Diapositiva 1

1
Study on search of a SM Higgs (120GeV) produced
via VBF and decaying in two hadronic taus
V.Cavasinni, F.Sarri, I.Vivarelli
2
OUTLINE

• Motivation
• Signal and backgrounds
• Preliminary results of analysis
• Trigger issues

3
Cross sections and branching ratios for the Higgs
boson
sVBF H120 4.35 pb (value from VV2H with CTEQ5L
for PDF )
BR(H 120?tt) 0.0679 (value from Hdecay)
4
Tau channel for low Higgs mass
Previous study (hep-ph/0402254) tt ? lnln , tt
? ln hn . In 30 fb-1 about 32 signal events and
22 background events
5
Mass reconstruction in collinear approximation
For tt?had had, if x1,2 are the visible energy
fraction of the two taus
6
VBF signature

• 2 high pT forward jets
• - depleted jet activity in the central region

Rapidity distribution for the tagged forward jets
for signal and for ttbar background.
7
HVBF ? tt ?hhnn
Tau decays 65 of times in hadrons The BR(tt?hh)
is 42.25

• Main features of this event
• 4 jets
• 2 forward jets
• 2 central tau jets
• Missing transverse energy due to neutrinos

Main background QCD
Other backgrounds ttbar, Z/?jets
8
Treatment of QCD backgrounds

• Due to the high cross section, impossible to
  generate all the needed events for the QCD, even
  with ATLFAST.
• After all the analysis cuts, very poor background
  statistics.
• To solve the problem when requiring tau jets,
  keep all the background events and weight them
  with the rejection factor for tau jets.
• The performance of the tau identification
  algorithm in FULLSIM depending on pT and h is
  parameterized and put directly into ATLFAST.
  (M.Heldmann work)

9
Tau id in full simulation - tauRec

• Reconstruct tau-candidate
• Start from Calorimetric Clusters (default).
• Associate tracks to the candidate.
• Calibrate candidate.
• Build the set of pT dependent variables for
  tau-identification (REM, FISO, Ntrack, Strip
  Width, Nstrip, Charge, Impact parameter,
  ET/pT(1sttrack)) and then calculates Likelihood.
• Apply a set of the basic cuts for
  tau-identification.

M.Heldmann
L gt 4
10
(No Transcript)
11
Typical cuts of the analysis (hep-ph/0402254)

• at least 4 jets in the event
• 2 tau tagged jets with pTs over given
  thresholds
• 2 candidate forward jets with pTs above given
  thresholds
• negative eta product for the forwards and Dh gt
  4
• taus between the 2 forwards
• 0 lt x1(2) lt 1 and taus not back to back
• invariant mass for the forwards gt 700 GeV
• missing transverse energy above a threshold
• no jet with pT gt 20 GeV between the 2 forward,
  apart from the taus
• cut on mass window.

Two different sets of cuts used for the analysis
12
Soft cuts

• at least 4 jets
• 2 tau tagged jets with pT above 35 and 30GeV
• missing transverse energy gt 45 GeV
• 2 forward candidates with pT above 50 and 30
  GeV
• negative eta product for the forwards and Dh gt 4
  
• taus between the forwards
• 0 lt x1(2) lt 1 and taus not back to back (Df lt
  3)
• cut on mass window 104 GeV lt mtt lt 150 GeV.
• dedicated trigger menu

13
Soft cuts
cut Signal efficiency Events 30 fb-1 (300 fb-1) QCD rejection Events 30 fb-1 (300 fb-1)
4jet (2 tau), Etmiss gt 45 GeV 4.4 3.4 10-8
Pt tau 2.5 1.21 10-10
Pt forward 1.9 4.81 10-11
h forward 0.94 3.56 10-12
t cuts 0.65 24 (241) 5.4 10-13 155 (1551)
Mass window 0.55 21 (206) 8.1 10-14 23 (231)
using the trigger thresholds for low luminosity
period
14
Soft cuts
15
Hard cuts

• at least 4 jets
• 2 tagged tau jets with pt over 45 and 35 GeV
• missing transverse energy over 45 GeV
• 2 forward candidates with pt above 60 and 40 GeV
• negative product of forward jet eta and Dh gt 4
• taus between 2 forward
• 0 lt x1(2) lt 1 and taus not back to back (Df lt
  3)
• forward jet invariant mass gt 700 GeV
• no jet with pt gt 30 GeV between the forwards,
  apart from the taus
• cut on mass window 104 GeV lt mtt lt 150 GeV.

16
Hard cuts
cut Signal efficiency Events 30 fb-1 (300 fb-1) QCD rejection Events 30 fb-1 (300 fb-1)
4jet (2 tau), Etmiss gt 45 GeV 4.4 3.4 10-8
Pt tau 1.6 3.96 10-11
Pt forward 0.97 1.22 10-11
h forward 0.49 7.8 10-13
t cuts 0.33 1.22 10-13
Mjj 0.3 9.5 10-14
Jet veto 0.26 10 (99) 5.1 10-14 14 (144)
Mass window 0.24 9 (89) 5.3 10-15 2 (15)
using the trigger thresholds for low luminosity
period
17
Hard cuts
18
ttbar background
Both Ws dacay in tn 1
In 30 fb-1 about 17106 events. Almost 40 10 6
generated events.
On ttbar events the rejection from
parameterization is worse by about a factor 2.
19
Soft cuts
cut Signal efficiency Events 30 fb-1 (300 fb-1) ttbar rejection Events 30 fb-1 (300 fb-1)
4jet (2 tau), Etmiss gt 45 GeV 4.4 (7.3/2) 10-4
Pt tau 2.5 (1.78/2) 10-4
Pt forward 1.9 (1.46/2) 10-4
h forward 0.94 (5.09/2) 10-6
t cuts 0.65 24 (241) (1.02/2) 10-6 64 (338)
Mass window 0.55 21 (206) (5.07/2) 10-8 4 (32)
using the trigger thresholds for low luminosity
period
20
Hard cuts
cut Signal efficiency Events 30 fb-1 (300 fb-1) ttbar rejection Events 30 fb-1 (300 fb-1)
4jet (2 tau), Etmiss gt 45 GeV 4.4 (7.3/2) 10-4
Pt tau 1.6 (1.1/2) 10-4
Pt forward 0.97 (6.8/2) 10-5
h forward 0.49 (2.4/2) 10-6
t cuts 0.33 (0.56/2) 10-6
Mjj 0.3 (4.05/2) 10-7
Jet veto 0.26 10 (99) (1.77/2) 10-7 6 (58)
Mass window 0.24 9 (89) 0
using the trigger thresholds for low luminosity
period
21
g/Z jets background

