Here's a question from Akshaya Rane:
I am doing polarization calibration and while trying to run 'pcal', I have encountered the following error. Does this has to do something with qualifiers?
PCAL 3: TOTAL INTENSITIES MUST BE PROVIDED
PCAL 3: Purports to die of UNNATURAL causes
Any thoughts?
I should add that Eric Greisen says that if you spend an hour trying to figure out something in AIPS, and you can't, you should feel free to email him and ask his help.
Showing posts with label calibration. Show all posts
Showing posts with label calibration. Show all posts
Wednesday, February 17, 2010
Error in PCAL
Saturday, May 23, 2009
Channels and CALIB
When one is self-calibrating a multi-channel data set, and one gives CALIB a BCHAN and ECHAN value, what is CALIB doing with the channels? Is it averaging them all before it calibrates?
If it is doing that, then why bother making a Channel 0 data set to calibrate off of? Simply because CALIB runs much faster if it doesn't have to deal with multiple channels? Is there something wrong with using a multi-channel data set to do the initial calibration (on the phase cal + flux cal sources?)
Saturday, March 7, 2009
More Missing Flux Cals
I'm reducing archival data, and a flux calibrator was unfortunately not taken at the beginning and the end of the observation. It was only taken at the beginning, and when I read the data in with FILLM, FILLM says that the antennas moved between my flux cal and my phase cal/data. FILLM then puts the flux cal in a different UV file from the data/phase cal.
I'm thinking the antennas couldn't have moved THAT much over 15 minutes (and all i want is a flux anyway), so I'd still like to use the flux cal (also there are no appropriate recent fluxes for the phase cal in any of the NRAO databases that Claudia suggested.) Is there a way to
a) apply a flux calibration solution from another uv file?
or
b) relax FILLM's requirements so that it will not be so sensitive to moving antennas and put everything in one file?
or
c) append a source to another UV file? I know uvglu appends in frequency, but i don't think it appends sources.
Thanks!
Monday, September 15, 2008
Flux calibration scales at the VLA?
I've run into what looks like a systematic offset between flux scales of earlier observations and my observations. I'm using the 1999.2 coefficients to set my flux scale and it looks like the earlier observations are using the coefficients from Baars et al. (1977). Any ideas on what the differences are between these two systems? 5%? 10%? 20%? I've tried searching the VLA calibration documentation for the information, but didn't find anything useful (although maybe I missed something).
Tuesday, June 17, 2008
Calibrating Data without a Flux Calibrator
Sometimes when I get data from the archive, it appears that the observers have not observed a flux calibrator. There is a phase calibrator. Is it possible to use this data? Is the data really only good for imaging and getting the morphology of an object, but you can not trust it for absolute fluxes? I feel like I should be able to do something clever to get fluxes...
Monday, June 2, 2008
Calibrating One Calibrator with Two UVranges
The VLA calibrator manual recommends that if you are using 3c286 as your flux calibrator at 20 cm in A configuration, you should use the uvrange 0-18 AND 90-180 (for the latter, decrease flux by 6%). This makes me wonder how on earth one would do this. First of all, I do not know how to calibrate on two separate uvranges and then merge that information. Second of all, I do not know how to reduce a flux by 6%. Anyone? Anyone?
Friday, January 4, 2008
Double Sources-- and not the AGN kind
I'm having this problem with some archival data from 1990. I have two nights of data on a galaxy, spaced a few nights apart from one another. I calibrate the nights separately, and image them separately. It rapidly becomes obvious that the images from the two nights are shifted relative to one another, but not in any simple way. The galaxy center, at the image center, is at the same position in both, but as you go out radially, the first night's sources are further out, radially. It's as if the image is stretched one night, relative to the other.
AIPS thinks that both images have the same pixel scale, so that means a source will have different R.A. and Dec (by about an arcminute, pretty bad!) between the two nights' images. If I DBCON the two nights, I get lots of double sources.
The observational setup claims to be exactly the same between the two nights-- same pointing center, same correlator set-up.
Anyone have any idea what might be causing this? I think I've had a similar problem with GMRT data before, and gave up temporarily and went on to a different project. I'm guessing it's a calibration issue?
