Here, parameters specific to VINCIA's QED and EW antenna showers are collected, including VINCIA's interleaved treatment of resonance decays. See the main VINCIA antenna shower page for more general parameters that are common to both the QCD and QED/EW showers.
VINCIA contains two alternative implementations of QED/EW shower effects. One is restricted to pure QED but allows for a fully coherent (multipole) treatment of that sector. The other is limited to dipole-style coherence in the photon sector but includes a full set of electroweak (EW) branching kernels with the appropriate (quasi-)collinear limits.
mode
Vincia:EWmode
(default = 2
; minimum = 0
; maximum = 3
)Vincia:QEDmodeMPI
below for the
handling of QED corrections in MPI systems.)
option
0 : No QED or weak showers at all.
option
1 : Dipole QED showers. Partially coherent, based on
maximally screening dipole pairs. Spin-dependent antenna terms can be
switched on/off using Vincia:QEDfullKernels
, see below.
option
2 : Multipole QED showers. Fully coherent but
computationally slower than the dipole approximation. Spin-dependent
antenna terms can be switched on/off using
Vincia:QEDfullKernels
, see below.
option
3 : Dipole QED + Weak showers. All electroweak
branchings are enabled, comprising emissions, splittings, and
self-interactions of W, Z, and H bosons, in addition to QED. Note
that the weak shower operates on helicity eigenstates; this is
particularly relevant for W and Z bosons for which the longitudinal
polarisations have different splitting amplitudes than the transverse
ones. Typically, the hard process does not provide helicity
information for the in- and outgoing legs in Pythia. In this case,
VINCIA will attempt to use its MG5 matrix-element interface and its
internal EW splitting amplitudes to assign such helicities. Note that
this requires Pythia to be configured using the
--with-mg5mes
option (see also the
example program
main404.cc
). If the given hard process is not available
in the linked MG5 library, or if the helicity assignment fails for any
other reason, then the QED dipole shower is used as a fallback.
mode
Vincia:QEDmodeMPI
(default = 1
; minimum = 0
; maximum = 2
)QEDmodeMPI
is forced to be less than or equal to the main
EWmode
switch above so that the treatment of QED
corrections for MPI cannot be more sophisticated than that of the hard
interaction. Also note that there is currently no option to include
weak showers for MPI.
option
0 : No QED showers in MPI systems.
option
1 : Dipole QED showers in MPI systems.
option
2 : Multipole QED showers in MPI systems. This is the
most advanced option, with full coherence, but is somewhat
computationally slower and would normally be overkill for MPI.
mode
Vincia:QEDmodeHadDec
(default = 2
; minimum = 1
; maximum = 2
)HadronLevel:QED = on
, this switch determines whether
QED multipole interference effects (similar to YFS) are taken into
account or not, for QED showers in hadron and tau decays (with spin
dependence according to Vincia:useSpinsQEDHadDec
).
option
1 : Dipole QED showers in hadron and tau decays.
option
2 : Multipole QED showers in hadron and tau decays.
mode
Vincia:nGammaToQuark
(default = 5
; minimum = 0
; maximum = 6
)mode
Vincia:nGammaToLepton
(default = 3
; minimum = 0
; maximum = 3
)flag
Vincia:convertGammaToQuark
(default = on
)flag
Vincia:convertQuarkToGamma
(default = on
)mode
Vincia:alphaEMorder
(default = 1
; minimum = 0
; maximum = 1
)option
0 : zeroth order, i.e. αem is kept
fixed.
option
1 : first order, i.e., one-loop running.
parm
Vincia:alphaEM0
(default = 0.00729735
; minimum = 0.0072973
; maximum = 0.0072974
)parm
Vincia:alphaEMmZ
(default = 0.00781751
; minimum = 0.00780
; maximum = 0.00783
)parm
Vincia:QminChgQ
(default = 0.5
; minimum = 0.1
; maximum = 2.0
)parm
Vincia:QminChgL
(default = 1e-6
; minimum = 1e-10
; maximum = 2.0
)fvec
Vincia:useSpinsQED
(default = {on, on, on}
)on
)
or just their (spin-independent, YFS-style) scalar eikonal terms
(off
), for QED in hard processes, resonance decays, and
MPI. The first entry switches the spin-dependent collinear terms
on/off for spin-1/2 radiators, the second for spin-1 radiators, and
the third for spin-3/2 ones. Note that the spin-3/2 case is not fully
implemented and currently defaults to use the same terms as the
spin-1/2 case.
fvec
Vincia:useSpinsQEDHadDec
(default = {on, on, on}
)Vincia:useSpinsQED
for QED in hadron and tau decays.
parm
Vincia:mMaxGamma
(default = 10.
; minimum = 0.001
; maximum = 5000.0
)mode
Vincia:kineMapEWFinal
(default = 3
; minimum = 1
; maximum = 3
)option
1 : The Ariadne angle.
option
2 : Longitudinal (dipole) map.
option
3 : The Kosower map.
flag
Vincia:doBosonicInterference
(default = on
)mode
Vincia:bwMatchingMode
(default = 2
; minimum = 1
; maximum = 3
)option
1 : Resonance-type branchings in the shower are
disabled. Resonances are instead decayed according to a Breit-Wigner
distribution exclusively.
option
2 : A suppression factor $\frac{Q^4}{(Q^2 +
Q^2_{\mathrm{EW}})^2} is applied to the resonance decay-type
branchings in the shower. Any resonance that does not disappear due to
a shower branching before its Breit-Wigner-sampled off-shellness is
instead decayed according to the Breit-Wigner distribution.
option
3 : No off-shellness is sampled from a
Breit-Wigner. Only the shower is ran without suppression factor. Does
not guarantee that all resonances decay.
parm
Vincia:EWScale
(default = 100.
; minimum = 80.
; maximum = 175.
)Vincia:bwMatchingMode = 2
.
flag
Vincia:EWoverlapVeto
(default = off
)on
in conjunction with Vincia:interleaveResDec =
off
.
parm
Vincia:EWoverlapVetoDeltaR
(default = 0.6
; minimum = 0.1
)flag
Vincia:BWstrongOrdering
(default = off
)parm
Vincia:EWheadroomF
(default = 1.1
; minimum = 1.
; maximum = 2.
)parm
Vincia:EWheadroomI
(default = 3
; minimum = 1.
; maximum = 5.
)As there are such a very large number of electroweak branchings, the technical problem of determining the overestimate function (used for generating trial branchings) has been automated. The overestimate function is first parameterised in terms of four functions; the coefficients of these are then extracted by numerically minimising the difference between the corresponding functional and the branching kernels themselves. The coefficients determined by this procedure are collected in the XML file and are read in during initialisation.