Dijet Azimuthal Decorrelations at Central Rapidities
We present a measurement of the normalized dijet cross section
as a function of the azimuthal angle between the two leading jets
(Delta Phi)
in the central region of rapidity |yjet|<0.5.
Jets are reconstructed by an iterative cone algorithm
with cone radius Rcone = 0.7.
The results are compared to pQCD calculations in fixed order alpha_s
(LO and NLO)
and to the predictions of the event generators PYTHIA and HERWIG.
The paper is available as
hep-ex0409040
(submitted to Phys. Rev. Lett. on September 16, 2004)
The figures from the paper as .eps files:
-
Fig. 1
The Delta Phi distributions in four regions of pTmax.
Data and predictions with pTmax > 100 GeV are scaled
by successive factors of 20 for purposes of presentation.
The solid (dashed) lines show the NLO (LO) pQCD predictions.
- Fig. 2
Ratios of data to the NLO pQCD calculation
for different regions of pTmax.
Theoretical uncertainties due to variation of the
renormalization and factorization scales are shown as the shaded regions;
the uncertainty due to the PDFs is indicated by the solid lines.
The points at large Delta Phi are excluded because the calculation has
non-physical behavior near the divergence at Pi.
- Fig. 3
The Delta Phi distributions in different pTmax ranges.
Results from HERWIG and PYTHIA are overlaid on the data.
Data and predictions with pTmax > 100 GeV are scaled by
successive factors of 20 for purposes of presentation.
Additional figures for presentations:
-
dphi_allmc.eps
Comparison of the data with all three theory predictions:
- NLO (predictions from NLOJET++) same curves as in the publication Fig.1
- HERWIG (default) predictions - same curves as in the publication Fig.3
- PYTHIA retuned:
In the publication it is demonstrated that the PYTHIA predictions
are very sensitive to the parameter PARP(67) which is
related to the maximum pT in the ISR parton shower.
PARP(67) was varied in the range from 1.0 - 4.0.
In this figure we show the results for
a re-tuned value of PARP(67)=2.5 for which PYTHIA gives
the best description of the data.
The following three figures have a closer look into the peak region
(near Pi).
Although HERWIG is slightly too narrowly peaked, it
gives a reasonable description of the data.
PYTHIA (default) is much too narrowly peaked near Delta Phi = Pi.
To investigate the possibilities for tuning PYTHIA we compare
various parameter variations to the data.
It is visible that none of these variations helps to get
agreement with PYTHIA and the data
in the region at large Delta Phi (near Pi).
- dphi03_mc_peak.eps
The data are compared with HERWIG (default), PYTHIA (default), and with
PYTHIA with increased ISR (increased pTmax in the ISR shower:
PARP(67)=4.0 (D=1.0) - these are the same settings as in Fig.3
in the publication)
- dphi03_pyth_isr.eps
Attempts to tune ISR related parameters in PYTHIA to the data.
The following parameters have been varied:
-
PARP(67)=4.0 (D=1.0) increase pTmax in ISR shower.
(at large Delta Phi: small effects only)
-
PARP(64)=0.5 (D=1.0) increase alpha_s in ISR shower by
reducing the renormalization scale factor (effect is negligible)
-
PARP(91)=4.0 (D=1.0) PARP(93)=8.0 (D=5.0)
increase the primordial kT and the upper cut-off for its gaussian
distribution (very small effect)
- dphi03_pyth_fsr.eps
The data are compared to HERWIG (default), PYTHIA (default), and
to PYTHIA with increased pTmax in the FSR shower:
PARP(71)=8.0 (D=4.0) (very small effect only at low pT)
back to the D0 Run II QCD results
Last updated September 21 , 2004 - wobisch@fnal.gov, dalton@fnal.gov