Tuesday, June 19, 2007
Where did R go?
Ok, this is kind of a strange thing. I'm working with Pband data. I've gone through initial flagging, and then I run CALIB. But for some reason it's not finding my R data... so I get 1/2 good solutions and 1/2 failed solutions (because it finds the L just fine). But I look at the data (and when I flagged it too) and it's *there*. Has anyone encountered this? What's going on and how to I fix it? I'm using pretty standard inputs...
Read more!Monday, June 4, 2007
WBR Step 6L: Calibration
Calibration is usually a somewhat iterative process. I will lightly flag my calibrator, then calibrate the calibrator, then see if any new funky looking data points have appeared. If they have, I will flag again and calibrate again. Wash, rinse, repeat until no funky data points show up.
So how do you calibrate your data anyway?
Assuming you've lightly flagged your phase and flux calibrators, according to Step 5, now we're ready to run SETJY. SETJY basically just writes to your data the most accurate flux for your flux calibrator. Important parameters to consider here:
--SOURCES '1331+305', '0137+331', or '0542+498' (whichever one you used. sometimes you will have two different flux calibrators, and you can enter them both here.
--OPTYPE 'CALC'
AIPS 1: SETJY Task to enter source info into source (SU) table.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: INNAME 'AC825_FQ1' Input image name (name)
AIPS 1: INCLASS 'CH 0' Input image name (class)
AIPS 1: INSEQ 1 Input image name (seq. #)
AIPS 1: INDISK 1 Input image disk unit #
AIPS 1: SOURCES '1331+305' Sources to modify.
AIPS 1: *rest ' '
AIPS 1: QUAL -1 Source qualifier -1=>all
AIPS 1: BIF 0 Low IF # for flux density
AIPS 1: EIF 0 High IF # for flux density
AIPS 1: ZEROSP *all 0 I,Q,U,V flux density (Jy)
AIPS 1: OPTYPE 'CALC' ' ' => use other adverbs
AIPS 1: for required operation
AIPS 1: 'CALC' => determine
AIPS 1: 3C286/3C48/1934 fluxes from
AIPS 1: standard formulae
AIPS 1: 'REJY' => reset source
AIPS 1: fluxes to zero.
AIPS 1: 'REVL' => reset velocity
AIPS 1: to zero
AIPS 1: 'RESE' => reset fluxes &
AIPS 1: velocities to zero.
AIPS 1: CALCODE ' ' New calibrator code:
AIPS 1: '----' => change to blank
AIPS 1: SYSVEL 0 Velocity of source (km/s)
AIPS 1: RESTFREQ 0 0 Line rest frequency (Hz)
AIPS 1: VELTYP ' ' Velocity type 'LSR,'HELIO'
AIPS 1: VELDEF ' ' Velocity definition 'RADIO',
AIPS 1: 'OPTICAL'
AIPS 1: FREQID 0 FQ table entry to use for
AIPS 1: velocity information and
AIPS 1: 'CALC' option
AIPS 1: APARM *all 0 (1): Pixel to which SYSVEL
AIPS 1: refers ( 0=>1)
AIPS 1: (2): Only for 'CALC' option:
AIPS 1: <= 0 => use latest VLA
AIPS 1: values (1999.2) or,
AIPS 1: for 1934-638, the
AIPS 1: ATCA value of 30Jul94.
AIPS 1: 1 => use Baars values
AIPS 1: or old ATCA/PKS values
AIPS 1: for 1934-638
AIPS 1: 2 => use VLA 1995.2
AIPS 1: values or for 1934-638
AIPS 1: the ATCA value of
AIPS 1: 30Jul94.
AIPS 1: >= 3 => use oldest VLA
AIPS 1: values (1990) or,
AIPS 1: for 1934-638, the
AIPS 1: ATCA value of 30Jul94.
AIPS 1: (3): Only for 'CALC' option:
AIPS 1: multiply the calculated
AIPS 1: fluxes by APARM(3) with
AIPS 1: 0 -> 1
Now it's time to run calib!
(I wouldn't use VLACALIB (like the AIPS cookbook tells you too) because some parts of VLACALIB are a bit of a black box, and can set certain parameters quite mysteriously (I've had it use SOLINT = 400 before, for no good reason).
