Further Parameters and Particle Data

The PYUPDA routine is the main tool for updating particle data tables. The following common blocks are maybe of a more peripheral interest, with the exception of the MDCY array, which allows a selective inhibiting of particle decays, and setting masses of not yet discovered particles, such as PMAS(25,1), the (Standard Model) Higgs mass.

Purpose:

to give you the ability to update particle data, or to keep several versions of modified particle data for special purposes (e.g. bottom studies).

MUPDA :

gives the type of action to be taken.

= 1 :

write a table of particle data, that you then can edit at leisure. For ordinary listing of decay data, PYLIST(12) should be used, but that listing could not be read back in by the program.
For each compressed flavour code KC = 1-500, one line is written containing the corresponding uncompressed KF flavour code (1X,I9) in KCHG(KC,4), the particle and antiparticle names (2X,A16,2X,A16) in CHAF, the electric (I3), colour charge (I3) and particle/antiparticle distinction (I3) codes in KCHG, the mass (F12.5), the mass width (F12.5), maximum broadening (F12.5) and average proper lifetime (1P,E13.5) in PMAS, the resonance width treatment (I3) in MWID and the on/off decay switch (I3) in MDCY(KC,1).
After a KC line follows one line for each possible decay channel, containing the MDME codes (10X,2I5), the branching ratio (F12.6) in BRAT, and the KFDP codes for the decay products (5I10), with trailing 0's if the number of decay products is smaller than 5.

= 2 :

read in particle data, as written with =1 and thereafter edited by you, and use this data subsequently in the current run. This also means e.g. the mapping between the full KF and compressed KC flavour codes. Reading is done with fixed format, which means that you have to preserve the format codes described for =1 during the editing. A number of checks will be made to see if input looks reasonable, with warnings if not. If some decay channel is said not to conserve charge, it should be taken seriously. Warnings that decay is kinematically unallowed need not be as serious, since that particular decay mode may not be switched on unless the particle mass is increased.

= 3 :

read in particle data, like option 2, but use it as a complement to rather than a replacement of existing data. The input file should therefore only contain new particles and particles with changed data. New particles are added to the bottom of the KC and decay channel tables. Changed particles retain their KC codes and hence the position of particle data, but their old decay channels are removed, this space is recuperated, and new decay channels are added at the end. Thus also the decay channel numbers of unchanged particles are affected.

= 4 :

write current particle data as data lines, which can be edited into BLOCK DATA PYDATA for a permanent replacement of the particle data. This option is intended for the program author only, not for you.

LFN :

the file number which the data should be written to or read from. You must see to it that this file is properly opened for read or write (since the definition of file names is platform dependent).

Purpose:

to give access to a number of flavour treatment constants or parameters and particle/parton data. Particle data is stored by compressed code KC rather than by the full KF code. You are reminded that the way to know the KC value is to use the PYCOMP function, i.e. KC = PYCOMP(KF).

KCHG(KC,1) :

three times particle/parton charge for compressed code KC.

KCHG(KC,2) :

colour information for compressed code KC.

= 0 :

colour-singlet particle.

= 1 :

quark or antidiquark.

= -1 :

antiquark or diquark.

= 2 :

gluon.

KCHG(KC,3) :

particle/antiparticle distinction for compressed code KC.

= 0 :

the particle is its own antiparticle.

= 1 :

a nonidentical antiparticle exists.

KCHG(KC,4) :

equals the uncompressed particle code KF (always with a positive sign). This gives the inverse mapping of what is provided by the PYCOMP routine.

PMAS(KC,1) :

particle/parton mass $m$ (in GeV) for compressed code KC.

PMAS(KC,2) :

the total width $\Gamma$ (in GeV) of an assumed symmetric Breit-Wigner mass shape for compressed particle code KC.

PMAS(KC,3) :

the maximum deviation (in GeV) from the PMAS(KC,1) value at which the Breit-Wigner shape above is truncated. Is used in ordinary particle decays, but not in the resonance treatment; cf. the CKIN variables.

PMAS(KC,4) :

the average lifetime $\tau$ for compressed particle code KC, with $c\tau$ in mm, i.e. $\tau$ in units of about $3.33 \times 10^{-12}$ s.

