2.3. Spectrometer Launch Pad

2.3.1. Does the observation data need re-processing?

There are two ways to obtain the most up-to-date processing of Spectrometer data (SDRG Section 7.3.1):

  • Using the ''on-demand'' reprocessing facility of the HSA - this is described in more detail in Chapter 1 of the Herschel Data Analysis Guide.
  • Using the User pipeline scripts in HIPE (SDRG Section 7.3.4)

Some further processing of the pipeline results may be required depending on the astronomical source:

2.3.2. Re-processing with the User Pipeline Scripts

This topic is covered in SDRG Section 7.3.1

The Spectrometer pipeline flowchart is shown in SDRG Figure 7.15.

Simplified User Pipeline Scripts for re-processing SPIRE data are provided in HIPE:

  • Accessed directly from the Pipeline>SPIRE menu at the top of the HIPE window
  • Include processing steps taking the data from Level 0.5, through to Level 1 and 2
  • Final results are saved as FITS files (not a pool) to any specified directory
  • The Observation Context can also be updated and saved to a new pool

Line by line descriptions of the Spectrometer scripts with example plots are shown in SDRG Section 7.3.4

The scripts assume the following:

  • The data are already stored in a Pool on your disk
  • The latest Calibration Tree is stored as a pool on your disk (to do this, see SDRG Chapter 5)

To run the User Scripts, several options must be set inside the script:

  • Observation ID, data pool name, output directory path (see SDRG Section 7.3.4 for other options)

Considerations for reprocessing with the Spectrometer User scripts are:

  • Memory usage can be reduced by limiting the number of detectors processed in the script
  • Apodization removes the ringing from the instrumental line function at the cost of reduced spectral resolution. Apodized line profiles can be fitted well by Gaussian functions
  • Second level deglitching parameters can be modified in the script if there are outlying scans seen in the level-1 spectra ( SDRG Section 7.3.1 )

Inspecting the data at various stages can provide diagnostic information (see SDRG Section 7.3.1)

General considerations for faint and medium strength sources are (see SDRG Section 7.5):

  • Optimisation of background subtraction
  • Check spectral noise with respect to expected HSpot values
  • Compare with SPIRE Photometer
  • Compare point source and extended calibration

Corrections for semi-extended sources are described in SDRG Section 7.6

And for maps of extended sources (see SDRG Section 7.8):

  • Understanding the SPIRE beam
  • Check for clipping in individual interferograms
  • Restrict the data made into the cube
  • Examine the actual positions on sky observed
  • Change the algorithm used for regridding
  • Examine coverage and redundancy in the cube
  • Maps with faint continuum

2.3.3. Further Analysis

SPIRE spectral analysis is described in SDRG Section 7.11 and cube analysis in SDRG Section 7.12.

Spectral analysis and visualisation tools are provided in HIPE (right click on product in Variables tab > Open With):

  • SPIRE specific SDI/SDS Explorer (SDRG Section 8.1) for viewing spectra/interferograms via a clickable footprint of the detector arrays
  • Spectrum Explorer (Herschel Data Analysis Guide Chapters 5 and 6) for viewing one, two, and three-dimensional spectral products
  • Spectrum Toolbox (Herschel Data Analysis Guide Chapters 5 and 6) for a set of mathematical functions that operate on Herschel spectral data (launched from Spectrum Explorer dialogs menu)
  • Spectrum Fitter GUI (Herschel Data Analysis Guide Chapter 7) for interactive line and continuum fitting (launched from Spectrum Explorer dialogs menu)
  • Cube Spectrum Analysis Toolbox (Herschel Data Analysis Guide Chapter 6) for spectral cube related tasks (launched from Spectrum Explorer dialogs menu)

Spectrometer useful scripts (Scripts > SPIRE Useful scripts menu at the top of the HIPE window):

Additional notes for SPIRE Spectrometer data:

  • The spectra have been corrected to the Local Standard of Rest only from HIPE v12 onwards - data processed with previous versions were not corrected (see SDRG Section 7.11)
  • Lines in high resolution FTS spectra should be fitted with a Sinc profile for unapodized data, and a Gaussian for apodized data. Various tools exist within and outside of HIPE to fit lines to spectra (see SDRG Section 7.11)
  • The beam size of the FTS changes across the band of each array in a non-trivial way (see SDRG Section 7.6 )
  • Each Spectrometer Useful Script contains an example observation ID. This observation will download from the Herschel Science Archive, if HIPE is connected to the internet, so the respective script can be tested out of the box