VLA CASA Bandpass Slope-CASA6.4.1: Difference between revisions
| Line 1: | Line 1: | ||
<b>This CASA Guide is for CASA version 6.4.1</b> | <b>This CASA Guide is for CASA version 6.4.1</b> | ||
== Read this first == | == Read this first == | ||
<source lang= | <source lang=html style="background-color: #AAffff;"> | ||
If you are new to CASA, or to VLA data reduction in CASA, it is | If you are new to CASA, or to VLA data reduction in CASA, it is <b>strongly</b> recommended that you start with the <a href='https://casaguides.nrao.edu/index.php?title=Getting_Started_in_CASA"> Getting Started in CASA</a> guide, the [https://casaguides.nrao.edu/index.php/VLA_high_frequency_Spectral_Line_tutorial_-_IRC%2B10216-CASA6.5.2 IRC+10216 spectral line tutorial], or the [https://casaguides.nrao.edu/index.php/VLA_Continuum_Tutorial_3C391-CASA6.4.1 VLA 3C391 Continuum Tutorial] before proceeding with this tutorial. | ||
</source> | </source> | ||
Revision as of 23:03, 12 January 2023
This CASA Guide is for CASA version 6.4.1
Read this first
If you are new to CASA, or to VLA data reduction in CASA, it is <b>strongly</b> recommended that you start with the <a href='https://casaguides.nrao.edu/index.php?title=Getting_Started_in_CASA"> Getting Started in CASA</a> guide, the [https://casaguides.nrao.edu/index.php/VLA_high_frequency_Spectral_Line_tutorial_-_IRC%2B10216-CASA6.5.2 IRC+10216 spectral line tutorial], or the [https://casaguides.nrao.edu/index.php/VLA_Continuum_Tutorial_3C391-CASA6.4.1 VLA 3C391 Continuum Tutorial] before proceeding with this tutorial.
Overview
For the standard VLA flux density calibrators 3C138, 3C147, 3C286 and 3C48, the CASA distribution includes angular and spectral models that are referenced during calibration. The models account for the observational source characteristics, resulting in improved calibration solutions that therefore more accurately represent the instrumental and atmospheric corrections. These VLA standard calibrators, however, exhibit a negative spectral index (flux density decreasing with increasing frequency) and become relatively weak at high frequencies.
Although the standard VLA flux density calibrators are usually still sufficiently bright enough for absolute flux density calibration even at the higher frequency bands, a good bandpass determination—which is important for spectral line observations or measuring the spectral index of continuum sources—requires high signal-to-noise ratio bandpass solutions derived from either a long integration time or a very bright source (refer to the Spectral Line (Bandpass Setup) section within the Guide to Observing with the VLA). Generally, instead of a long integration on a fainter source, additional observations of non-standard, but bright bandpass calibrators are used. Unfortunately, the spectral characteristics of these sources will be undetermined, and no a-priori flux density model is available. If not accounted for, using undetermined spectral knowledge will introduce bias in the bandpass calibration. This tutorial describes how to determine a source's spectral characteristics and how to correct the bandpass solution for such an effect using these additional sources.
Obtaining the Data
The data used in this guide are taken for a target field in Ka-band with four 2-GHz basebands paired to two chunks of 4 GHz contiguous frequency coverage centered at 29 and 36.5 GHz for a total of 8 GHz frequency coverage. That is, each baseband consists of a series of 16 consecutive 128-MHz subbands linked to another baseband for a total of 64 individual spectral windows.</ br> As this tutorial only concerns bandpass calibration, all scans on fields other than the flux density and bandpass calibrator fields were removed from the measurement set (MS). All pre-calibration steps including flagging, antenna position offsets, requantizer gains, opacity corrections, and gain-elevation curves were applied. The original data (TVER0004.sb14459364.eb14492359.56295.26287841435) can be obtained through the NRAO archive and has a raw size of 57.04 GB.
