Jupiter: continuum polarization calibration 5.5.0
Data Import
ADD DATA LOCATION
Import the data into CASA. Task importuvfits() will read the original AIPS friendly UVFITS format and create a CASA native MS.
importuvfits(fitsfile='planets_6cm.fits', vis='jupiter6cm.demo.ms', antnamescheme='old')
Data Inspection and Editing
First, check the observation set up and print verbose summary of the observations to the CASA logger
listobs(vis='jupiter6cm.demo.ms')
The output of this task is fairly long, the abridged version is shown below since we may need some of this information for reference later on during our calibration.
########################################## ##### Begin Task: listobs ##### Observer: FLUX99 Project: Observation: VLA Computing scan and subscan properties... Data records: 2021424 Total elapsed time = 85136.5 seconds Observed from 15-Apr-1999/23:15:25.0 to 16-Apr-1999/22:54:21.6 (TAI) [...] Fields: 13 ID Code Name RA Decl Epoch SrcId nRows 0 A 0137+331 01:37:41.299500 +33.09.35.13400 J2000 0 72946 1 T 0813+482 08:13:36.051800 +48.13.02.26200 J2000 1 227524 2 A 0542+498 05:42:36.137900 +49.51.07.23400 J2000 2 227058 3 T 0437+296 04:37:04.174700 +29.40.15.13600 J2000 3 113424 4 VENUS 04:06:54.109428 +22.30.35.90609 J2000 4 121098 5 A 0521+166 05:21:09.886000 +16.38.22.05200 J2000 5 81258 6 T 1411+522 14:11:20.647700 +52.12.09.14100 J2000 6 243608 7 A 1331+305 13:31:08.288100 +30.30.32.96000 J2000 7 307958 8 MARS 14:21:41.365747 -12.21.49.45444 J2000 8 118570 9 NGC7027 21:07:01.593000 +42.14.10.18600 J2000 9 136082 10 NEPTUNE 20:26:01.136316 -18.54.54.21127 J2000 10 157736 11 URANUS 21:15:42.828572 -16.35.05.59272 J2000 11 99412 12 JUPITER 00:55:34.043951 +04.45.44.70633 J2000 12 114750 Spectral Windows: (2 unique spectral windows and 1 unique polarization setups) SpwID Name #Chans Frame Ch0(MHz) ChanWid(kHz) TotBW(kHz) CtrFreq(MHz) Corrs 0 none 1 TOPO 4885.100 50000.000 50000.0 4885.1000 RR RL LR LL 1 none 1 TOPO 4835.100 50000.000 50000.0 4835.1000 RR RL LR LL Sources: 26 ID Name SpwId RestFreq(MHz) SysVel(km/s) 0 0137+331 0 0 0 0 0137+331 1 0 0 1 0813+482 0 0 0 1 0813+482 1 0 0 2 0542+498 0 0 0 2 0542+498 1 0 0 3 0437+296 0 0 0 3 0437+296 1 0 0 4 VENUS 0 0 0 4 VENUS 1 0 0 5 0521+166 0 0 0 5 0521+166 1 0 0 6 1411+522 0 0 0 6 1411+522 1 0 0 7 1331+305 0 0 0 7 1331+305 1 0 0 8 MARS 0 0 0 8 MARS 1 0 0 9 NGC7027 0 0 0 9 NGC7027 1 0 0 10 NEPTUNE 0 0 0 10 NEPTUNE 1 0 0 11 URANUS 0 0 0 11 URANUS 1 0 0 12 JUPITER 0 0 0 12 JUPITER 1 0 0 Antennas 27 [....] ##### End Task: listobs ##### ##########################################
Check also the array configuration and pick a reference antenna. Here, we will use antenna '11' as a reference since it's located in the middle of the array.
plotants(vis='jupiter6cm.demo.ms', figfile='jupiter6cm.demo.ant.png')
ADD FLAGGING
Calibration
Set the Flux Density Scale
Set the absolute flux density scale, but only for Stokes I at the moment (total flux density model). Our primary flux calibrator here is 1331+305 (3C286). The default model for CASA 5.3+ is 'Perley-Butler 2017'.
setjy(vis='jupiter6cm.demo.ms', field='1331+305', model='3C286_C.im')
Initial gain calibration
At this stage the data have an overall flux density scaling determined, but full gain solutions aren't there yet. The relevant task is gaincal() (analogous to the AIPS task CALIB). Gaincal() will produce a separate table with solutions, and we will use appropriate extensions to keep track of what is what.
NOTE: Since we have only two single-channel continuum spw, we do not want to do bandpass calibration.