PARF(1) - PARF(60) :

give a parameterization of the $\d\overline{\mathrm{d}}$- $\u\overline{\mathrm{u}}$- $\mathrm{s}\overline{\mathrm{s}}$ flavour mixing in production of flavour-diagonal mesons. Numbers are stored in groups of 10, for the six multiplets pseudoscalar, vector, axial vector ($S = 0$), scalar, axial vector ($S = 1$) and tensor, in this order; see section . Within each group, the first two numbers determine the fate of a $\d\overline{\mathrm{d}}$ flavour state, the second two that of a $\u\overline{\mathrm{u}}$ one, the next two that of an $\mathrm{s}\overline{\mathrm{s}}$ one, while the last four are unused. Call the numbers of a pair $p_1$ and $p_2$. Then the probability to produce the state with smallest KF code is $1-p_1$, the probability for the middle one is $p_1 - p_2$ and the probability for the one with largest code is $p_2$, i.e. $p_1$ is the probability to produce either of the two `heavier' ones.

PARF(61) - PARF(80) :

give flavour SU(6) weights for the production of a spin 1/2 or spin 3/2 baryon from a given diquark-quark combination. Should not be changed.

PARF(91) - PARF(96) :

(D = 0.0099, 0.0056, 0.199, 1.23, 4.17, 165 GeV) default nominal quark masses, used to give the starting value for running masses calculated in PYMRUN.

PARF(101) - PARF(105) :

contain $\d$, $\u$, $\mathrm{s}$, $\c$ and $\b$ constituent masses, in the past used in mass formulae for undiscovered hadrons, and should not be changed.

PARF(111), PARF(112) :

(D=0.0, 0.11 GeV) constant terms in the mass formulae for heavy mesons and baryons, respectively (with diquark getting 2/3 of baryon).

PARF(113), PARF(114) :

(D=0.16, 0.048 GeV) factors which, together with Clebsch-Gordan coefficients and quark constituent masses, determine the mass splitting due to spin-spin interactions for heavy mesons and baryons, respectively. The latter factor is also used for the splitting between spin 0 and spin 1 diquarks.

PARF(115) - PARF(118) :

(D=0.50, 0.45, 0.55, 0.60 GeV), constant mass terms, added to the constituent masses, to get the mass of heavy mesons with orbital angular momentum $L = 1$. The four numbers are for pseudovector mesons with quark spin 0, and for scalar, pseudovector and tensor mesons with quark spin 1, respectively.

PARF(121), PARF(122) :

(D=0.1, 0.2 GeV) constant terms, which are subtracted for quark and diquark masses, respectively, in defining the allowed phase space in particle decays into partons (e.g. $\mathrm{B}^0 \to \overline{\mathrm{c}}\d\u\overline{\mathrm{d}}$).

PARF(201) - PARF(1960) :