The trimmed measurement set used here can be downloaded directly from http://casa.nrao.edu/Data/EVLA/G192/G192-BP.ms.tar.gz (dataset size: 3.4 GB)
Your first step will be to uncompress and untar the file in a terminal (before you start CASA):
# in a terminal, outside of CASA: tar -xzvf G192-BP.ms.tar.gz
Then move to your working directory (if not the same as you put the raw data in) and optionally move/copy or link the data to a preferred shortcut name. This is particularly useful if the original observation name has it's full extent like ,TVER0004.sb14459364.eb14492359.56295.26287841435, for example:
# in a terminal, outside of CASA: ln -s <path-to-data>/G192-BP.ms <i>my-vis.ms</i>
Starting CASA
To start CASA, type:
# in a terminal, outside of CASA: casa
or better, to make sure to be running the CASA version for which this tutorial was designed (list the installed versions to choose from by typing casa -ls):
# in a terminal, outside of CASA: casa -r 6.4.1-12-pipeline-2022.2.0.64
This will initialize CASA and set the necessary paths appropriately. This will also create two files; one is ipython-<unique-timestamp>.log which contains a record of all the text entered at the CASA prompt, and another as casapy-<unique-timestamp>.log which will contain all the messages that are printed to the CASA logger window. It is recommended to keep these log files intact as a reminder of the steps completed in the data reduction. These log files may also be helpful when submitting a helpdesk ticket.
Once CASA has started, a logger window will appear. Note that you can rescale this window or change the font size (under the View menu option) as desired. I personally like to remove clutter columns by moving the mouse to the logger window and press "Ctrl-N", "Ctrl-Y","Ctrl-E", (New Years Eve is a way to remember).
Examining the Measurement Set (MS)
From here on, instead of the G192-BP.ms name in the directory, will be replacing it with the variable name id_ms in this tutorial to make the instructions less dependent on the name of the file so that if another measurement set name is used, only the definition here needs to be edited to the other file name. In order to do that we will define some parameters as variables inside of CASA:
# In CASA visbase = 'G192-BP' (or whatever base name the data set or link to be used has, like <i>my-vis</i> above) id_ms = visbase + '.ms' (this should match and find the measurement set created earlier)
We use listobs to summarize our MS:
# In CASA: listobs on the initial data set
listobs(vis=id_ms, listfile=visbase+'_list.txt')
This will write the output to a file called <visbase>_list.txt, which we can examine using various Unix/Linux print commands such as cat, less, or more or use an editor and keep it on the desktop for frequent consultation with e.g. emacs, vi, or nedit:
# In CASA
cat G192-BP_list.txt
================================================================================
MeasurementSet Name: <path-to-data>/G192-BP.ms MS Version 2
================================================================================
Observer: Dr. Observer Project: uid://evla/pdb/7303457
Observation: EVLA
Data records: 1769355 Total elapsed time = 4563 seconds
Observed from 03-Jan-2013/06:31:48.0 to 03-Jan-2013/07:47:51.0 (UTC)
ObservationID = 0 ArrayID = 0
Date Timerange (UTC) Scan FldId FieldName nRows SpwIds Average Interval(s) ScanIntent