Firstly, generate an antenna zenith-angle dependent VLA gain curve calibration table
gencal(vis='jupiter6cm.demo.ms', caltable='jupiter6cm.demo.gc', caltype='gc')
Now, solve for antenna gains on 1331+305 and 0137+331, using the just generated gain curve table (.gc).
gaincal(vis='jupiter6cm.demo.ms', caltable='jupiter6cm.demo.G', field='1331+305,0137+331', spw='', gaintype='G', calmode='ap', solint='inf', combine='', refant='11', minsnr=3, gaintable=['jupiter6cm.demo.gc'], parang=False)
# And check the solutions
plotcal(caltable='jupiter6cm.demo.G', xaxis='time', yaxis='amp', subplot='333', iteration='antenna')
If all looks good, bootstrap the flux density scale of the flux calibrator onto the phase calibrators. AIPS called it GETJY, but CASA calls it fluxscale().
myFluxscale = fluxscale(vis='jupiter6cm.demo.ms', caltable='jupiter6cm.demo.G', fluxtable='jupiter6cm.demo.Gflx', reference='1331+305', transfer='0137+331', append=False, display=False)
The output is displayed in the logger as well as stored in the myFluxscale python dictionary.
Beginning fluxscale--(MSSelection version)------- Found reference field(s): 1331+305 Found transfer field(s): 0137+331 Flux density for 0137+331 in SpW=0 (freq=4.8851e+09 Hz) is: 5.29665 +/- 0.00449217 (SNR = 1179.09, N = 54) Flux density for 0137+331 in SpW=1 (freq=4.8351e+09 Hz) is: 5.34890 +/- 0.00176819 (SNR = 3025.07, N = 54) Fitted spectrum for 0137+331 with fitorder=1: Flux density = 5.32271 +/- 1.70229e-07 (freq=4.86004 GHz) spidx=-0.954124 (degenerate) Storing result in jupiter6cm.demo.Gflx Writing solutions to table: jupiter6cm.demo.Gflx
When executing fluxscale(), the calibration table with the extension .G is modified and stored as a new table with the extension .Gflx. So far, solutions have been generated only for the calibrators, and have not yet been transferred to the target source(s).
Before proceeding, inspect the flux density calibration and save results to a file.
plotcal(caltable='jupiter6cm.demo.Gflx', xaxis='time', yaxis='amp', showgui=True, figfile='jupiter6cm.demo.Gflx.amp.png')
plotcal(caltable='jupiter6cm.demo.Gflx', xaxis='time', yaxis='phase', plotrange=[-1,-1,-200,200], showgui=True, figfile='jupiter6cm.demo.Gflx.phase.png')
Polarisation calibration
Set the polarisation model
Now, set the polarisation model for the polarised position-angle calibrator (here 1331+305=3C286 which is also our primary flux calibrator). For polarisation properties of your primary polarisation calibrator see VLA Polarimetry Guide.
i0=7.3109 # Stokes I value for spw 0 ch 0
f0=4.8851 # Frequency for spw0 ch0 (note that in our data the 'lower' spw is actually higher frequency)
alpha=log(i0/7.35974932)/log(4.8351/f0) # Values from our setjy() run on Stokes I earlier
c0=0.114 # Fractional polarisation 11.4% for 5GHz
d0=33*pi/180 # Polarisation angle 33deg in radians
myPolSetjy = setjy(vis='jupiter6cm.demo.ms', field='1331+305', standard='manual', spw='0', fluxdensity=[i0,0,0,0], spix=[alpha,0], reffreq=str(f0)+'GHz', polindex=[c0,0], polangle=[d0,0], scalebychan=True, usescratch=False)
The results are displayed in the CASA logger as well as myPolSetjy python dictionary
#In CASA
CASA <49>: myPolSetjy
Out[49]:
{'7': {'0': {'fluxd': array([ 7.3109 , 0.33899165, 0.7613877 , 0. ])},
'fieldName': '1331+305'},
'format': "{field Id: {spw Id: {fluxd: [I,Q,U,V] in Jy}, 'fieldName':field name }}"}
Solve for the cross-hand delays
Solve for the cross-hand delays (RL,LR) delays due to residual delay difference between R and L. For this use the 3C286 calibrator since it had a strong polarised signal in the RL and LR correlations.
gaincal(vis='jupiter6cm.demo.ms', caltable='jupiter6cm.demo.K', field='1331+305', spw='', gaintype='KCROSS', solint='inf', combine='scan', refant='11', minsnr=3, gaintable=['jupiter6cm.demo.gc','jupiter6cm.demo.G'], gainfield=['','1331+305'], parang=False)
Check the results in the CASA logger window
For solint = inf, found 2 solution intervals. Time=1999/04/16/06:38:49.9 Spw=0 Global cross-hand delay=0.102352 nsec Time=1999/04/16/06:36:28.4 Spw=1 Global cross-hand delay=-9.36806e-06 nsec Found good Kcross Jones solutions in 2 solution intervals. Writing solutions to table: jupiter6cm.demo.K Finished solving.
Solve for the leakage terms (D terms)
Solve for polarization leakage on 0137+331; assume it has unknown polarisation (poltype='D+QU' if good parallactic coverage, 'D' otherwise, consult polcal() help for more information on options).
Solve for the R-L polarisation angle (X term)
Template:Checked 5.3.0
