From Jet Scaling to Jet Vetos - thphys.uni-heidelberg.deplehn/includes/talks/2012/atlas_12.pdf ·...

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

From Jet Scaling to Jet Vetos

Tilman Plehn

Heidelberg

ATLAS Higgs-Tau Workshop, 3/2012

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Jet counting

Why count jets

– complete event reconstruction crucial at LHC [Higgs plus 0,1,2 jets; jets plus /~pT ]

– utilize tagging and recoil jets in Higgs searches

– reduce t t and gg backgrounds

– identify decay jets in BSM searches

⇒ dσ/dnjets just another distribution to cut on?

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Jet counting

Why count jets






Why not [intro: arXiv:0910.4182, Springer Lecture Notes]

– remember qg → qZ

– collinear divergence from g → qq splitting [overlapping with soft divergence]Z pmaxT

pminT

dp2T

Cp2

T

= 2Z pmax

T

pminT

dpT pTCp2

T

' 2CZ pmax

T

pminT

dpT1

pT= 2C log

pmaxT

pminT

universal form following factorization

σn+1 =

Zσn

dp2a

p2a

dzαs

2πP(z)

– universally divergent, fixed order poorly defined

⇒ find object to ‘renormalize’ [i.e. absorbe universal divergence]

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Jet counting

Why count jets






DGLAP equation and jet-inclusive rates

– re-organize perturbation series [sum collinear logs]

σn+1(x, µ) ∼1n!

1

2πb0log

αs(µ20)

αs(µ2)

!n Z 1

x0

dxn

xnP„

xxn

«· · ·Z 1

x0

dx1

x1P„

x2

x1

«σ1(x1, µ0)

– DGLAP equivalent to ‘infrared RGE’

– factorization scale as inclusive momentum cutoff [vanishing at all orders]

σtot(µ) =

Z 1

0dx1

Z 1

0dx2

Xij

fi (x1, µ) fj (x2, µ) σij (x1x2S, µ)

⇒ collinear jets automatically included [veto on hard jets ok only if pmaxT ,j > hard scale]

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Jet counting

Why count jets






DGLAP equation and jet-inclusive rates

– re-organize perturbation series [sum collinear logs]

σn+1(x, µ) ∼1n!

1

2πb0log

αs(µ20)

αs(µ2)

!n Z 1

x0

dxn

xnP„

xxn

«· · ·Z 1

x0

dx1

x1P„

x2

x1

«σ1(x1, µ0)

– DGLAP equivalent to ‘infrared RGE’

– factorization scale as inclusive momentum cutoff [vanishing at all orders]

σtot(µ) =

Z 1

0dx1

Z 1

0dx2

Xij

fi (x1, µ) fj (x2, µ) σij (x1x2S, µ)

⇒ collinear jets automatically included [veto on hard jets ok only if pmaxT ,j > hard scale]

⇒ but still...

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Poisson scaling

Theory: soft gluon radiation [Peskin & Schroeder Ch 6]

– example: photons off hard electrononly abelian diagrams, successive radiation

– eikonal approximation

Mn+1 = gsT aε∗µ(k) u(q)

qµ +O( /k)

(qk) +O(k2)Mn

– factorization of ‘hard process’ and soft radiation factors

– Poisson distribution [normalized pdf for n if n expected]

σn =nne−n

n!⇐⇒ R(n+1)/n =

σn+1

σn=

nn + 1

Basis of Poisson distribution

1– radiation matrix element nn:abelian fine, non-abelian a little tricky [ISR-FSR, see Ioffe, Fadin, Lipatov]

2– phase space factor 1/n!:only combinatorics, matrix element ordered [angular ordering, color suppression]

3– normalization factor e−n

– just like parton shower in collinear regimes

s

s

1/pT

1/pT

1/pT

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling

Experiment: from UA1 to ATLAS/CMS [Steve Ellis, Kleiss, Stirling]