03-Jan-2013/06:31:48.0 - 06:36:42.0 6 0 3C147 704865 [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63] [6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 5.94, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6] [CALIBRATE_FLUX#UNSPECIFIED,OBSERVE_TARGET#UNSPECIFIED]
07:40:27.0 - 07:47:51.0 64 1 3c84-J0319+413 1064490 [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63] [6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6] [CALIBRATE_BANDPASS#UNSPECIFIED,OBSERVE_TARGET#UNSPECIFIED]
(nRows = Total number of rows per scan)
Fields: 2
ID Code Name RA Decl Epoch SrcId nRows
0 E 3C147 05:42:36.137916 +49.51.07.23356 J2000 0 704865
1 F 3c84-J0319+413 03:19:48.160102 +41.30.42.10305 J2000 1 1064490
Spectral Windows: (64 unique spectral windows and 1 unique polarization setups)
SpwID Name #Chans Frame Ch0(MHz) ChanWid(kHz) TotBW(kHz) CtrFreq(MHz) BBC Num Corrs
0 EVLA_KA#A1C1#2 128 TOPO 34476.000 1000.000 128000.0 34539.5000 10 RR LL
1 EVLA_KA#A1C1#3 128 TOPO 34604.000 1000.000 128000.0 34667.5000 10 RR LL
2 EVLA_KA#A1C1#4 128 TOPO 34732.000 1000.000 128000.0 34795.5000 10 RR LL
3 EVLA_KA#A1C1#5 128 TOPO 34860.000 1000.000 128000.0 34923.5000 10 RR LL
<snip>
13 EVLA_KA#A1C1#15 128 TOPO 36140.000 1000.000 128000.0 36203.5000 10 RR LL
14 EVLA_KA#A1C1#16 128 TOPO 36268.000 1000.000 128000.0 36331.5000 10 RR LL
15 EVLA_KA#A1C1#17 128 TOPO 36396.000 1000.000 128000.0 36459.5000 10 RR LL
16 EVLA_KA#A2C2#18 128 TOPO 36476.000 1000.000 128000.0 36539.5000 11 RR LL
17 EVLA_KA#A2C2#19 128 TOPO 36604.000 1000.000 128000.0 36667.5000 11 RR LL
18 EVLA_KA#A2C2#20 128 TOPO 36732.000 1000.000 128000.0 36795.5000 11 RR LL
<snip>
29 EVLA_KA#A2C2#31 128 TOPO 38140.000 1000.000 128000.0 38203.5000 11 RR LL
30 EVLA_KA#A2C2#32 128 TOPO 38268.000 1000.000 128000.0 38331.5000 11 RR LL
31 EVLA_KA#A2C2#33 128 TOPO 38396.000 1000.000 128000.0 38459.5000 11 RR LL
32 EVLA_KA#B1D1#34 128 TOPO 26976.000 1000.000 128000.0 27039.5000 13 RR LL
33 EVLA_KA#B1D1#35 128 TOPO 27104.000 1000.000 128000.0 27167.5000 13 RR LL
34 EVLA_KA#B1D1#36 128 TOPO 27232.000 1000.000 128000.0 27295.5000 13 RR LL
<snip>
45 EVLA_KA#B1D1#47 128 TOPO 28640.000 1000.000 128000.0 28703.5000 13 RR LL
46 EVLA_KA#B1D1#48 128 TOPO 28768.000 1000.000 128000.0 28831.5000 13 RR LL
47 EVLA_KA#B1D1#49 128 TOPO 28896.000 1000.000 128000.0 28959.5000 13 RR LL
48 EVLA_KA#B2D2#50 128 TOPO 28976.000 1000.000 128000.0 29039.5000 14 RR LL
49 EVLA_KA#B2D2#51 128 TOPO 29104.000 1000.000 128000.0 29167.5000 14 RR LL
50 EVLA_KA#B2D2#52 128 TOPO 29232.000 1000.000 128000.0 29295.5000 14 RR LL
<snip>
61 EVLA_KA#B2D2#63 128 TOPO 30640.000 1000.000 128000.0 30703.5000 14 RR LL
62 EVLA_KA#B2D2#64 128 TOPO 30768.000 1000.000 128000.0 30831.5000 14 RR LL
63 EVLA_KA#B2D2#65 128 TOPO 30896.000 1000.000 128000.0 30959.5000 14 RR LL
Sources: 128
ID Name SpwId RestFreq(MHz) SysVel(km/s)
0 3C147 0 - -
0 3C147 1 - -
0 3C147 2 - -
<snip>
0 3C147 61 - -
0 3C147 62 - -
0 3C147 63 - -
1 3c84-J0319+413 0 - -
1 3c84-J0319+413 1 - -
1 3c84-J0319+413 2 - -
<snip>
1 3c84-J0319+413 61 - -
1 3c84-J0319+413 62 - -
1 3c84-J0319+413 63 - -
<snip>
This small MS contains only scans on the flux density calibrator 3C147 (field 0) and the bandpass calibrator 3C84 (field 1). Both fields have 64 spectral windows (spws); each spw is comprised of 128 channels, each channel being 1 MHz wide, for a total bandwidth of 128 MHz / spw.
