2000 fb ^-1 SM Data Samples at Ecm= 250, 350, 500 & 1000 GeV

Introduction

Stdhep files for 2000 fb ^-1 Standard Model data samples at Ecm= 250 & 500 GeV have been produced and are currently located on SLAC mass storage. Some
Standard Model data samples at Ecm= 350 GeV have also been produced. A complete list of the processes can be found in ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC250/doc/integ_index_0250_01,
ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/ILC350/doc/integ_index_0350_SB2009_nTF_extbunches_01, and ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/integ_index_0500_01 . The five columns are process_id, initial_state,
the variable IDRUPLH, an internal sequence number (1 - 4) and a bit indicating whether or not events were produced for this particular initial state polarization sign combination (0 = events generated, 1= events not generated).
Subsets of these events are available via ftp; please see the table of derived stdhep files.

The events are produced assuming 100% polarization for the initial state electron and positron; different files for the same final state correspond to different polarization sign combinations. The variable IDRUPLH indexes the different final states and polarizations. Assume, for example, that a process has IDRUPLH=14995 . The stdhep file is ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/w14995_01.stdhep and the information about the generation of this file can be found in the directory ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/run_output/w14995/run_01/ .
The log file is ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/run_output/w14995/run_01/whizard.log ,
the whizard input file is ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/run_output/w14995/run_01/whizard.in
and cross section information is in ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/run_output/w14995/run_01/whizard.n3n3n3n3ss_o.out. For Ecm= 350 GeV, the event generation log files can be found in ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/ILC350/doc/run_output .

The WHIZARD Monte Carlo version 1.40 is used for parton generation for all processes except SUSY processes. WHIZARD version 1.51 is used for SUSY processes. For processes with (without) an on-shell Higgs boson in the final state the Higgs mass is assumed to be 120 GeV (2000 GeV). The Makefile and build log files for this implementation of WHIZARD can be found in ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/doc/whizard-v1r4p0 .

Electron/Positron Beam Properties: Beamstrahlung and LINAC Energy Spread

The following lines in whizard.in control the properties of the colliding electron/positron beams:

USER_spectrum_on = T
USER_spectrum_mode = -2

The first line indicates that a user-supplied function is used to simulate the beams. A copy of this function can
be found in ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/whizard/v1r4p0/whizard-v1r4p0/whizard-src/user.f90 for the WHIZARD 1.40 version and in ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/whizard/whizard-1.95/whizard-src/user.f90 for the WHIZARD 1.95 version. This function refers to links with the names ..._linker_000... which can be found in
ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/whizard/guinea-pig/energy_spread.

The absolute value of USER_spectrum_mode determines which energy spectrum is used, with the sign +/- indicating
electron/positron beam, respectively. For the Ecm = 500 GeV SM data sample this absolute value is always 2, and corresponds
to the Guinea-Pig data contained in the directory ftp://ftp-lcd.slac.stanford.edu/ilc/ILC500/StandardModel/guinea-pig/ilc_0500_may05_run05_seed06/
This spectrum represents the default ILC design for Ecm=500 GeV circa August 2005, and includes both incoming LINAC
energy spread and beamstrahlung.

For the Ecm = 250 GeV SM data sample the absolute value of USER_spectrum_mode is 4 for the LOI samples with do_isr = T, and USER_spectrum_mode is 5 for the post-LOI samples with do_isr = F. The Guinea-Pig data for USER_spectrum_mode=5 is in ftp://ftp-lcd.slac.stanford.edu/ilc3/ILC250/LOI_backgrounds/guineapig/postLOI/guineapig/lumionly/raw . USER_spectrum_mode=6 corresponds to SB2009_350_nTF_extbunches. USER_spectrum_mode=7 corresponds to SB2009_500_nTF_extbunches. To summarize, the USER_spectrum_mode numbers, aka the Lumi_linker numbers, are as follows:

Final State Parton Showering and Fragmentation

PYTHIA 6.205 is used for final state QED/QCD parton showering and for the fragmentation of quarks and gluons. Parton showering is performed for all final state fermions with the exception of electrons. Final state QED showering of electrons is normally turned off because the PYTHIA final state showering code indiscriminately uses the invariant mass of final state fermion-antifermion pairs for the maximum virtuality scale; however, PYTHIA final state QED showering of electrons was turned on for all Ecm = 250 GeV processes so that its effects could be studied in e+e- --> e+e-H. Otherwise default parameters are used for parton showering and fragmentation.

The interface to PYTHIA is contained in ftp://ftp-lcd.slac.stanford.edu/ilc3/whizdata/whizard/whizard-1.95/whizard-src/user.f90 . The source code for referenced subroutines can be found in ftp://ftp-lcd.slac.stanford.edu/ilc/ILC500/StandardModel/a6f/include .

Color flow information was not available in WHIZARD 1.40, and so kinematic and parton id information is used to identify color singlet systems (see ftp://ftp-lcd.slac.stanford.edu/ilc/ILC500/StandardModel/a6f/include/ilc_fragment_call.f90 and ftp://ftp-lcd.slac.stanford.edu/ilc/ILC500/StandardModel/a6f/include/calc_a1sq_a2sq.f90 ).

