Release Notes CASA 4.6.0
The main feature improvements of CASA 4.6.0 are listed below.
We occasionally issue patches and 'monthly' versions of CASA. To get notified, please subscribe to the 'casa-users' mailing list. 'Monthlies' as well as CASA releases are available at Obtaining CASA. Releases will be announced via the 'casa-announce' mailing list. To subscribe, please visit the mailing lists webpage. For feedback, and help please go to the NRAO helpdesk; for ALMA questions please use the ALMA helpdesk
Note that in its current incarnation CASA is designed to support ALMA, and VLA data, as well as single dish data from ALMA and Nobeyama. Data from other telescopes, be it single dish or interferometers can be imported from uvfits, FITS-IDI, or sdfits formats into measurements sets (ms) or scantables in CASA. Given the variety of non-standard fits formats, we cannot guarantee that CASA will fully support data from all telescopes. However, efforts are made to support data formats from other facilities.
Obtaining the CASA Release
This version of CASA is available from our Download page. There, we also list the supported operating systems.
CASA Release Notes:
Valid for CASA Release 4.6.0 and later: The CASA 4.6.0 release emphasized improved stability, robustness, and code refactoring. Many bugs were fixed and development such as parallelization of the calibration steps are being introduced. tclean will eventually replace clean and has underwent a period of intensive testing and improvement. Major feature improvements over the previous version of CASA include:
- tclean is a new imaging task, eventually replacing clean. It has a new, more straightforward interface, allows more combinations of algorithms, include options for output files, and has a more flexible outlier approach. tclean has also been designed to support parallel computing. tclean can be used by everyone but note that implementation, improvements, and testing are still ongoing. If you find any problems with tclean please send us a note through the CASA helpdesk.
- Data examination/editing/import/export
- New task importatca has been implemented to directly convert ATNF RPFITS files to CASA measurement sets.
- Source Table coordinates for ephemeris objects are now updated from the ephemeris as part of the SDM filler step.
- imhead now returns a python dictionary with the image parameters.
- plotms has now options to define the dpi, width, and height for exported plots.
- plotms now displays the average channel number when averaging across channels.
- plotms has now an added feed selection parameter.
- plotms responsiveness for canceling a plot has been improved.
- plotms can restarted if necessary without restarting the casa session.
- plotms can plot averaged WEIGHTs.
- CASA now starts to support image histories. A new task imhistory is now available that can list and append history information. Image analysis tasks start to populate the image history sections.
- The msmd tool can now retrieve the rest frequency and line transition from the MS.
- exportuvfits now has an overwrite parameter.
- imstat has a listfile parameter to write the output in machine- (and human)-readable form into a file.
- Generalized pre-averaging (channel and time) is now available for all autoflagging modes (clip, rflag, tfcrop) in flagdata
- flagdata: New sub-parameter to savepars called "overwrite". The default behaviour of appending to the file has changed. It will either overwrite the existing file or give an error when overwrite=False.
- setjy takes frequency dependent polarization fraction and angle for manual flux settings.
- New parameter offsetstable in wvrgcal has been implemented to allow a user to specify a table with temperature offsets that will be subtracted before the phase corrections are calculated.
- new weight modes 'tsys' and 'tinttsys' in task initweights. The modes initialize WEIGHT(_SPECTRUM) and SIGMA(_SPECTRUM) columns using Tsys values in calibration table
- CASA can read the new format of the ALMA OT pointing files.
- Data and Image Manipulation
- split and hanningsmooth are now what was named split2 and hanningsmooth2 in previous CASA releases. The previous task implementations are still available via oldsplit and oldhanningsmooth.
- ia.subimage has a new parameter keepaxes that controls which degenerate axes shall be dropped when dropdeg=True.
- partition and importasdm can now create a Multi-MS separated across baselines.
- A user can now specify what image header is being used for an immath output.
- Data analysis
- ia.histograms is now inline with the histogram code in the viewer.
- When region files are used and chans or stokes parameters are set, the latter will override the region file specifications. A specification of box will be a logical AND to the regions defined in the region file.
- Single Dish
- a new task, importasap, is available for converting scantables to MS
- enhancements to single dish tasks that operate on MSes:
- a new smoothing kernel, kernel='boxcar', is available in task tsdsmooth. It supports smoothing with pill box kernel function
- a new fitting function, fitfunc='lorentzian', is available in task tsdfit. It supports spectral line fitting by Lorentzian functions
- removed tasks sdtpimaging and msmoments
- introduced a new parameter, intent in tasks tsdbaseline, tsdfit and tsdsmooth to support data selection by scan intent
- removed parameter singledish in task importasdm
- spline interpolation is adopted in sdimaging to interpolate pointing directions in time for gridding
- parallelization for continuum modes implemented in tclean.
- The use of simple_cluster has been deprecated in 4.6 and its use has been removed from all parallelized tasks. Use the MPI framework instead.
- pclean and msmoments are obsolete and have been removed from CASA.
- Experimental tasks
- cvel2 to provide the functionality of cvel based on the mstransform task.
- tclean is now more mature but still under testing. Since tclean is the platform for imaging code parallelization, pclean has been removed from CASA.
- New task msuvbin grids the visibilities independent of the imaging task clean. msuvbin can be useful when large datasets that were observed over multiple epochs are gridded on a common grid.