As the intents for these fields are clearly included in the listing, we now will introduce new variables to recall the general use of these intents without having to remember their details like field ID numbers or name strings - variable names aren't string-values so therefore using them also avoids typing the quotes:
# In CASA id_flx = '3C147' (or '0' for using the field name or field ID respectively) id_bp = '3c84-J0319+413' (or '1')
Setting the Model of the Flux Density Calibrator
To start, we insert the spectral (using the 'Perley-Butler 2017' standard) and spatial (3C147_A.im for Ka-band) models for the flux density calibrator 3C147 (field 0) with the setjy task:
# In CASA: model for the flux density calibrator
setjy(vis=id_ms, field=id_flx, scalebychan=True,
standard='Perley-Butler 2017', model=id_flx+'_A.im')
- scalebychan=True: If scalebychan=False setjy would use a single value per spectral window.
Inspecting the logger report shows that 3C147 has amplitudes ranging from ~1.0–1.47 Jy across all spws.
We can plot the model data using plotms (Figure 1):
# In CASA
plotms(vis=id_ms, field=id_flx, antenna='ea03',
xaxis='freq', yaxis='amp', ydatacolumn='model',coloraxis='ant2')
This plot shows baselines to antenna ea03. Since we provided both a spectral and a spatial model for this well resolved calibrator, each baseline has a somewhat different behavior.
Calibrating delays and initial bandpass solutions
As a first step, we need to specify a reference antenna for all phase calibrations. Since observations were performed with VLA in the most extended configuration A, it is desirable to use an antenna that is near the center of the array, and has valid calibrator data on all baselines with the least amount of data flagged. The array can be mapped with plotants:
# In CASA: plotting antenna locations
plotants(vis=id_ms)
Although the plot is a bit crowded (Figure 2), zooming in (with the magnifying glass icon) shows that ea05 is located close to the center and if we use the flagdata task with mode='summary' it can be seen that ea05 has a comparably small number of flags. Therefore we will use this antenna as our reference and introduce a variable name for it:
# In CASA: id_ref = 'ea05' (or '3' which is the antenna ID number for ea05)
We start with a phase-only, time-dependent calibration solution for the bandpass calibrator. Note that the calibration solutions are written to a calibration table, cal.ign, where 'ign' stands for integration gain. This table is not a variable but a string/name so it needs to be quoted in the input parameter. Solutions for each integration will remove most of the decorrelation of the signal. For best results, we will derive the phase variations from a narrow range of channels (60~68) near the centers of each spws:
# In CASA: phase only calibration
gaincal(vis=id_ms, caltable='cal.ign',
field=id_bp, spw='*:60~68', \
gaintype='G', refant=id_ref, calmode='p',
solint='int', minsnr=3)
- refant=id_ref : Use the variable defined as reference antenna (ea05) as the reference antenna
- solint='int' : Do a per-integration solve (every 6 seconds, since we've time-averaged the data).
- minsnr=3 : Apply a minimum signal-to-noise cutoff. Solutions with less than this value will be flagged.
- gaintable is not set here as we have already applied pre-calibrations.
Plot the phase solutions (using full phase range -180 to 180 instead of autorange):
# In CASA
plotms(vis='cal.ign', xaxis='time', yaxis='phase',
iteraxis='antenna', plotrange=[-1,-1,-180,180])
Note that antenna ea01 is fully flagged. Click on the green arrows to page through the rest of the antennas (also note the reference antenna ea05 has all phases set to 0 deg).
