The Ge Reprocessing Tools (GeRT) allow users to carry out custom offline data reduction of Spitzer MIPS-Ge (70 micron and 160 micron) data. The GeRT is contributed software from the MIPS Instrument Support Team made available to the public to help users maximize their science returns with MIPS-Ge data. Users should be careful when processing their data offline. If used properly, the GeRT can yield enhanced data products, but users can also corrupt the quality and calibration of their data if used improperly.
The GeRT uses an offline version of the SSC pipeline to produce the basic calibrated data products (BCDs), following the algorithms derived by the MIPS Instrument Team and the MIPS Instrument Support Team (Gordon et al. 2005, PASP, 117, 503). With the GeRT, users can reprocess their data completely from scratch starting with the "raw" data from the archive or can cleanup (re-filter) the BCD products from the archive. The GeRT scripts are comprised of Perl and tcsh wrappers that use copies of SSC downlink binaries and libraries. The GeRT package includes all of the calibration files (e.g., DARK, IC, PMASK) and "namelist" files (namelists are ASCII parameter files that control the way the processing is done. They can be found in the cdf/ directory of your GeRT distribution).
When new calibration files, namelist files, and/or new software techniques are available, users can use the GeRT to reprocess their data offline instead of waiting for the online re-processing of their data. In addition, expert users may find it useful to optimize the namelist files for their specific science application. The SSC online pipelines have been optimized primarily for faint point sources in low background regions. The online pipelines are robust for many science applications, but observers of bright, extended regions may find that offline processing with the GeRT is helpful. The MIPS-IST encourages users of the GeRT to provide feedback (contact the helpdesk at http://irsa.ipac.caltech.edu/data/SPITZER/docs/spitzerhelpdesk/), so that we can improve the processing of MIPS-Ge data.
The most common reasons for using the GeRT are:
Re-filtering the data to optimize it for extended sources;
Correcting poor stim flash calibration in the BCD frames;
Re-filtering the data to correct for poor calibration due to bright sources;
Recovering flux densities from saturated sources.
Important Documentation
The following documents and webpages are recommended reading, and are referred to throughout this User’s Guide:
GeRT announcements are made online on the GeRT main page. The Spitzer Data Analysis Cookbook contains worked examples of common uses of the GeRT. If you have any questions about the GeRT, you can get help by submitting a ticket to the Spitzer Helpdesk. Please let us know which operating platform you are running (MacOSX/Linux/Solaris), as much information as possible about the problem, the AOR ID of your data, plus any error messages you saw. We endeavor to reply to all queries within 48 hours. Please note that our system does not autoreply. If you receive a message back within a few minutes, please check it for further information.
1.2 GeRT Updates
Version 070904 is the most recent release for the Mac version of the GeRT, and contains updated binaries and calibration files that are/will be implemented for the S16/S17 version of the online MIPS pipeline. S16 changes are restricted to changes of some cal files (160 micron flux conversion factor and various SED mode calibration files). For S17, there are some changes to the Interp module to better handle erroneous stimflash minus background calculations and stimflash extrapolations.
The 060415 version of the GeRT is the most recent version for all other platforms, and uses the latest SSC S14.0.0 software version and the current online S13.2 calibration files (see MIPS data handbook for calibration details). This version of the GeRT was designed to process all types of input MIPS-Ge data, e.g., PHT, SCAN, FINE, TPM, SED, DARK, and IC data. The standard science modes (PHT, SCAN) have been tested extensively, but the other modes have yet to be fully validated. In addition, the GeRT is now independent of the MOPEX modules.
The GeRT can now use Perl THREADs to carry out multiple processing jobs in parallel, significantly increasing the speed of the processing. This is particularly useful if running the GeRT on multi-CPU machines. Set the GERT_THREAD parameter to a value of approximately number_of_CPUs + 1 ---> 2 x number_of_CPUs for optimal performance. To use THREADs your system Perl must include Thread.pm; if you run into "Can't locate Thread.pm in @INC" errors, then your system does not support Perl threads, and you will need to set GERT_THREAD = 0.
The old version of the GeRT only worked for Solaris with the FORTE compilers and was tied to a specific version of Perl used by our database system. The new version of the GeRT uses the Perl version on your system. Not all Solaris systems support the FORTE compile. GNU-compiled versions of the GeRT for Solaris, Linux, and Mac have also been made recently. The GNU-software versions make the GeRT more portable, but some system specific issues still exist given the use of shared system libraries.
The webpage provides full installation and setup instructions for all of the supported operating systems (Solaris, Linux, Mac). .
The GeRT package includes the following directories and files:
cal/ Calibration files used for MIPS-70 and MIPS-160
cdf/ Namelist directory
cdf/2 Namelists for MIPS-70 bcd pipeline processing
cdf/3 Namelists for MIPS-160 bcd pipeline processing
cdf/*.nl Namelists for filtering scripts and postbcd processing
scripts/ Scripts for MIPS-Ge processing
bin/ MIPS-Ge binaries
lib/ MIPS-Ge shared libraries (if applicable)
gert_setup.csh Environment setup file for running GeRT software
gert_doc Documentation for this package (ASCII)
1.4 BCD Processing
The majority of users will find the online BCD products sufficient for their purposes (before or after re-filtering; see §2.1). However, since it is simple to reprocess data, users may want to check if the GeRT offline processing yields better products than those currently available in the SSC archive. In particular, the GeRT has proven useful for the following cases:
1. Data where the online stimflash calibration was not done optimally (e.g., extrapolated stimflash calibration or stimflash calibration corrupted by bright emission regions).
2. New updated calibration files and/or pipeline corrections which are not available online yet.
3. Optimization of the pipeline namelist files for non-standard data and/or specific science applications (e.g., ultra-deep data with high redundancy vs low redundancy data).
4. Data brighter than the advertised saturation limits (see §2.4).