(D=1760*0) relative probabilities for flavour production in the MSTJ(15)=1 option; to be defined by you before any PYTHIA calls. (The standard meaning is changed in the advanced baryon popcorn code, described in subsection , where many of the PARF numbers are used for other purposes.)
The index in PARF is of the compressed form
$120 + 80 \times$KTAB1$+ 25 \times$KTABS$+$KTAB3.
Here KTAB1 is the old flavour, fixed by preceding fragmentation history, while KTAB3 is the new flavour, to be selected according to the relevant relative probabilities (except for the very last particle, produced when joining two jets, where both KTAB1 and KTAB3 are known). Only the most frequently appearing quarks/diquarks are defined, according to the code $1 = \d$, $2 = \u$, $3 = \mathrm{s}$, $4 = \c$, $5 = \b$, $6 = \t$ (obsolete!), $7 = \d\d _1$, $8 = \u\d _0$, $9 = \u\d _1$, $10 = \u\u _1$, $11 = \mathrm{s}\d _0$, $12 = \mathrm{s}\d _1$, $13 = \mathrm{s}\u _0$, $14 = \mathrm{s}\u _1$, $15 = \mathrm{s}\mathrm{s}_1$, $16 = \c\d _0$, $17 = \c\d _1$, $18 = \c\u _0$, $19 = \c\u _1$, $20 = \c\mathrm{s}_0$, $21 = \c\mathrm{s}_1$, $22 = \c\c _1$. These are thus the only possibilities for the new flavour to be produced; for an occasional old flavour not on this list, the ordinary relative flavour production probabilities will be used.
Given the initial and final flavour, the intermediate hadron that is produced is almost fixed. (Initial and final diquark here corresponds to `popcorn' production of mesons intermediate between a baryon and an antibaryon). The additional index KTABS gives the spin type of this hadron, with
0 = pseudoscalar meson or $\Lambda$-like spin 1/2 baryon,
1 = vector meson or $\Sigma$-like spin 1/2 baryon,
2 = tensor meson or spin 3/2 baryon.
(Some meson multiplets, not frequently produced, are not accessible by this parameterization.)
Note that some combinations of KTAB1, KTAB3 and KTABS do not correspond to a physical particle (a $\Lambda$-like baryon must contain three different quark flavours, a $\Sigma$-like one at least two), and that you must see to it that the corresponding PARF entries are vanishing. One additional complication exist when KTAB3 and KTAB1 denote the same flavour content (normally KTAB3$=$KTAB1, but for diquarks the spin freedom may give KTAB3$=$KTAB1$\pm 1$): then a flavour neutral meson is to be produced, and here $\d\overline{\mathrm{d}}$, $\u\overline{\mathrm{u}}$ and $\mathrm{s}\overline{\mathrm{s}}$ states mix (heavier flavour states do not, and these are therefore no problem). For these cases the ordinary KTAB3 value gives the total probability to produce either of the mesons possible, while KTAB3$=$23 gives the relative probability to produce the lightest meson state ($\pi^0$, $\rho^0$, $\mathrm{a}_2^0$), KTAB3$=$24 relative probability for the middle meson ($\eta$, $\omega$, $\mathrm{f}_2^0$), and KTAB3 = 25 relative probability for the heaviest one ($\eta'$, $\phi$, $f'^0_2$). Note that, for simplicity, these relative probabilities are assumed the same whether initial and final diquark have the same spin or not; the total probability may well be assumed different, however.
As a general comment, the sum of PARF values for a given KTAB1 need not be normalized to unity, but rather the program will find the sum of relevant weights and normalize to that. The same goes for the KTAB3$=$23-25 weights. This makes it straightforward to use one common setup of PARF values and still switch between different MSTJ(12) baryon production modes, with the exception of the advanced popcorn scenarios.

VCKM(I,J) :

squared matrix elements of the Cabibbo-Kobayashi-Maskawa flavour mixing matrix.

I :

up type generation index, i.e. $1 = \u$, $2 = \c$, $3 = \t$ and $4 = \t '$.

J :

down type generation index, i.e. $1 = \d$, $2 = \mathrm{s}$, $3 = \b$ and $4 = \b '$.

Purpose:

to give access to particle decay data and parameters. In particular, the MDCY(KC,1) variables may be used to switch on or off the decay of a given particle species, and the MDME(IDC,1) ones to switch on or off an individual decay channel of a particle. For quarks, leptons and gauge bosons, a number of decay channels are included that are not allowed for on-mass-shell particles, see MDME(IDC,2)=102. These channels are not directly used to perform decays, but rather to denote allowed couplings in a more general sense, and to switch on or off such couplings, as described elsewhere. Particle data is stored by compressed code KC rather than by the full KF code. You are reminded that the way to know the KC value is to use the PYCOMP function, i.e. KC = PYCOMP(KF).

MDCY(KC,1) :

switch to tell whether a particle with compressed code KC may be allowed to decay or not.

= 0 :

the particle is not allowed to decay.

= 1 :

the particle is allowed to decay (if decay information is defined below for the particle).

Warning:

these values may be overwritten for resonances in a PYINIT call, based on the MSTP(41) option you have selected. If you want to allow resonance decays in general but switch off the decay of one particular resonance, this is therefore better done after the PYINIT call.

MDCY(KC,2) :

gives the entry point into the decay channel table for compressed particle code KC. Is 0 if no decay channels have been defined.

MDCY(KC,3) :

gives the total number of decay channels defined for compressed particle code KC, independently of whether they have been assigned a non-vanishing branching ratio or not. Thus the decay channels are found in positions MDCY(KC,2) to MDCY(KC,2)+MDCY(KC,3)-1.