Kinematic Cuts

Kinematic cuts are applied to massless particle configurations where some cutoff is required to avoid an infinite cross section. In the WHIZARD event generation stage we set the masses of all first and second generation fermions to zero, so that photons, gluons, electrons, muons and u,d,s,c quarks are all subject to kinematic cuts. The cuts are given by the WHIZARD input parameters default_jet_cut, default_mass_cut and default_q_cut . We use the WHIZARD default value
default_jet_cut=10 GeV for the minimum invariant mass of a pair of colored particles. We use default_mass_cut=4 GeV for the minimum invariant mass of a pair of colorless particles, and we use default_q_cut=4 GeV for the minimum sqrt(-Q**2) for
massless t-channel processes.

The massless assumption for the first and second fermion generations can produce some odd results given the kinematic cut values we have chosen. For example, the cross section for gamma gamma -> tau tau is significantly larger than the cross section for gamma gamma -> mu mu since the cut value of 4 GeV for default_mass_cut and default_q_cut is much larger than the tau mass.
Also the cross section for gamma gamma -> qq is suppressed relative to the corresponding cross section for lepton pair
production because the cut value of 10 GeV for default_jet_cut is larger than the cut value of 4 GeV for default_mass_cut .

Mixed stdhep files.

There are over 3500 files in ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/, each produced with 100% electron and positron polarization. To use these
stdhep files in practice one must read the correct number of events from a subset of these files, or read a mixed stdhep file. A mixed stdhep file is built from the files
in ftp://ftp-lcd.slac.stanford.edu/ilc2/whizdata/ILC500/ and corresponds to a particular subset of final states with a particular initial state electron and positron polarization
combination (such as -80% electron and +30% positron).

Event Weight

Due to the presence of some high cross section processes, the events in a mixed stdhep file are not completely unweighted.
The event weight must therefore always be considered when analyzing events.
This weight is stored in the variable EVENTWEIGHTLH in the stdhep common block HEPEV4.

Process Identification

For each event in a mixed stdhep file the variable IDRUPLH from the common block HEPEV4 is used to identify the process. (See the
description of the IDRUPLH variable in the introduction above.)

LOI Mixed stdhep files with randomized final states.

The following table lists Mixed Stdhep files that were produced for the LOI. The Raw Stdhep information corresponds to the original stdhep files, each with a single final state and 100% initial state polarization.

$\mathbf{ e^-}$

$\mathbf{e^+} $

The lumi values in this table refer to the event weight normalization, and may not correspond in some instances to the ratio of the number of events to the cross section. Each Mixed Stdhep
File directory contains files inv_ab_stdhep_files_XXX_ecmEEE_80L_30R and inv_ab_stdhep_files_XXX_ecmEEE_80R_30L which describe the raw stdhep files used to build the mixed files (for example ftp://ftp-lcd.slac.stanford.edu/ilc/ILC250/Large_Stdhep_SM/inv_ab_stdhep_files_Large_Stdhep_SM_ecm250_80L_30R).
The fields in the ..._inv_ab_stdhep_files are:

The "index" link in the "Raw Stdhep Files" column takes you to the list of processes in the mixed stdhep file. The five columns in this list are: process_id, initial_state,
the variable IDRUPLH, an internal sequence number (1 - 4) and a bit indicating whether or not events were produced for this particular initial state polarization sign combination (0 = events generated, 1= events not generated). The internal sequence number denotes the initial state electron or positron helicity and/or the type of initial state photon (EPA or bremsstrahlung):

DBD Mixed stdhep files with randomized final states.

The following table lists Mixed Stdhep files that have been produced for the DBD.

$\mathbf{ e^-}$

$\mathbf{e^+} $

FAQ

Whizard 1.40 has no gluon emission by default, leading to potentially incorrect multiplicity distributions.

The WHIZARD version 1.40 that was used to generate this sample indeed did not include gluon emission. However gluon radiation was simulated using PYTHIA's parton showering algorithm. WHIZARD versions 1.50 and higher include gluon emission, and, starting with version 1.91, WHIZARD has its own parton showering code.

Whizard 1.40 has an incorrect implementation of the CKM matrix. Only diagonal terms of the matrix are present (and = 1!), giving wrong W decays.

Although true for the Whizard version 1.40 that was used to generate this data sample, it is extremely doubtful that this will have any effect on the current analyses. WHIZARD versions 1.51 and higher include the correct CKM matrix, and so future data samples will include the rarer W decays.

This sample has generator level cuts a la SiD, providing a potential bias when used for ILD.

There are, indeed, some kinematical cuts for processes with divergent cross-sections, which can be seen by looking at the whizard.in file as described above. However, the only kinematic cut that leads to a genuine loss of events is a 4 GeV minimum invariant mass cut on final state fermion-antifermion pairs.