Total Power to Visibilities (TP2VIS): an ALMA Cycle 4 Development Study

Jin Koda, Peter Teuben, Adele Plunkett, Tsuyoshi Sawada, Crystal Brogan, Ed Formalont

This project provides tools to create visibilities from a single dish cube using the method of Koda et al. 2011 and Koda et al. 2019 (PASP in press). The TP visibilities can then be combined with the interferometric visibilities in a joint deconvolution using for example CASA's tclean() method. TP2VIS requires CASA 5.4 or above and as powerful a computer as what the CASA Feather guide requires.

Our github repo for distribution : https://github.com/tp2vis/distribute

This release assumes that users have experience on interferometer data reduction with CASA so that they can catch any anomaly by themselves. This release is ready for scientific use, but please let us know if you encounter any problem.

The following acknowledgement would be appreciated if you decide to use TP2VIS: "This work made use of TP2VIS (Koda et al. 2011 ApJS, 139, 19; Koda et al. 2019, PASP in press)".

Click "Clone or download" on top page for download options, or run

   git clone https://github.com/tp2vis/distribute.git

You need only one script "tp2vis.py".

To give you a quick idea how to run TP2VIS, here are the basic flow of commands in CASA, broken into 6 pieces. This document gives an overall flow, but look at examples listed below for more info.

   listobs('calibrated_12m.ms',listfile='calibrated_12m.log')

after which you can cut and paste the appropriate Fields in the following format into a pointing file, 12m.ptg, which has to be a simple text file:

   cat 12m.ptg
   
   J2000 05h39m45.660s -70d07m57.524s
   J2000 05h39m54.340s -70d07m57.524s
   J2000 05h39m41.320s -70d07m19.175s
   ...

This can be a little cumbersome, so in the examples listed below you can see examples using awk and grep.

We need to know the RMS in the TP cube from some line free channels. For example, you might be able to use the first 10 channels of your TP cube

   importfits(fitsimage='tp.fits',imagename='tp.im')              # convert to CASA image format
   imstat('tp.im',axes=[0,1])['rms'][:10].mean()
   -> 0.67

Cut down unnecessary spws from measurement sets as tp2vis assumes all spws to be used for imaging. CASA tasks such as split() and mstransform() can be useful.

   execfile('tp2vis.py')                                          # load tp2vis 
   tp2vis('tp.im','tp.ms','12m.ptg',rms=0.67)                     # make visibilities

   tp2viswt('tp.ms', ...)                                         # set weights
   tp2vispl(['12m.ms','7m.ms','tp.ms'])                           # (optional) plot weights

"tp2viswt" shows weight statistics and manipulates weights. There are several modes how you can set weight, described in example1. "tp2vispl" plots the weights.

In case tclean crashes in the next step (perhaps due to an inconsistency in CASA versions), there is no single workaround. The below worked in some cases (more in example1, section 4):

   concat(vis=['12m.ms','7m.ms','tp.ms'], concatvis='all.ms',copypointing=False)

The "copypointing=False" option is important. No worries, important pointing info still remain in 'all.ms' (one of CASA mysteries!).

   tclean(vis='all.ms', imagename='all_clean', ...)                     # run clean the way you like it

Where ... represents the large number of options to control the deconvolution. For example, users may try "robust" and "uvtaper" options.

Make dirty images as well

   tclean(vis='all.ms', imagename='all_dirty', niter=0, ...)                     # make dirty map

This step is for general interferometer imaging, not only for TP2VIS.

Once dirty and cleaned images are generated, one may correct for the discrepancy between dirty and clean/restore beam areas (see Jorsater and van Moorsel 1995).

   tp2vistweak('all_dirty','all_clean')                   # adjust dirty beam size in residual image

It assumes that all outputs from tclean() are kept intact (no name change). It creates .tweak.image and .tweak.residual, which have a correct flux scale.

• example1: M100 (data from CASA guide on Feather)

The spiral galaxy M100:

Negative sidelobes around strong emissions in the 7m+12m map (left), but not in the TR+7m+12m map (right).

A giant molecular cloud in the Large Magellanic Cloud:

Most extended emissions are not recovered in the 7m+12m map (left), but recovered in the TR+7m+12m map (right).

• Koda et al. 2011, ApJS, 193, 19 : http://adsabs.harvard.edu/abs/2011ApJS..193...19K

• Koda, Teuben, Sawada, Plunkett & Fomalont 2019, PASP, 131, 54505: https://ui.adsabs.harvard.edu/abs/2019PASP..131e4505K

• CASA reference manual and cookbook : http://casa.nrao.edu/docs/cookbook/

  • Measurement Set: https://casa.nrao.edu/casadocs/latest/reference-material/measurement-set

• Jorsater and van Moorsel 1995, AJ, 110, 2037 : http://adsabs.harvard.edu/abs/1995AJ....110.2037J

We thank the NRAO staff, in particular, Remy Indebetouw, Kumar Golap, Jennifer Donovan Meyer, Crystal Brogan, and John Carpenter for their help. We also thank Kazuki Tokuda, Fumi Egusa, Manuel Fernández, and Mercedes Vazzano for feedback on an early version of TP2VIS.