MDME(IDC,1) :

on/off switch for individual decay channel IDC. In addition, a channel may be left selectively open; this has some special applications in the event generation machinery. Effective branching ratios are automatically recalculated for the decay channels left open. Also process cross sections are affected; see section . If a particle is allowed to decay by the MDCY(KC,1) value, at least one channel must be left open by you. A list of decay channels with current IDC numbers may be obtained with PYLIST(12).

= -1 :

this is a non-Standard Model decay mode, which by default is assumed not to exist. Normally, this option is used for decays involving fourth generation or $\H ^{\pm}$ particles.

= 0 :

channel is switched off.

= 1 :

channel is switched on.

= 2 :

channel is switched on for a particle but off for an antiparticle. It is also on for a particle which is its own antiparticle, i.e. here it means the same as =1.

= 3 :

channel is switched on for an antiparticle but off for a particle. It is off for a particle which is its own antiparticle.

= 4 :

in the production of a pair of equal or charge conjugate resonances in PYTHIA, say $\mathrm{h}^0 \to \mathrm{W}^+ \mathrm{W}^-$, either one of the resonances is allowed to decay according to this group of channels, but not both. If the two particles of the pair are different, the channel is on. For ordinary particles, not resonances, this option only means that the channel is switched off.

= 5 :

as =4, but an independent group of channels, such that in a pair of equal or charge conjugate resonances the decay of either resonance may be specified independently. If the two particles in the pair are different, the channel is off. For ordinary particles, not resonances, this option only means that the channel is switched off.

Warning:

the two values -1 and 0 may look similar, but in fact are quite different. In neither case the channel so set is generated, but in the latter case the channel still contributes to the total width of a resonance, and thus affects both simulated line shape and the generated cross section when PYTHIA is run. The value 0 is appropriate to a channel we assume exists, even if we are not currently simulating it, while -1 should be used for channels we believe do not exist. In particular, you are warned unwittingly to set fourth generation channels 0 (rather than -1), since by now the support for a fourth generation is small.

Remark:

all the options above may be freely mixed. The difference, for those cases where both make sense, between using values 2 and 3 and using 4 and 5 is that the latter automatically include charge conjugate states, e.g. $\mathrm{h}^0 \to \mathrm{W}^+ \mathrm{W}^- \to \mathrm{e}^+ \nu_e \d\overline{\mathrm{u}}$ or $\overline{\mathrm{d}}\u\mathrm{e}^- \overline{\nu}_e$, but the former only one of them. In calculations of the joint branching ratio, this makes a factor 2 difference.

Example:

to illustrate the above options, consider the case of a $\mathrm{W}^+ \mathrm{W}^-$ pair. One might then set the following combination of switches for the $\mathrm{W}$:

channel	value	comment
$\u\overline{\mathrm{d}}$	1	allowed for $\mathrm{W}^+$ and $\mathrm{W}^-$ in any combination,
$\u\overline{\mathrm{s}}$	0	never produced but contributes to $\mathrm{W}$ width,
$\c\overline{\mathrm{d}}$	2	allowed for $\mathrm{W}^+$ only,
$\c\overline{\mathrm{s}}$	3	allowed for $\mathrm{W}^-$ only, i.e. properly $\mathrm{W}^- \to \overline{\mathrm{c}}\mathrm{s}$,
$\t\overline{\mathrm{b}}$	0	never produced but contributes to $\mathrm{W}$ width
		if the channel is kinematically allowed,
$\nu_{\mathrm{e}}\mathrm{e}^+$	4	allowed for one of $\mathrm{W}^+$ or $\mathrm{W}^-$, but not both,
$\nu_{\mu}\mu^+$	4	allowed for one of $\mathrm{W}^+$ or $\mathrm{W}^-$, but not both,
		and not in combination with $\nu_{\mathrm{e}}\mathrm{e}^+$,
$\nu_{\tau}\tau^+$	5	allowed for the other $\mathrm{W}$, but not both,
$\nu_{\chi}\chi^-$	$-1$	not produced and does not contribute to $\mathrm{W}$ width.

