21 February 2013, 14:30 CET
Chair: Ivan Valtchanov (HSC)

On the call: Ivan Valtchanov, Daizhong Liu, Dominique Benielli, Ed Polehampton, Bruno Merín, Ros Hoopgood, Stephan Ott

Video tutorials from the webinar

SPIRE FTS Spectral mapping on-line tutorial part 1/3


- Minute 0:00: how to get an observation from the Herschel Science Archive and then load it in HIPE

- Minute 8:30: how to do quickly check the observation summary and the spectral cube

- Minute 9:10: how to extract 1-d spectra from spaxels or at a particular sky position RA,Dec

- Minute 20:00: how to convert the extracted 1-d spectrum from extended calibration to point source

- Minute 21:30: how to subtract a particular spectrum from the cube

- Minute 25:35: end of video

SPIRE FTS Spectral mapping on-line tutorial part 2/3


- Minute 0:00: how to use the cube line fitting useful script

- Minute 4:50: how to fit lines and continuum to all cube pixels

- Minute 9:20: how to make line maps

- Minute 13:05: how to compare a line map with the photometer observation

- Minute 22:45: end of video

SPIRE FTS Spectral mapping on-line tutorial part 3/3


- Minute 0:00: how you can recreate the cube with a different WCS with bigger pixels and/or different reference position.

- Minute 10:02: end of video.

Demo script

A short script used in the webinar:

Questions and Answers

Q - Daizhong Liu: we have observations of galaxies and would like to know what are the difference between HIPE 9 and 10.
R - Ivan Valtchanov: There are not-very-significant improvements in the telescope model but certainly you should in principle update the data with the latest HIPE.
Ivan demoes the use of the on-demand reprocessing to show what the differences are between data processed with HCSS 9 and HCSS 10 with test data M82.

Q- Daizhong Lui: when you put data on the shopping basket you can download it only right?
R - Ivan Valtchanov: in fact, you can also request an on-demand reprocessing of the observation with the latest operational HCSS version at the ESAC grid.

Q- Daizhong Lui: when dealing with extended sources, when should we use the point-source spectrum of the mapping spectral cube?
Ivan Valtchanov: For these things we need to look into the Level 1 data directly.

Q - Daizhong Lui: when there is a jump between the short and long wavelength, how do I know tackle that? A case when the source is semi-point like, with brightness in homogeneously distributed.
R- Ivan Valtchanov: it depends on how big is the galaxy
R - Ed Polehampton: ideally you need to correct for the coupling between the beam shape and the shape of the source, we are working on a tool to correct for that.

Q - Daizhong Lui: where is the tool?
R - Ed Polehampton: there is a prototype in version 10 and the full version will be in 11. It is equivalent to using some filling factors yourself but this tool does it for you.
beam SSW - 18 arcsec
beam SLW - 20 arcsec
In principle you can do a calculation between size of source and beams to estimate the correction factors.

Ivan Valtchanov/Ed Polehampton demo the semiExtendedCorrectionToolPrototype for M82 with a gaussian source model of 18 to 28 arcsec.
When the tool is doing a good job, the spectra from both detectors match and there is no jump at 1250 GHz, when there is a beam edge.

earlier OD 1011: very similar between 9 and 10
later than OD 1011: larger differences

Q - Bruno Merín: what do the different lines in the SPIRE FTS browse products in the archive mean?
A - Ed Polehampton: The line color in the SPIRE FTS browse products should be explained.
The HR + LR browse product says HR and the number of repetitions is not correct either. This will be corrected in a future HCSS version.


I will assume that people will be familiar with the SPIRE Spectrometer observing modes and observational context data structure, i.e. first couple of sections in Chapter 6 of the SPIRE Data Reduction Guide(DRG).

Data to be used for the webinar

HD37041 (aka Orion Bar), OBSID 1342204919, OD495, full map in high resolution.
Photometer map: OBSID 1342239930

Please make sure you already have the data available in your local machine, the data volume is big and using the Herschel Science Archive will be slow. So, do get the data either from the archive as tar files (data are public) or from within HIPE:

obs = getObservation(1342204919,useHsa=True)
saveProduct(product=obs, pool='OrionBar', tag='OrionBar HR Full OD0495')

you can rename the pool and the tag if you wish.

Note, saving this product can take a very long time, it's preferable to get it as a tar file from the Herschel Science Archive, well in advance of the start time of the webinar.

Webinar Agenda

  1. Quick look at a spectral cube using the Spectrum Explorer and the Cube Toolbox

  2. Illustration of some cube analysis tasks, following the DRG, Section 6.6
    1. Extract spectra from cube spaxels, near a (RA, Dec) position or from a region.
    2. Merge SSW and SLW extracted spectra
    3. Convert to point source calibration
    4. Subtract a spectrum (dark sky) from the cube

  3. Illustration of the lines fitting useful script to cubes
    1. Making line maps
    2. Compare line maps with photometer maps, important considerations!

  4. Reprocess/recreate the cube: when and why.


comments powered by Disqus