Run calib on your flux calibrator first. Here are some parameters you should definitely worry about and set:
CALSOUR '1331+305' (name of flux calibrator)
UVRANGE (find this in the VLA calibrator manual. the best calibrators are ones where you don't have to set this.)
DOCALIB 2 (apply weights)
REFANT 6 (you chose a refant in step 3, use it!)
SOLINT 1 (this is how often to solve for a calibration solution. 1 seems to work well often.)
SOLMODE 'A&P' (calibrate amplitudes and phases)
AIPS 1: CALIB: Task to determine calibration for data.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: Input uv data.
AIPS 1: INNAME 'AC825_FQ1' UV file name (name)
AIPS 1: INCLASS 'CH 0' UV file name (class)
AIPS 1: INSEQ 1 UV file name (seq. #)
AIPS 1: INDISK 1 UV file disk drive #
AIPS 1: Data selection (multisource):
AIPS 1: CALSOUR '1331+305' Calibrator sources
AIPS 1: *rest ' '
AIPS 1: QUAL -1 Calibrator qualifier -1=>all
AIPS 1: CALCODE ' ' Calibrator code ' '=>all
AIPS 1: SELBAND -1 Bandwidth to select (kHz)
AIPS 1: SELFREQ -1 Frequency to select (MHz)
AIPS 1: FREQID 1 Freq. ID to select.
AIPS 1: TIMERANG *all 0 Time range to use.
AIPS 1: BCHAN 0 Lowest channel number 0=>all
AIPS 1: ECHAN 0 Highest channel number
AIPS 1: ANTENNAS *all 0 Antennas to select. 0=all
AIPS 1: DOFIT *all 0 Subset of ANTENNAS list for
AIPS 1: which solns are desired.
AIPS 1: ANTUSE *all 0 Mean gain is calculated
AIPS 1: (CPARM(2)>0) using only the
AIPS 1: listed antennas. See explain.
AIPS 1: SUBARRAY 0 Subarray, 0=>all
AIPS 1: UVRANGE 0 0 Range of uv distance for full
AIPS 1: weight
AIPS 1: WTUV 0 Weight outside UVRANGE 0=0.
AIPS 1: WEIGHTIT 0 Modify data weights function
AIPS 1: Cal. info for input:
AIPS 1: DOCALIB 2 If >0 calibrate data
AIPS 1: = 2 calibrate weights
AIPS 1: GAINUSE 0 CL table to apply.
AIPS 1: DOPOL -1 If >0 correct polarization.
AIPS 1: BLVER -1 BL table to apply.
AIPS 1: FLAGVER 1 Flag table version
AIPS 1: DOBAND -1 If >0 apply bandpass cal.
AIPS 1: Method used depends on value
AIPS 1: of DOBAND (see HELP file).
AIPS 1: BPVER 1 Bandpass table version
AIPS 1: SMOOTH *all 0 Smoothing function. See
AIPS 1: HELP SMOOTH for details.
AIPS 1:
AIPS 1: CLEAN map. See HELP.
AIPS 1: IN2NAME ' ' Cleaned map name (name)
AIPS 1: IN2CLASS ' ' Cleaned map name (class)
AIPS 1: IN2SEQ 0 Cleaned map name (seq. #)
AIPS 1: IN2DISK 0 Cleaned map disk unit #
AIPS 1: INVERS 0 CC file version #.
AIPS 1: NCOMP *all 0 # comps to use for model.
AIPS 1: 1 value per field
AIPS 1: FLUX 0 Lowest CC component used.
AIPS 1: NMAPS 0 No. Clean map files
AIPS 1: CMETHOD ' ' Modeling method:
AIPS 1: 'DFT','GRID',' '
AIPS 1: CMODEL ' ' Model type: 'COMP','IMAG'
AIPS 1: SMODEL *all 0 Source model, 1=flux,2=x,3=y
AIPS 1: See HELP SMODEL for models.
AIPS 1:
AIPS 1: Output uv data file.