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling


– W/Z+jets production– many equivalent descriptions

R(n+1)/n =σn+1

σn= const

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling



R(n+1)/n =σn+1

σn= const

– same for inclusive and exclusive rates [Blackhat-Sherpa]

R incl(n+1)/n =

P∞j=n+1 σ

(excl)j

σ(excl)n +

P∞j=n+1 σ

(excl)j

= Rexcl(n+1)/n

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling



R(n+1)/n =σn+1

σn= const


R incl(n+1)/n =

P∞j=n+1 σ

(excl)j

σ(excl)n +

P∞j=n+1 σ

(excl)j

= Rexcl(n+1)/n

– confirmed by ATLAS/CMS

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling



R(n+1)/n =σn+1

σn= const


R incl(n+1)/n =

P∞j=n+1 σ

(excl)j

σ(excl)n +

P∞j=n+1 σ

(excl)j

= Rexcl(n+1)/n


Theoretical studies [Englert, TP, Schichtel, Schumann]

– CKKW/MLM merging [we used Sherpa]

– phase space effects? → moderate

– αs uncertainties? → small

– scale uncertainties? → tuning parameter?jetsn

2 3 4 5 6

jets

dndN

210

310

410

510

610

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

jetsn2 3 4 5 6

jets

dndN

210

310

410

510

610

jetsn2 3 4 5 6

var

iati

onsα

0.5

1

1.5 theoretical uncertainty

statistical uncertainty

jetsn2 3 4 5 6

var

iati

onsα

0.5

1

1.5

jetsn2 3 4 5 6

var

iati

onµ

1

10

jetsn2 3 4 5 6

var

iati

onµ

1

10

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Staircase scaling



R(n+1)/n =σn+1

σn= const


R incl(n+1)/n =

P∞j=n+1 σ

(excl)j

σ(excl)n +

P∞j=n+1 σ

(excl)j

= Rexcl(n+1)/n


Theoretical studies [Englert, TP, Schichtel, Schumann]

– CKKW/MLM merging [we used Sherpa]

– phase space effects? → moderate

– αs uncertainties? → small

– scale uncertainties? → tuning parameter?

– same for QCD jets

– correctly described by ME-PS merging!?

jetsn2 3 4 5 6

jets

dndN

410

510

610

710

810

910

1010

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

QCD

µQCD, 1/4

µQCD, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

QCD

µQCD, 1/4

µQCD, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

QCD

µQCD, 1/4

µQCD, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

QCD

µQCD, 1/4

µQCD, 4

jetsn2 3 4 5 6

jets

dndN

410

510

610

710

810

910

1010

jetsn2 3 4 5 6

var

iati

onsα 0.8

1



jetsn2 3 4 5 6

var

iati

onsα 0.8

1

1.2

jetsn2 3 4 5 6

var

iati

onµ 1

10

210

jetsn2 3 4 5 6

var

iati

onµ 1

10

210

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Interpolating scaling patterns

Scaling for photon plus jets [Englert, TP, Schichtel, Schumann]

– naively, no scaling at all [CMS, private complaint]

– after appropriate cuts, great playground

1– staircasedemocratic γ and jet acceptancelarge γ-jet separation [m or ∆R, no large logs]

no reason for ordered emission [1/n! in Poisson form]

dominant: non-abelian splitting of ISR gluonhelping: pdf effect shifting to high-njets

2/1 3/2 4/3 5/4 6/5 7/6

nn+

1R

0

0.1

0.2

0.3

0.4

0.5

=-0.0064dndR

=0.15970R> 70 GeV, ,jγm

=-0.0007dndR

=0.14880R> 90 GeV, ,jγm

=0.0029dndR

=0.14070R>110 GeV, ,jγm

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Interpolating scaling patterns

Scaling for photon plus jets [Englert, TP, Schichtel, Schumann]

– naively, no scaling at all [CMS, private complaint]

