crush

Section: User Manuals (1)
Updated: 22 JUL 2019
Return to Main Contents

NAME

crush - Data Reduction and Imaging

SYNOPSIS

crush <instrument> [options] [scans] ...

Parameters:

<instrument>: sharc2, laboca, saboca, polka, gismo, scuba2, hawc+, or hirmes.
[options]: various configuration options (see GLOSSARY). Global settings must precede all scans on argument list. Each scan will use all options listed before it on the command line.
[scans]: A list of scans (file names or scan numbers or ranges).

DESCRIPTION

CRUSH is an astronomical data reduction/imaging tool for select cameras. Currently, it supports SHARC-2, LABOCA, SABOCA, PolKa, GISMO, SCUBA-2, SOFIA/HAWC+, and SOFIA/HIRMES. CRUSH is written entirely in java(1) , allowing it to run on virtually any platform. It is normally run from the command-line with several arguments. For a full documentation please refer to the README file inside the distribution directory (usually /usr/share/crush).

OPTIONS

Here are some of the most commonly used options available. For full and detailed description of all options please consult the GLOSSARY.

Location of input files:

-datapath=: Specify the path to the raw data. You may include environment variables enclosed in curly brackets with a $ sign (e.g. {$HOME} for the user's home directory), as well as the shorthands '~' (also for the user's home) or, in most cases, also '~jsmith' for the home of user jsmith.
-date=: GISMO: Specify the date of observation as 'YYYY-MM-DD'.
SCUBA-2: Specify the date of observation as 'YYYYMMDD'.
-object=: GISMO: Specify the object name as it was entered into PaKo.
-project=: APEX: Specify the project ID (case insensitive). E.g. 'T-79.F-0002-2007'.
-file=: SHARC: Specify the data file (relative to -datapath) containing the scans.

Optimize reduction by source type:

-bright: Reduce bright sources (S/N > 1000).
-faint: Use with faint sources (S/N < 10).
-deep: Use with deep fields. Optimize the reduction for the recovery of faint point sources. Emission on larger scales may be strongly filtered.
-extended: Assume extended structures (>FoV/2). The recovery of emission on large scales is limited, and usually results in significantly increased noise levels/baselines (especially on the extended scales). Please refer to the README for a detailed explanation.
-moving: Target is a moving object (e.g. planet, asteroid, or moon). CRUSH will use offsets to the tracking center, instead of absolute coordinates, to ensure that all data is properly aligned, especially over multiple scans. A nominal source position (e.g. for maps) is taken from the first scan listed.

Options for the output map:

-outpath=: Specify the directory where output files will go. See the -datapath option on including environment variables and shorthands.
-name=: Specify the output FITS map file name (rel. to outpath).
-gzip: Compress the output with gzip if possible (.gz exztension added as needed).
-projection=: The spherical projection to use (SFL, SIN, TAN, GLS, MER, CAR, AIT, STG, ARC or ZEA).
-grid=: The map pixelization (arcsec).
-altaz: Reduce in horizontal coordinates (e.g. for pointing).
-ecliptic: Reduce in Ecliptic coordinates.
-galactic: Reduce in Galactic coordinates.
-final:smooth=: Smoothing of the map in the final iteration, either as a FWHM (arcsec), or one of the predefined values: minimal, halfbeam, 2/3beam, beam.

Commonly used options for scans:

-tau=: Specify an in-band zenith tau value, source ID, or interpolation table to use. E.g.: 1.036 or 225GHz or ${CRUSH}/gismo/tau-2014-04.dat
-tau.<id>=: Specify a zenith tau value or interpolation table for<id>. E.g.: tau.225GHz=0.075.
-scale=: Apply a calibration correction factor to the scan(s), or provide a table to interpolate from.
-pointing=: x,y incremental pointing corrections (in telescope's natural coordinate system (e.g. AZ/EL) in arcsec.

Alternative reduction modes:

-point: Reduce pointing/calibration scans and suggest pointing corrections.
-skydip: Reduce skydips to obtain in-band zenith opacity.
-pixelmap: Derive pixel position data from beam maps.
-split: Indicate that the scans are part of a larger dataset, which cannot be reduced in a single go. The resulting maps from such a split can be combined by coadd(1), and fed back into crush for another round of reduction (if desired) via the source.model option (see GLOSSARY).

Other useful options:

-show: Display the result (if possible) at the end of the reduction.
-forget=: Comma separated list of options to unset.
-blacklist=: Comma separated list of options to ignore.
-config=: Load configuration file.
-poll: Poll the currently set options.
-conditions: List all conditional settings.

FILES

/usr/share/crush: Directory where CRUSH resides if installed by a package manager, or via the installation script by a super-user.
~/.crush2/: Location of user configuration files (e.g. default.cfg).
~/.crush2/<instrument>/: Location of instrument-specific user configurations.
~/.crush2/startup/crush/: Location of user-specific java(1) runtime configurations.

BUGS

If you find them, please report them to the author (see below).

AUTHOR

Attila Kovács <attila[AT]sigmyne.com>