AIPS 1: OUTNAME ' ' UV file name (name)
AIPS 1: OUTCLASS ' ' UV file name (class)
AIPS 1: OUTSEQ 0 UV file name (seq. #)
AIPS 1: OUTDISK 0 UV file disk drive #
AIPS 1:
AIPS 1: Solution control adverbs:
AIPS 1: REFANT 6 Reference antenna
AIPS 1: SOLINT 1 Solution interval (min)
AIPS 1: SOLSUB 0 Solution subinterval
AIPS 1: SOLMIN 0 Min solution interval
AIPS 1: APARM *all 0 General parameters
AIPS 1: 1=min. no. antennas
AIPS 1: 2 > 0 => data divided
AIPS 1: 3 > 0 => avg. RR,LL
AIPS 1: 5 > 0 => avg. IFs.
AIPS 1: 6=print level, 1=good,
AIPS 1: 2 closure, 3 SNR
AIPS 1: 7=SNR cutoff (0=>5)
AIPS 1: 8=max. ant. # (no AN)
AIPS 1: 9 > 0 => pass failed soln
AIPS 1: Phase-amplitude Parameters:
AIPS 1: DOFLAG 0 Flag on closure error?
AIPS 1: SOLTYPE ' ' Soln type,' ','L1','GCON',
AIPS 1: 'R', 'L1R', 'GCOR'
AIPS 1: SOLMODE 'A&P' Soln. mode: 'A&P','P','P!A',
AIPS 1: 'GCON',
AIPS 1: SOLCON 0 Gain constraint factor.
AIPS 1: MINAMPER 0 Amplitude closure error
AIPS 1: regarded as excessive in %
AIPS 1: MINPHSER 0 Phase closure error regarded
AIPS 1: as excessive in degrees
AIPS 1: CPARM *all 0 Phase-amp. parameters
AIPS 1: 1 = Min el for gain
AIPS 1: normalization (deg)
AIPS 1: 2 >0 => normalize gain
AIPS 1: 3 avg. amp. closure err
AIPS 1: 4 avg. ph. closure err
AIPS 1: 5 >0 => scalar average
AIPS 1: 6 limit clipping in robust
AIPS 1: SNVER 0 Output SN table, 0=>new table
AIPS 1: ANTWT *all 0 Ant. weights (0=>1.0)
AIPS 1: GAINERR *all 0 Std. Dev. of antenna gains.
AIPS 1: BADDISK *all 0 Disk no. not to use for
AIPS 1: scratch files.
You will get some output that looks like this:
localh> CALIB1: Writing SN table 1
localh> CALIB1: RPOL, IF= 1 The average gain over these antennas is 3.080E+00
localh> CALIB1: RPOL, IF= 2 The average gain over these antennas is 3.088E+00
localh> CALIB1: LPOL, IF= 1 The average gain over these antennas is 3.033E+00
localh> CALIB1: LPOL, IF= 2 The average gain over these antennas is 3.086E+00
localh> CALIB1: Found 2321 good solutions
localh> CALIB1: Failed on 7 solutions
Some ways to tell if your data is really acting up is if you have tons of bad solutions (if you do, you might want to consider increasing your solint). Additionally, for some telescopes (not really for the VLA), if you have lots of bad solutions, you might have to tell CALIB to accept points of lower quality.
aparm(1) = 4 (CALIB default is a minimum of 6 antennas for a good solution, you could decrease it to 4.)
aparm(7) = 3 (CALIB default is a signal-to-noise in your solutions of 5. This is a bit high for some telescopes, and can reasonably be decreased to 3.)
soltype = 'L1' (this should be a more robust method for finding solutions (robust against outlying points). However, AIPS wants you to know that you will "lose some statistical efficiency". You could try other robust soltypes, too, but be warned they might be slow.)
Additionally, in my experience, it's a good thing that those 4 numbers (one for each polarization and IF) are all about the same (around 3.08). If one is really different than the others, than it might imply that their is bad data in that if/polarization.
All CALIB does is spit out a 'SN' table. The way we have it right now (SNVER = 0), every time you run CALIB it will spit out a new SN table. If you set SNVER = 1, every time you run CALIB, it would just append new information to this table.