• Important background
• g/Z jets with g/Z ? tt has a s 1742 pb
• 736 pb with two true hadronic taus.
• UNDER STUDY

22
Hadronic Tau Trigger (I) (ATL-COM-DAQ-2003-030)
LVL1 trigger look at 4X4 matrix of calorimetric
towers (DhDf 0.1 x 0.1 each). ET threshold for
the central core (EMHad) and isolation
thresholds between core and 12 external towers
for e.m. and had. calorimeters.
second layer of EM calorimeter
LVL2 trigger look at the shower shape in the
2nd layer of e.m. calorimeter and at the track
multiplicity inside the RoI defined at LVL1. Cut
on the ratio between ET contained in a 3x7 cell
cluster and ET in a 7x7 cell cluster and on track
multiplicity
h
track multiplicity in the RoI
f
23
Hadronic Tau Trigger (II) (ATL-COM-DAQ-2003-030)

• LVL3 (Event Filter)
• look at the complete event.
• By now the variables of the offline algorithms
  are used as an approximation LVL3 trigger five
  variables
• number of reconstructed tracks, within DR 0.3
  of the candidate calorimeter cluster, between 1
  and 3
• cut on isolation fraction, defined as the
  difference between the ET contained in a cone
  size of DR0.2 and 0.1 normalized to the total
  jet ET
• cut on EM jet radius, an energy weighted radius
  calculated only in the e.m. calorimeter
• cut on EM energy fraction, defined as the
  fraction of the total jet energy in the e.m.
  calorimeter
• threshold on the pT of the highest pT track.

24
LVL1 Trigger Rates
Illustrative menu
M.Bosman talk_at_ATLAS Physics Workshop June 2005
25
Inclusive High Level Trigger Event Selection
Selection 2x1033 cm-2s-1 Rates (Hz)
Electron e25i, 2e15i 40
Photon g60i, 2g20i 40
Muon m20i, 2m10 40
Jets j400, 3j165, 4j110 25
Jet ETmiss j70 xE70 20
tau ETmiss t35 xE45 5
B-physics 2m6 with mB /mJ/y 10
Others pre-scales, calibration, 20
Total 200
No safety factors - large uncertainties !
Current global understanding of trigger rates
M.Bosman talk_at_ATLAS Physics Workshop June 2005
26
Trigger efficiency
While the trigger rate depends on the QCD
backgrounds, the trigger efficiency depends on
the physics channel under study. By now, only
studies on Z?tt and A/H?tt?hh. The trigger
impact on the VBF H ? tt ? hh has to be studied.
27
Conclusions
The channel VBF Higgs ? tt ?hh appears to be
promising to improve the statistical significance
in the critical low Higgs mass region. Fast
simulation shows that QCD and ttbar backgrounds
may be rejected at the desired level. Zjet
background still under study. Efficiency of the
trigger has to be studied for this
channel. Future work redo the analysis with
full simulated events and with ME QCD generators.
28
BACKUP SLIDES
29
t id in full simulation - tauRec

• Builds set of variables for t-identification
  they are pT dependant
• Calculate Likelihood from REM, FISO, Ntrack,
  Strip Width, Nstrip, Charge, Impact parameter,
  ET/pT(1sttrack)

REM
FISO
Charge
Ntrack
Nstrip
Impact parameter
Strip Width
ET/pT(1sttrack)
Signal A?tt?, background QCD, 0lt pTlt44 full line
and 134lt pTlt 334 dashed line
30
t id in fast Simulation - Parameterization
Rejection vs. h (jet)
Rejection vs. pT (jet)
31
Parametrisation (R vs. pT)
20
25
30
35
45
40
50
55
60
R falls again with high pT
Rmax and slope depend on e
Rmax300GeV
R grows with pT
32
Parametrisation (R vs. ?)

• d

20
25
30
35
45
40
50
55
60
a lot of structure in R vs. h, but taken to be
flat and than falling like a gauss
R falls for h gt 1.5 (endcap, higher granularity,
lower ID resolution)
At h 1.5 crack region should be excluded
33
Hadronic Tau Trigger

• Previous studies (ATL-DAQ-98-127) used
• ETCore(emh) gt 50
• fr(Core) ETCore(em) / ETRoI(em) gt 0.85
• 1 Ntrk 3
• Now using infrastructure of LVL2 calo algorithm
  (TrigT2Calo)
• Adding AlgTools for Tau calculations at LVL2.
• Variables considered
• ETCore(em), ETCore(h), ETRoI(em), ETRoI(h) in
  regions set in the jobOptions.
• Evaluating offline variables em radius of the
  cluster, width in energy deposition, isolation
  fraction.
• Procedure in developing HLT code
• Use offline athena for programming
• Test the code in the multithreading environment
  with athenaMT.
• Testbed integration.

Pilar Casado, Martine Bosman Carlos Osuna