A $\mathrm{W}^+ \mathrm{W}^-$ final state $\u\overline{\mathrm{d}}+ \overline{\mathrm{c}}\mathrm{s}$ is allowed, but not its charge conjugate $\overline{\mathrm{u}}\d + \c\overline{\mathrm{s}}$, since the latter decay mode is not allowed for a $\mathrm{W}^+$. The combination $\nu_{\mathrm{e}}\mathrm{e}^+ + \bar{\nu}_{\tau}\tau^-$ is allowed, since the two channels belong to different groups, but not $\nu_{\mathrm{e}}\mathrm{e}^+ + \bar{\nu}_{\mu}\mu^-$, where both belong to the same. Both $\u\overline{\mathrm{d}}+ \bar{\nu}_{\tau}\tau^-$ and $\overline{\mathrm{u}}\d + \nu_{\tau}\tau^+$ are allowed, since there is no clash. The full rulebook, for this case, is given by eq. (). A term $r_i^2$ means channel $i$ is allowed for $\mathrm{W}^+$ and $\mathrm{W}^-$ simultaneously, a term $r_i r_j$ that channels $i$ and $j$ may be combined, and a term $2 r_i r_j$ that channels $i$ and $j$ may be combined two ways, i.e. that also a charge conjugate combination is allowed.

MDME(IDC,2) :

information on special matrix-element treatment for decay channel IDC. Is mainly intended for the normal-particle machinery in PYDECY, so many of the codes are superfluous in the more sophisticated resonance decay treatment by PYRESD, see section . In addition to the outline below, special rules apply for the order in which decay products should be given, so that matrix elements and colour flow is properly treated. One such example is the weak matrix elements, which only will be correct if decay products are given in the right order. The program does not police this, so if you introduce channels of your own and use these codes, you should be guided by the existing particle data.

= 0 :

no special matrix-element treatment; partons and particles are copied directly to the event record, with momentum distributed according to phase space.

= 1 :

$\omega$ and $\phi$ decays into three pions, eq. ().

= 2 :

$\pi^0$ or $\eta$ Dalitz decay to $\gamma \mathrm{e}^+ \mathrm{e}^-$, eq. ().

= 3 :

used for vector meson decays into two pseudoscalars, to signal non-isotropic decay angle according to eq. (), where relevant.

= 4 :

decay of a spin 1 onium resonance to three gluons or to a photon and two gluons, eq. (). The gluons may subsequently develop a shower if MSTJ(23)=1.

= 11 :

phase-space production of hadrons from the quarks available.

= 12 :

as =11, but for onia resonances, with the option of modifying the multiplicity distribution separately.

= 13 :

as =11, but at least three hadrons to be produced (useful when the two-body decays are given explicitly).

= 14 :

as =11, but at least four hadrons to be produced.

= 15 :

as =11, but at least five hadrons to be produced.

= 22 - 30 :

phase-space production of hadrons from the quarks available, with the multiplicity fixed to be MDME(IDC,2)-20, i.e. 2-10.

= 31 :

two or more quarks and particles are distributed according to phase space. If three or more products, the last product is a spectator quark, i.e. sitting at rest with respect to the decaying hadron.

= 32 :

a $\mathrm{q}\overline{\mathrm{q}}$ or $\mathrm{g}\mathrm{g}$ pair, distributed according to phase space (in angle), and allowed to develop a shower if MSTJ(23)=1.

= 33 :

a triplet $\mathrm{q}X \overline{\mathrm{q}}$, where $X$ is either a gluon or a colour-singlet particle; the final particle ( $\overline{\mathrm{q}}$) is assumed to sit at rest with respect to the decaying hadron, and the two first particles ($\mathrm{q}$ and $X$) are allowed to develop a shower if MSTJ(23)=1. Nowadays superfluous.

= 41 :

weak decay, where particles are distributed according to phase space, multiplied by a factor from the expected shape of the momentum spectrum of the direct product of the weak decay (the $\nu_{\tau}$ in $\tau$ decay).

= 42 :

weak decay matrix element for quarks and leptons. Products may be given either in terms of quarks or hadrons, or leptons for some channels. If the spectator system is given in terms of quarks, it is assumed to collapse into one particle from the onset. If the virtual $\mathrm{W}$ decays into quarks, these quarks are converted to particles, according to phase space in the $\mathrm{W}$ rest frame, as in =11. Is intended for $\tau$, charm and bottom.

= 43 :

as =42, but if the $\mathrm{W}$ decays into quarks, these will either appear as jets or, for small masses, collapse into a one- or two-body system. Nowadays superfluous.