Let's see if this calibration brought up any new bad data that needs to be flagged. If we want to apply calibration to our data in tasks like UVPLT and TVFLG, the calibration needs to be in a 'CL' table, not a "SN' table (yeah, stupid AIPS rule). To copy our 'SN' table to a 'CL' table, use VLACLCAL. First, you will need to type:
run vlaprocs
to load up the task VLACLCAL. Then set three important parameters. In this case, we'll use:
SOURCES = '1331+305' (the name of your flux cal)
CALSOUR = '1331+305' (the name of your flux cal)
REFANT = 6 (your reference antenna)
AIPS 1: VLACLCAL Procedure to run CLCAL and LISTR for VLA data.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: Use RUN VLAPROCS first
AIPS 1: INNAME 'AC825_FQ1' Input UV file name (name)
AIPS 1: INCLASS 'CH 0' Input UV file name (class)
AIPS 1: INSEQ 1 Input UV file name (seq. #)
AIPS 1: INDISK 1 Input UV file disk unit #
AIPS 1: SOURCES '1331+305' Source list to calibrate
AIPS 1: *rest ' '
AIPS 1: SOUCODE ' ' Source "Cal codes"
AIPS 1: CALSOUR '1331+305' Cal sources for calibration
AIPS 1: *rest ' '
AIPS 1: QUAL -1 Source qualifier -1=>all
AIPS 1: CALCODE ' ' Calibrator code ' '=>all
AIPS 1: FREQID 1 Unique frequency code
AIPS 1: TIMERANG *all 0 Time range to calibrate
AIPS 1: SUBARRAY 0 Subarray, 0=>all
AIPS 1: INTERPOL ' ' Interpolation function
AIPS 1: SAMPTYPE ' ' Smoothing function
AIPS 1: DOBLANK 0 Blanked value interpolation
AIPS 1: DOBTWEEN 1 > 0 -> smooth all sources
AIPS 1: together; else separate them
AIPS 1: SMOTYPE ' ' Data to smooth
AIPS 1: BPARM *all 0 Smoothing parameters
AIPS 1: ICUT .1 Cutoff for functional forms
AIPS 1: GAINVER 0 Input Cal table 0=>1
AIPS 1: GAINUSE 0 Output CAL table 0=>2
AIPS 1: REFANT 6 Reference antenna 0=>pick.
AIPS 1: DOPRINT 1 >0 Print messages to a file
AIPS 1: or to the printer.
AIPS 1: OUTPRINT *all ' ' Printer disk file to save
AIPS 1: BADDISK *all 0 Disks to avoid for scratch
This should write out a CL table 2. If you want to apply this caibration in, say, TVFLG, you would just use
DOCAL = 2
GAINUSE = 2
and that applies the CL table 2.
At this point, I also like to flag on phase. I've had a few data sets where amplitudes looks ok for the calibrators, but there are still some quite aberrant points in phase. In general, any points with phases > +/-10 degrees are probably bad (for your calibrators, which are point sources!!! unfortunately, you won't be able to flag on phase at all for your target source, because your target source will not be a point source and will therefore have phases running the whole gamut from -180 to +180 degrees.)
If you want to 'undo' the calibration (let's say because you flagged some data, and now you want to do a new and improved calibration and no longer need your old one), just delete your SN tables, and all CL tables which are greater than 1. BUT DON'T DELETE CL TABLE 1!!! This has important basic information in it.
When you think you have the very best calibration for your flux calibrator, I like to then repeat this same process for my phase calibrator. Iterate by running CALIB and VLACLCAL on your phase calibrator, and then flagging it until in uvplt there are very no aberrant points left in amplitude or phase. (Make sure you check each IF and each polarization!)
Now, when you trust your calibration/flagging for both your calibrators, Delete your SN tables and CL table #2. Let's do final runs of CALIB. At this point I set SNVER = 1 so that CALIB will write out solutions for both calibrators to the same SN table.
--Run CALIB on your flux calibrator.
--Run CALIB on your phase calibrator.