– after appropriate cuts, great playground

1– staircasedemocratic γ and jet acceptancelarge γ-jet separation [m or ∆R, no large logs]

no reason for ordered emission [1/n! in Poisson form]

dominant: non-abelian splitting of ISR gluonhelping: pdf effect shifting to high-njets

2/1 3/2 4/3 5/4 6/5 7/6

nn+

1R

0

0.1

0.2

0.3

0.4

0.5

=-0.0064dndR

=0.15970R> 70 GeV, ,jγm

=-0.0007dndR

=0.14880R> 90 GeV, ,jγm

=0.0029dndR

=0.14070R>110 GeV, ,jγm

2– Poissongenerate ‘hard process’ [m, pT ,∆R, ...]

lower general pminT for soft-collinear logarithm

rely on lot-enhanced radiation

dominant: successive ordered ISRremaining: high-n staircase tail

⇒ staircase–Poisson transition tunable!

2/1 3/2 4/3 5/4 6/5 7/6 8/7n

n+1

R0

0.5

1

1.5 =1.7138n

=0.01310R>100 GeV,

T,j1p

=2.361n

=-0.05090R>150 GeV,

T,j1p

=2.6287n

=-0.05360R>200 GeV,

T,j1p

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Jet veto in Higgs searches

Jet veto in Higgs analyses [Barger, Phillips, Zeppenfeld; Rainwater]

– particularly useful for WBF signals [remove t t and Z+jets backgrounds]

– veto central (semi-hard) jetsestimate Pvetoapply as ‘efficiency factor’

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics





– implicit in LHC Higgs searchesindividual searches for exclusive fixed njets

– problem in Higgs phenomenology [mine for more than 10 years]

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics







Using jet counting [Gerwick, TP, Schumann]

– avoid Pveto as single number

– study exclusive njets distribution:1– understand basic features:

staircase for inclusive samplesPoisson for radiation processes

2– predict from theory [including error]

3– validate simulation4– estimate alowed parameter range5– extrapolate to Higgs

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics







Using jet counting [Gerwick, TP, Schumann]

– avoid Pveto as single number

– study exclusive njets distribution:1– understand basic features:

staircase for inclusive samplesPoisson for radiation processes

2– predict from theory [including error]

3– validate simulation4– estimate alowed parameter range5– extrapolate to Higgs

staircase scaling Poisson scaling

σn σ0 e−bn σtote−n nn

n!

Rexcl(n+1)/n e−b n

n + 1

Rincl(n+1)/n e−b

0@ (n + 1) e−n n−(n+1)

Γ(n + 1) − nΓ(n, n)+ 1

1A−1

〈njets〉1

2

1

cosh b − 1n

Pveto 1 − e−b e−n

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics


Example: WBF H → ττ [Gerwick, TP, Schumann]

– staircase scaling before WBF cuts [QCD and e-w processes]

– first emission sensitive to cuts

– e-w Zjj production with too many structures

1

n!/n+1!-1 0 1 2 3 4 5 6

) n!/

n+1

!R(

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Higgs gluon fusion

Z QCD

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

R(n

+1)

/n

n + 1

n

1/0 2/1 3/2 4/3 5/4 6/5

1

n!/n+1!-1 0 1 2 3 4 5 6

) n!/

n+1

!R(

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Higgs WBF

Z EW

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

R(n

+1)

/n

n + 1

n

1/0 2/1 3/2 4/3 5/4 6/5

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics






Forward tagging jets

– count add’l veto jets to reduce backgrounds

pvetoT > 20 GeV min y1,2 < yveto

< max y1,2

– Poisson for QCD processes [‘radiation’ pattern]

1

n!/n+1!-1 0 1 2 3 4 5

) n!/

n+1

!R(

0

0.5

1

1.5

2

2.5

Higgs gluon fusion

Z QCD

n(Z QCD) = 1.42

n(Higgs gg fusion) = 1.80

0.5

1.0

1.5

2.0

2.5

R(n

+1)