We will now produce multipanel plots of the phase solutions, writing the plots to output files as well as on the screen (Figures 3a, 3b, & 3c). The output files generated are PNG files and can be viewed within CASA by executing an external viewer program, e.g., !xv plotG192_plotcal_G0p1.png; or by running an image viewing application such as xv, Preview, Gimp, Photoshop, etc., external to CASA at the OS level.
# In CASA
plotms(vis='cal.ign', xaxis='time', yaxis='phase',
antenna='0~11', gridrows=3, gridcols=3, iteraxis='antenna',
coloraxis='spw',
plotfile=visbase+'_cal.ign_p1.png', width=1300, height=800)
Note that recently the name of the output file gets appended with the selection of the iteraxis and there unfortunately seems to be no possibility to switch that off; to get the file name as specified in the plotfile-parameter one would edit the file name using the operating system tools.
plotms(vis='cal.ign', xaxis='time', yaxis='phase',
antenna='12~20', gridrows=3, gridcols=3, iteraxis='antenna',
coloraxis='spw',
plotfile=visbase+'_cal.ign_p2.png', width=1300, height=800)
#
plotms(vis='cal.ign', xaxis='time', yaxis='phase',
antenna='21~26', gridrows=2, gridcols=3, iteraxis='antenna',
coloraxis='spw',
plotfile=visbase+'_cal.ign_p3.png', width=1300, height=800)
 |
 |
 |
Here the width=1300 and height=800 parameters specify the dimensions, in pixels, of the output file.
We can now solve for the residual delays using the parameter gaintype='K' option in gaincal and store the solutions in the cal.dly table. Note that this currently does not do a global fringe-fitting solution for delays, but instead does a baseline-based delay solution per spw for all baselines to the reference antenna, treating these as antenna-based delays. In most cases, with high enough S/N to get baseline-based delay solutions, this will suffice. We avoid the edge channels of each spectral window by selecting channels 5~122:
# In CASA: residual delays
gaincal(vis=id_ms, caltable='cal.dly',
field=id_bp, spw='*:5~122', gaintype='K',
gaintable=['cal.ign'],
refant=id_ref, solint='inf', minsnr=3)
Note that we are now applying our initial phase table cal.ign.
Alternatively, you can derive a delay across all spws of a baseband. If this is desired, use parameter combine='spw' in gaincal and run the task for each baseband separately. The solutions from the second and following runs can be appended to the same calibration table via parameter append=True.
Now plot the delays, in nanoseconds, as a function of antenna index (you will get one for each spw and polarization):
# In CASA
plotms(vis='cal.dly', xaxis='ant1', yaxis='delay', coloraxis='spw')
The delays range from around -5 to 4 nanoseconds (Figure 4).
Now solve for the antenna bandpasses using the previously generated tables cal.ign and cal.dly and store the solutions in cal.bp:
# In CASA: antenna bandpasses
bandpass(vis=id_ms, caltable='cal.bp',
gaintable=['cal.ign', 'cal.dly'],
field=id_bp, refant=id_ref, solnorm=False,
bandtype='B', solint='inf')
WARNING: You must set solnorm=False here or later on you will find some offsets among spws due to the way the amplitude scaling adjusts weights internally during solving.
You will see in the terminal window some reports of solutions failing due to "Insufficient unflagged antennas"—note that these are for bad channels that have been pre-flagged.