Now, run GETJY to get a flux for your phase calibrator. Parametrs to set:
SOURCES = your phase calibrator
CALSOUR = your flux calibrator
AIPS 1: GETJY Task to determine source flux densities.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: INNAME 'AC825' Input UV file name (name)
AIPS 1: INCLASS 'CH 0' Input UV file name (class)
AIPS 1: INSEQ 1 Input UV file name (seq. #)
AIPS 1: INDISK 1 Input UV file disk unit #
AIPS 1: SOURCES '1033+395' Source list to find fluxes
AIPS 1: *rest ' '
AIPS 1: SOUCODE ' ' Source "Cal codes"
AIPS 1: CALSOUR '1331+305' Cal sources for calibration
AIPS 1: *rest ' '
AIPS 1: QUAL -1 Source qualifier -1=>all
AIPS 1: CALCODE ' ' Calibrator code ' '=>all
AIPS 1: BIF 0 Lowest IF number 0=1
AIPS 1: EIF 0 Highest IF number
AIPS 1: TIMERANG *all 0 Time range of solutions.
AIPS 1: ANTENNAS *all 0 Antennas to use
AIPS 1: SUBARRAY 0 Subarray, 0=>all
AIPS 1: SELBAND -1 Bandwidth to select (kHz)
AIPS 1: SELFREQ -1 Frequency to select (MHz)
AIPS 1: FREQID 2 Freq. ID to select.
AIPS 1: SNVER 2 Input SN table, 0=>all.
You will get some output that looks like this:
localh> GETJY1: Task GETJY (release of 31DEC05) begins
localh> GETJY1: Source:Qual CALCODE IF Flux (Jy)
localh> GETJY1: 1033+395 : 0 A 1 .40437 +/- .00091
localh> GETJY1: 2 .39109 +/- .00074
localh> GETJY1: Appears to have ended successfully
localh> GETJY1: localhost 31DEC05 TST: Cpu= .0 Real= 0
At this point, what you want to worry about is the errors in the fluxes determined for your phase cal. If they are big, something has gone awry. What you do NOT need to worry about is if the fluxes are significantly different from those in the VLA calibrator manual. Remember that these calibrators are AGN-- they vary in flux!! Flux calibrators have been chosen because their fluxes are hopefully quite steady, but phase calibrators have only been chosen to be pointy. Their fluxes may vary. That's why you are comparing them with the flux calibrator, to get the best flux determination for them at the time of your observations!
A fun exercise is to look up your phase calibrator using this Java applet. It will show you how the flux of your calibrator has changed over time (their are only data points here when someone has actually used your phase calibrator for their VLA observations).
Finally, do final runs of VLACLCAL.
Do one run with
SOURCES = your phase cal, your target sources
CALSOUR = your phase cal
Do one more run with
SOURCES = your flux cal
CALSOUR = your flux cal
The defaults on VLACLCAL are smart enough to always output to CL table #2 (even if it already exists).
A good way to do a final check of your calibration (although it should be quite good if you have done it the iterative way I suggest!) is to run LISTR in 'MATX' mode.
OPTYPE = 'MATX'
SOURCES = your phase cal
DOCAL = 2
GAINUSE = 2
DPARM(1) = 5
DOCRT = -1
OUTPRINT = a file name
AIPS 1: LISTR: Task to print UV data and calibration tables.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: USERID 0 User number.
AIPS 1: INNAME 'AC825' UV data (name).
AIPS 1: INCLASS 'CH 0' UV data (class).
AIPS 1: INSEQ 1 UV data (seq. #). 0 => high
AIPS 1: INDISK 1 Disk unit #. 0 => any
AIPS 1: OPTYPE 'MATX' List type:
AIPS 1: 'MATX','LIST','GAIN','SCAN'
AIPS 1: INEXT 'CL' CL, SN or TY table for 'GAIN'
AIPS 1: INVER 0 CL, Sn or TY table version
AIPS 1: SOURCES '1033+395' Source list
AIPS 1: *rest ' '
AIPS 1: CALCODE ' ' Calibrator code ' '=>all
AIPS 1: TIMERANG *all 0 Time range to list
AIPS 1: STOKES ' ' Stokes type to list.
AIPS 1: SELBAND -1 Bandwidth to select (kHz)
AIPS 1: SELFREQ -1 Frequency to select (MHz)
AIPS 1: FREQID 0 Freq. ID to select.
AIPS 1: None selected => 1.