/n

n + 1

n

1/0 2/1 3/2 4/3 5/4

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics






Forward tagging jets

– count add’l veto jets to reduce backgrounds

pvetoT > 20 GeV min y1,2 < yveto

< max y1,2

– Poisson for QCD processes [‘radiation’ pattern]

– staircase-like for e-w processes [generic]

– features of njets distributions understood

⇒ cut on njets fine, your job now1

n!/n+1!-1 0 1 2 3 4 5

) n!/

n+1

!R(

0

0.5

1

1.5

2

2.5

Higgs gluon fusion

Z QCD

n(Z QCD) = 1.42

n(Higgs gg fusion) = 1.80

0.5

1.0

1.5

2.0

2.5

R(n

+1)

/n

n + 1

n

1/0 2/1 3/2 4/3 5/4

1

n!/n+1!-1 0 1 2 3 4 5

) n!/

n+1

!R(

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Higgs WBF

Z EW

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

R(n

+1)

/n

n + 1

n

1/0 2/1 3/2 4/3 5/4

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Higgs couplings

Veto probabilities crucial for coupling extraction [SFitter: Lafaye, TP, Rauch, Zerwas (2009ff)]

– theory:Higgs portal? [φ†φ gauge singlet]

γγH and ggH couplings w/o naive decouplingSUSY or strongly interacting Higgs simple coupling measurements

– experiment:Higgs decays to fermions crucialassociated Higgs-jet channels/WBF crucial

– current observables: σ × BRmore relevant: gjjH

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Higgs couplings






Dedicated parameter analysis in progress [SFitter+Duhrssen, Klute]

– general fit of all couplings [Moriond 2012]

gjjH = gSMjjH`1 + ∆j

´– prelim: SM central values, proper errors

– pre-Moriond

0

0.5

1

1.5

2

110 115 120 125 130 135 140 145 150

mH [GeV]

68% CL ATLAS+CMS

∆W∆Z∆t∆b∆

τ∆H

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Higgs couplings










– pre-Moriondend of 2012

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

110 115 120 125 130 135 140 145 150

mH [GeV]

68% CL ATLAS+CMS

∆W∆Z∆t∆b∆

τ∆H

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Higgs couplings










– pre-Moriondend of 2012assuming mH = 125 GeV

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

∆H

∆W

∆Z

∆t

∆b

∆τ

∆γ

∆g

∆Z/W

∆τ/b

∆b/W

gj = gjSM

(1+∆j)

68% CL: ATLAS + CMS

7 TeV, 20 fb-1

14 TeV, 30 fb-1

no D5

14 TeV, 30 fb-1

with D5

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Higgs couplings










– pre-Moriondend of 2012assuming mH = 125 GeV

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

∆H

∆W

∆Z

∆t

∆b

∆τ

∆γ

∆g

∆Z/W

∆τ/b

∆b/W

gj = gjSM

(1+∆j)

68% CL: ATLAS + CMS

7 TeV, 20 fb-1

14 TeV, 30 fb-1

no D5

14 TeV, 30 fb-1

with D5

⇒ LHC task for coming years

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

New physics

Effective mass [Englert, TP, Schichtel, Schumann]

– obviously sensitive to new physics massesbut awful variable, after theory uncertainty