= 44 :

weak decay matrix element for quarks and leptons, where the spectator system may collapse into one particle for a small invariant mass. If the first two decay products are a $\mathrm{q}\overline{\mathrm{q}}'$ pair, they may develop a parton shower if MSTJ(23)=1. Was intended for top and beyond, but is nowadays superfluous.

= 46 :

$\mathrm{W}$ (KF = 89) decay into $\mathrm{q}\overline{\mathrm{q}}'$ or $\ell \nu_{\ell}$ according to relative probabilities given by couplings (as stored in the BRAT vector) times a dynamical phase-space factor given by the current $\mathrm{W}$ mass. In the decay, the correct $V-A$ angular distribution is generated if the $\mathrm{W}$ origin is known (heavy quark or lepton). This is therefore the second step of a decay with MDME=45. A $\mathrm{q}\overline{\mathrm{q}}'$ pair may subsequently develop a shower if MSTJ(23)=1. Was intended for top and beyond, but is nowadays superfluous.

= 48 :

as =42, but require at least three decay products.

= 50 :

(default behaviour, also obtained for any other code value apart from the ones listed below) do not include any special threshold factors. That is, a decay channel is left open even if the sum of daughter nominal masses is above the mother actual mass, which is possible if at least one of the daughters can be pushed off the mass shell. Is intended for decay treatment in PYRESD with PYWIDT calls, and has no special meaning for ordinary PYDECY calls.

= 51 :

a step threshold, i.e. a channel is switched off when the sum of daughter nominal masses is above the mother actual mass. Is intended for decay treatment in PYRESD with PYWIDT calls, and has no special meaning for ordinary PYDECY calls.

= 52 :

a $\beta$-factor threshold, i.e. $\sqrt{ (1-m_1^2/m^2-m_2^2/m^2)^2 - 4 m_1^2 m_2^2/m^4}$, assuming that the values stored in PMAS(KC,2) and BRAT(IDC) did not include any threshold effects at all. Is intended for decay treatment in PYRESD with PYWIDT calls, and has no special meaning for ordinary PYDECY calls.

= 53 :

as =52, but assuming that PMAS(KC,2) and BRAT(IDC) did include the threshold effects, so that the weight should be the ratio of the $\beta$ value at the actual mass to that at the nominal one. Is intended for decay treatment in PYRESD with PYWIDT calls, and has no special meaning for ordinary PYDECY calls.

= 101 :

this is not a proper decay channel, but only to be considered as a continuation line for the decay product listing of the immediately preceding channel. Since the KFDP array can contain five decay products per channel, with this code it is possible to define channels with up to ten decay products. It is not allowed to have several continuation lines after each other.

= 102 :

this is not a proper decay channel for a decaying particle on the mass shell (or nearly so), and is therefore assigned branching ratio 0. For a particle off the mass shell, this decay mode is allowed, however. By including this channel among the others, the switches MDME(IDC,1) may be used to allow or forbid these channels in hard processes, with cross sections to be calculated separately. As an example, $\gamma \to \u\overline{\mathrm{u}}$ is not possible for a massless photon, but is an allowed channel in $\mathrm{e}^+\mathrm{e}^-$ annihilation.

BRAT(IDC) :

give branching ratios for the different decay channels. In principle, the sum of branching ratios for a given particle should be unity. Since the program anyway has to calculate the sum of branching ratios left open by the MDME(IDC,1) values and normalize to that, you need not explicitly ensure this normalization, however. (Warnings are printed in PYUPDA(2) or PYUPDA(3) calls if the sum is not unity, but this is entirely intended as a help for finding user mistypings.) For decay channels with MDME(IDC,2)$> 80$ the BRAT values are dummy.

KFDP(IDC,J) :

contain the decay products in the different channels, with five positions J= 1-5 reserved for each channel IDC. The decay products are given following the standard KF code for partons and particles, with 0 for trailing empty positions. Note that the MDME(IDC+1,2)=101 option allows you to double the maximum number of decay product in a given channel from 5 to 10, with the five latter products stored KFDP(IDC+1,J).

Purpose:

to give access to character type variables.

CHAF(KC,1) :

particle name according to KC code.

CHAF(KC,2) :

antiparticle name according to KC code when an antiparticle exists, else blank.