Project

General

Profile

The SBND flux files

Neutrino flux

While flux has a specific meaning in general physics, neutrino physicists normally use 'flux' as shorthand when referring to the neutrino rays impinging on a detector, often from a beam-line. 'Flux' does also get used in a more literal sense to describe the number of neutrinos passing through a unit area, particularly relevant when measuring a cross-section.

The SBND flux files

In a nutshell, each flux file contains a tree where each entry describes a neutrino as it passes through a 2D window (usually) defined to be upstream of a detector. LArSoft's implementation of GENIE randomly samples these neutrino rays to generate interactions in a detector volume.
How the flux files are produced changed in 2017 but they are still used in LArSoft/sbndcode in exactly the same way. More information about how the flux files used to be generated and how they are generated now can be found in docdb 5672

Where to find the flux files and beam configurations

All of the flux files live on dcache as the files need to be readily available when running grid jobs. All are stored in subdirectories of the following location:

/pnfs/sbnd/persistent/stash/fluxFiles

The files are separated into several categories: BooNEToGSimple (which are the newest and recommended kind of BNB flux file), the older gsimple type (which were are the older flux files and were generated using NuMI style decay algorithms) and the raw files used in the generation of the former two categories.
The flux files are further separated into 'configurations' which describe a particular setup. The configuration encapsulates all of the necessary parameters used in propagating the neutrinos from their parent's decay point to the flux window they pass through. The amount of associated parameters would make the directory structure unwieldy so each configuration is given a unique identifier with a quantitive description given in the gsimple XML file and a more qualitative description given on this wiki page below.

/pnfs/sbnd/persistent/stash/fluxFiles/bnb/BooNEtoGSimple/configD-v1/april07/neutrinoMode

In each of the configuration directories, XML files are located which describe the flux window the neutrinos are propagated to.
An example location is:

/pnfs/sbnd/persistent/stash/fluxFiles/bnb/BooNEtoGSimple/configD-v1/configD-v1.xml

In all cases the XML file accurately describes the flux window but the XML files were used in a literal sense only in the older gsimple files. For the older gsimple files, the main XML file is stored in the sbndutil repository along with the request files used to generate the gsimple files.

The recommended flux files are the BooNEToGSimple files. At time of writing, the corresponding configurations are ConfigF-v1, ConfigG-v1

A note about pre F/G configurations

The beam centre is in the incorrect place along the X-axis for all configuration prior to F/G; it is shifted too far along the +ve x-axis (1.3m rather than 0.457m). 1.3m is actually the beam's x-centre in the uboone coordinate system.

Beam configurations

ConfigA-100m-v1

The baseline length (target to SBND TPC distance) for this configuration is 100 m. Additionally, the beam's axis is offset to the SBND origin such that the beam centre is at +1.3 m along the x-axis in SBND's coordinate system.
The flux window is defined to be somewhat larger than the SBND TPC volume; a 10 m x 10 m flux window but is positioned such that the beam's axis cuts through the exact centre of the flux window. Additionally, the flux window is placed 10 m upstream of the SBND TPC front face.

ConfigB-v1

This configuration copies the settings from ConfigA-100m-v1 but uses a 110 m baseline.

ConfigC-v1

This configuration copies the settings from ConfigB-v1 but uses a very very large flux window (80 m x 80 m) flux window pressed up against the front face of the TPC. The flux files produced with this beam configuration should be used for event generation which includes dirt-based interactions. The reason for placing the flux window up against the TPC is that it's possible to ask GENIE to push rays back along Z when generating interactions in the volume. The idea is that it's better to leave the flux window in some well known location and then have whatever generator you are using decide how far upstream the flux rays actually start.

ConfigD-v1

Files were produced using BooNEtoGSimple. The file uses an identical window setup to ConfigB-v1. A new config has been used due to a newer flux generation method.

ConfigE-v1

Files were produced using BooNEtoGSimple. The file uses an identical window setup to ConfigC-v1. A new config has been used due to a newer flux generation method.

ConfigF-v1

Files were produced using BooNEtoGSimple. The centre of the window has been shifted along X so that it is inline with the engineer drawings (the X,Y beam centre is now at (0.457, 0) rather than (1.3,0)), but otherwise uses an identical window setup to ConfigD-v1.

ConfigG-v1

Files were produced using BooNEtoGSimple. The centre of the window has been shifted along X so that it is inline with the engineer drawings (the X,Y beam centre is now at (0.457, 0) rather than (1.3,0)), but otherwise uses an identical window setup to ConfigE-v1.