Introduction to Background Analysis Scripts
Description of Individual Background Analysis Scripts

Introduction to Background Analysis Scripts

• EPIC background datasets have been produced by Dave Lumb by co-adding several source-subtracted datasets. These `blank sky' event lists can be used within the scripts presented here. They can be obtained from here, along with useful explanatory notes. More recent, long-exposure EPIC background maps and datasets have been produced by Andy Read. These products, together with related software and explanotary notes, can be obtained here.

• Furthermore, event lists combining several CLOSED observations have been created by Phillipe Marty, and these have also proven very useful in the analysis and background-subtraction of extended objects. A zipped tarball containing MOS1, MOS2 and PN (full-frame and extended full-frame) CLOSED data is available here.

• Other useful documents include Silvano Molendi's presentation `Spatially Resolved Spectroscopy with EPIC' at the November 2001 XMM-Newton Science Analysis Workshop.

• A very brief summary outlining current thinking as to the temporal, spatial and spectral properties of the major components contributing to the XMM-Newton background can be found here.

There are a couple of useful points to note when using the following scripts:

• To see the full task help and the input arguments required by a script, type the script name on the command line with no arguments
• It is always a good idea to 'pipe' the output of a script to a log file, using e.g. tee. For example:

script.csh argument1 argument2 root |& tee rootlog.txt
This means you have a record of what the script did, and the arguments you gave it.

A zipped tarball containing all of the background analysis scripts is available here.

The scripts are written in TCSH shell, and should run on solaris or linux machines.

back to top

Description of Individual Background Analysis Scripts

• skycast
• Casts XMM EPIC background datasets (or indeed any EPIC event dataset) onto the sky, at the position given by an input template event dataset or image.
• imageBGsub
• Performs rigorous background subtraction of an image, using blank sky background fields and filter-CLOSED observations.
• compareoutofFOV
• Compares counts in the out-of-field-of-view regions of two event files (e.g. a source and blank-sky background file). Useful for background scaling.
• createspectra
• Creates spectra in specific annuli from two event files (e.g. source and background spectra). A scaling factor (from e.g. compareoutofFOV) is applied, and scaled background and source minus scaled background (`soft excess') spectra are produced. Arf and rmf files are also produced.
• doublesubtraction
• Given source and scaled background spectra from the same area (e.g. from createspectra) and a large-radius `source minus scaled background' spectrum (e.g. from createspectra), a `double-background' spectrum is created, associated with the source spectrum (along with arf and rmf files) and grouped together as input to xspec.

Before running any of these scripts, the XMM-SAS needs to be correctly configured, and all relevant SAS environment variables (e.g. SAS_ODF, SAS_CCF) need to be correctly set.
To create a background subtracted image, the relevant blank-sky background and filter-CLOSED datsets need to be skycasted. Also, the same filtering (for flares, pattern etc.) performed on the user's source dataset should be performed on the blank-sky background dataset. Once this is done, imageBGsub can be run.
To create XSPEC-compatible source and background spectra, the relevant blank-sky background dataset needs to be filtered to the same degree as the user's source dataset. A scaling value, obtained via e.g. compareoutofFOV, can then be fed into createspectra, firstly far from any source region, to create a large-radius `soft excess' spectrum. Annular source and blank-sky background spectra can then be created from the source regions via successive small-radius runs of createspectra. Finally, doublesubtraction creates a `double-background', comprising of the source background and the correctly area-scaled `soft excess' background, and associates this with a XSPEC-compatible source spectrum.

back to top

Further pages on XMM-Newton Data Analysis at Birmingham can be found here.