Plot the resulting bandpasses in amplitude and phase. Note that the first panel with ea01 is empty as it is completely flagged. Proceed to ea06 to see the plots as shown in Figures 5a, 5b, 6a, and 6b:
# In CASA
plotms(vis='cal.bp', xaxis='freq', yaxis='amp',
spw='0~31', iteraxis='antenna', coloraxis='spw')
#
plotms(vis='cal.bp', xaxis='freq', yaxis='amp',
spw='32~63', iteraxis='antenna', coloraxis='spw')
#
plotms(vis='cal.bp', xaxis='freq', yaxis='phase',
spw='0~31', iteraxis='antenna', coloraxis='spw',
plotrange=[-1,-1,-180,180])
#
plotms(vis='cal.bp', xaxis='freq', yaxis='phase',
spw='32~63', iteraxis='antenna', coloraxis='spw',
plotrange=[-1,-1,-180,180])
 |
 |
 |
 |
Bootstrapping the bandpass calibrator spectrum
Since there is no a priori spectral information for our chosen bandpass calibrator of 3C84, we need to bootstrap to find its spectral index, then recalibrate with this information in order to avoid folding the intrinsic spectral shape of 3C84 into our calibration.
First, we again do a phase-only calibration solution, this time for both the bandpass and the flux density calibrator (which need to be separated by a comma) and store the solutions in cal.ign2pha. This will correct for decorrelation of the signals. We again use the channel range 60~68 and apply the bandpass and delay calibration tables:
# In CASA: flux and bandpass calibrators gain
gaincal(vis=id_ms, caltable='cal.ign2pha', field=id_flx+','+id_bp,
gaintable=['cal.dly', 'cal.bp'],
gaintype='G', refant=id_ref, calmode='p', solint='int', minsnr=3)
Now we are ready to solve for both phase and gain for each scan and store the solutions in cal.ign2:
# In CASA: flux and bandpass calibrators gain
gaincal(vis=id_ms, caltable='cal.ign2', field=id_flx+','+id_bp,
gaintable=['cal.dly', 'cal.bp','cal.ign2pha'],
gaintype='G', refant=id_ref, calmode='ap', solint='inf', minsnr=3)
With gain solutions for the flux density and bandpass calibrators, we can now use fluxscale to scale the gain amplitudes of the bandpass calibrator using those of the flux density calibrator and store the solutions in cal.flx2 and the model for the bandpass calibrator in visbase+'_'+id_bp+'_fluxinfo.txt':
# In CASA: bandpass calibrator gain amplitudes scaling
flux1 = fluxscale(vis=id_ms, caltable='cal.ign2',
fluxtable='cal.flx2', reference=id_flx, transfer=id_bp,
listfile=visbase+'_'+id_bp+'_fluxinfo.txt', fitorder=1)
- flux1 = fluxscale(...): the Python dictionary returned by the fluxscale task will be stored in the flux1 variable. Inspect the dictionary flux1 by typing "print flux1" at the CASA command line. Alternatively, when not interested in the dictionary type fluxscale(...); with a semicolon after the parentheses and not assigning it to a variable. This will prevent the dictionary to be printed in the terminal. Here however the dictionary is captured in flux1 as it is going to be used below
- fluxtable='cal.flx2': this is the output scaled gain table. Since we are only using this to find the spectral index of the bandpass calibrator source field, we won't be using this table.
- listfile=visbase+'_'+id_bp+'_fluxinfo.txt': an output file that contains the derived flux values and fit information.
- fitorder=1: only find a (linear) spectral index, ignoring curvature in the spectrum.
- reference=id_flx: the reference field from which the flux scaling is transferred (here: the flux density calibrator 3C147, field 0, assigned by id_flx)
- transfer=id_bp: the target field to which the flux scaling is transferred (here: the bandpass calibrator 3C84, field 1, assigned by id_bp)
The last line in the file (and displayed in the logger) shows:
Fitted spectrum for 3c84-J0319+413 with fitorder=1: Flux density = 29.0286 +/- 0.0308709 (freq=32.5128 GHz) spidx=-0.538803 +/- 0.00883078
Using the information in the returned flux1 dictionary, we can plot the derived spectrum seen in Figure 7; for that the two loops below fill the parameter variables used in the plotting:
(Note, IPython 5.1.0 requires the use of "%cpaste" to allow the copy/paste of loops, conditions, etc.. Press "Enter" (or "Return") after "--" at the end, this stops (or exits) the copy/paste prompt and will return you to the CASA prompt.)
