310

ee→(1S) ψJ/

µµ →(1S) ψJ/

(1S) rest frameψtotal photon energy radiated in the J/

θ0 0.5 1 1.5 2 2.5 3

θdΓd to

tΓ1

-510

-410

-310

-210

-110

approximated matrix element

exact matrix element

eikonal factors only

radiated in Z rest frame-e+e→photon angular distribution in decay Z

F. Krauss IPPP



Summary & outlookMany interesting signals at LHC “spoiled” by QCD.

Simulation tools mandatory for success of LHC

Various new OO-projects in C++.

New methods of merging of ME& PS extremely powerful(Complementary to MC@NLO)

Sherpa a versatile tool - new features becoming available:higher ME multis, new showers, new Underlying Eventmodel (based on k⊥-factorization), more BSM models.

Plan: Go to NLO

automatic dipole subtraction implemented and testedbuild library of virtual corrections.

F. Krauss IPPP


new development in monte carlos for the lhc · new development in monte carlos for the lhc....

Documents