– correlation meff ∼ 〈pT 〉 × njets

– use merged sample for meffestimate scale and αs uncertainties

[GeV]effm200 400 600 800

[1/

100

GeV

]ef

fdmdN 310

410

510

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

2≥ jn

> 50 GeVT, j

p

-1L = 1 fb

W+jets

µW+jets, 1/4

µW+jets, 4

[GeV]effm200 400 600 800

[1/

100

GeV

]ef

fdmdN 310

410

510

[GeV]effm200 400 600 800

var

iati

onsα 0.9

1



[GeV]effm200 400 600 800

var

iati

onsα 0.9

1

1.1

[GeV]effm200 400 600 800

var

iati

onµ

1

10

[GeV]effm200 400 600 800

var

iati

onµ

1

10

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

New physics





– estimate SUSY significance as function of meff

[GeV]effm200 400 600 800 1000 1200 1400

[1/

100

GeV

]ef

fdmdN

-210

-1101

10

210

310

410

510 signalleptonictt

hadronicttW+jetsZ+jetsQCD

[GeV]effm200 400 600 800 1000 1200 1400

[1/

100

GeV

]ef

fdmdN

-210

-1101

10

210

310

410

510

[GeV]effm200 400 600 800 1000 1200 1400

BS+

B

0.5

1

1.5

2

2.5

3

= 2jn

-1L = 1 fb

[GeV]effm200 400 600 800 1000 1200 1400

BS+

B

0.5

1

1.5

2

2.5

3 [GeV]effm200 400 600 800 1000 1200 1400

[1/

100

GeV

]ef

fdmdN

-210

-1101

10

210

310

410

510 signalleptonictt

hadronicttW+jetsZ+jetsQCD

[GeV]effm200 400 600 800 1000 1200 1400

[1/

100

GeV

]ef

fdmdN

-210

-1101

10

210

310

410

510

[GeV]effm200 400 600 800 1000 1200 1400

BS+

B

0.5

1

1.5

2

2.5

3

= 3jn

-1L = 1 fb

[GeV]effm200 400 600 800 1000 1200 1400

BS+

B

0.5

1

1.5

2

2.5

3

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

New physics






– multijet studies establishing meff

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

New physics






– multijet studies establishing meff

Mass vs color charge

– now, significance as function of njets

– representing new physics color charge [gluino does not decay via gluon]

– exclusive 2D likelihood including all information

2 3 4 5 6nj0.2

0.4

0.6

0.8

1.

1.2

1.4

1.6meffHTeVL

SPS1a, 1 inverse fb

0Σ

6Σ

2 3 4 5 6nj0.2

0.4

0.6

0.8

1.

1.2

1.4

1.6meffHTeVL

SPS1a, squarks+squarks, 1 inv. fb

0Σ

4Σ

2 3 4 5 6nj0.2

0.4

0.6

0.8

1.

1.2

1.4

1.6meffHTeVL

SPS1a, squarks+gluino, 1 inv. fb

0Σ

3.2Σ

2 3 4 5 6nj0.2

0.4

0.6

0.8

1.

1.2

1.4

1.6meffHTeVL

SPS1a, gluino+gluino, 1 inv. fb

0Σ

0.6Σ

Jet Scaling

Tilman Plehn

Counting jets

Poisson

Staircase

Interpolating

Jet veto

Higgs couplings

New physics

Exclusive jet counting

staircase (non-abelian) vs Poisson (ordered)

– start by measuring dσ/dnjets

– test both regimes in photon+jets

– test ME-PS merging [CKKW/MLM, unchanged by NLO]

– key to jet vetos

– key to coupling measurements [SFitter+friends]

Understanding Jet Scaling and Jet Vetos in Higgs SearchesE Gerwick, TP, S SchumannPRL 108 (2012)

Establishing Jet Scaling Patterns with a PhotonC Englert, TP, P Schichtel, S SchumannarXiv:1108.5473, JHEP in print

Jets plus Missing Energy with an AutofocusC Englert, TP, P Schichtel, S SchumannPRD83 (2011)

Much of this work was funded by the BMBF Theorie-Verbund which is ideal for hard and relevant LHC work

From Jet Scaling to Jet Vetos - thphys.uni-heidelberg.deplehn/includes/talks/2012/atlas_12.pdf ·...

Documents

Transcript of From Jet Scaling to Jet Vetos - thphys.uni-heidelberg.deplehn/includes/talks/2012/atlas_12.pdf ·...