# In CASA
%cpaste
# Press Enter or Return, then copy/paste the following:
freq = flux1['freq'] / 1e9
spw_list = range(0,64)
spw_str = []
for i in spw_list:
thisspw = str(i)
spw_str.append(thisspw)
--
# In CASA
%cpaste
# Press Enter or Return, then copy/paste the following:
bootstrapped_fluxes = []
for j in spw_str:
thisflux = flux1['1'][j]['fluxd'][0]
if thisflux ==None:
continue
else:
bootstrapped_fluxes.append(thisflux)
--
# In CASA - this section creates the plot seen in Figure 7
import matplotlib.pyplot as pl
%cpaste
# Press Enter or Return, then copy/paste the following:
pl.clf()
pl.plot(freq, bootstrapped_fluxes, 'bo')
pl.xlabel('Frequency (GHz)')
pl.ylabel('Flux Density (Jy)')
pl.title(id_bp)
pl.show()
--
We can use the model from fluxscale to fill the MODEL column with the banpass source's spectral information using setjy. With standard='fluxscale', we can directly use the flux1 Python dictionary as input via fluxdict:
# In CASA: spectral information
setjy(vis=id_ms, field=id_bp, scalebychan=True,
standard = 'fluxscale', fluxdict=flux1);
Check with plotms that the data have been appropriately filled (Figure 8):
# In CASA
plotms(vis=id_ms, field=id_bp, antenna='ea05&ea02',
xaxis='freq', yaxis='amp', ydatacolumn='model')
Next, we redo the previous calibration using this new model information. Although the commands are the same as issued earlier, keep in mind that the model values for the bandpass calibrator have changed and, therefore, the results of these calibration calculations will differ:
# In CASA: phase only recalibration
gaincal(vis=id_ms, caltable='cal.ign.redo',
field=id_bp, spw='*:60~68',
gaintype='G', refant=id_ref, calmode='p',
solint='int', minsnr=3)
# In CASA: residual delays recalibration
gaincal(vis=id_ms, caltable='cal.dly.redo',
gaintable=['cal.ign.redo'],
field=id_bp, spw='*:5~122', gaintype='K',
refant=id_ref, solint='inf', minsnr=3)
# In CASA: antenna bandpasses recalibration
bandpass(vis=id_ms, caltable='cal.bp.redo',
gaintable=['cal.ign.redo', 'cal.dly.redo'],
field=id_bp, refant=id_ref, solnorm=False,
bandtype='B', solint='inf')
Finally, we inspect these solutions (Figures 9a, 9b, 10a, and 10b; again note that antenna ea01 is fully flagged):
# In CASA - Figure 9a
plotms(vis='cal.bp.redo', xaxis='freq', yaxis='amp',
spw='0~31', iteraxis='antenna', coloraxis='spw')
#
plotms(vis='cal.bp.redo', xaxis='freq', yaxis='amp',
spw='32~63', iteraxis='antenna', coloraxis='spw')
#
plotms(vis='cal.bp.redo', xaxis='freq', yaxis='phase',
spw='0~31', iteraxis='antenna', coloraxis='spw',
plotrange=[-1,-1,-180,180])
#
plotms(vis='cal.bp.redo', xaxis='freq', yaxis='phase',
spw='32~63', iteraxis='antenna', coloraxis='spw',
plotrange=[-1,-1,-180,180])
 |
 |
 |
 |
They look virtually unchanged from the previous solutions with the exception that the amplitude scaling is corrected for the spectrum of the bandpass calibrator. We have the final version of our delay and bandpass calibration tables, cal.dly.redo and cal.bp.redo, which can be used for all subsequent calibration steps.
Last checked on CASA Version 6.4.1