AIPS 1: BIF 1 Lowest IF number 0=1
AIPS 1: EIF 1 Highest IF number
AIPS 1: BCHAN 1 Low channel number 0=>1
AIPS 1: ECHAN 0 High channel number
AIPS 1: ANTENNAS *all 0 Antennas to list
AIPS 1: BASELINE *all 0 Baselines with ANTENNAS
AIPS 1: UVRANGE 0 0 UV range in kilolambda
AIPS 1: SUBARRAY 0 Subarray, 0=>1
AIPS 1: Cal. info for input:
AIPS 1: DOCALIB 2 If >0 calibrate data
AIPS 1: = 2 calibrate weights
AIPS 1: GAINUSE 2 CAL (CL or SN) table to apply
AIPS 1: DOPOL -1 If >0 correct polarization.
AIPS 1: BLVER -1 BL table to apply.
AIPS 1: FLAGVER 0 Flag table version
AIPS 1: DOBAND -1 If >0 apply bandpass cal.
AIPS 1: Method used depends on value
AIPS 1: of DOBAND (see HELP file).
AIPS 1: BPVER -1 Bandpass table version
AIPS 1: SMOOTH *all 0 Smoothing function. See
AIPS 1: HELP SMOOTH for details.
AIPS 1: DPARM 5 1 Control info:
AIPS 1: *rest 0 (1) 0=amp, 1=phase, 2=rms
AIPS 1: 3=amp+rms, 4=phase+rms
AIPS 1: 5=amp+phase, 6=delay,
AIPS 1: 7=rate, 8=SNR, 9=par. ang
AIPS 1: 10 = Tsys, 11=elev,
AIPS 1: 12 = multiband delay
AIPS 1: 13 = Tant 14 = weights
AIPS 1: N.B. Options 3 and 4 only
AIPS 1: valid for OPTYPE = 'MATX'
AIPS 1: (2) 0=vec avg, 1=scalar
AIPS 1: (3) No. col (4-10) def=4.
AIPS 1: (4) avg. time min (MATX,LIST)
AIPS 1: *** significant
AIPS 1: (5) Type of gain listings:
AIPS 1: 0 => all with same scale
AIPS 1: like DEC-10 listings;
AIPS 1: 1 => scan/source info at
AIPS 1: source/scan boundaries.
AIPS 1: (6) Matrix scaling control:
AIPS 1: 0 => plot amp & rms
AIPS 1: matrices with same scale;
AIPS 1: 1 => self-scale each
AIPS 1: separately.
AIPS 1: (7) > 0 self-scale phase
AIPS 1: displays, else degrees.
AIPS 1: DOACOR -1 > 0 include autocorrelations
AIPS 1: FACTOR 0 When DPARM(5)=0 Multiply
AIPS 1: gain listings by FACTOR,
AIPS 1: 0=> use scale factor from
AIPS 1: first record printed.
AIPS 1: DOCRT -1 > 0 -> use the terminal,
AIPS 1: else use the line printer
AIPS 1: > 72 => terminal width
AIPS 1: OUTPRINT 'MATX_CAL1'
AIPS 1: Printer disk file to save
AIPS 1: BADDISK *all 0 Disk to avoid for scratch.
When you look at this list of numbers, amplitudes should be near constant between baselines and in time. Phases should be around zero. Lots of 0's and 1's are good! Teens and 20's are bad! If you have nice phases that are mostly small numbers, then...
Congratulations! You have calibrated your data!!!
Sunday, May 27, 2007
WBR Step 4. One File per Each Day of Observations
Sometimes, your data is taken over multiple days. You want to calibrate each day separately. You can do this by setting 'timerang' in calib, but that scares me a little bit and has lots of room for mistakes. What I prefer to do is SPLIT off each day of observing, so that you'll have a UV file for each day of data. After you've calibrated each file individually, you can combine all the data back together using DBCON.
So how do you split your data according to observation day?
Using SPLIT!
Well, first of all, check in that scan list you got from LISTR to see what days you observed on.
Then run split with 'timerang' set to only include one of those days. For example, only include data taken on day 1:
timerang = 1 0 0 0 1 23 0 0
will include all data take on day 1 between 0 hr and 23 hr.
Everything else in SPLIT you will probably want to leave as default. Except for maybe 'outclass', which you could change to something like, say, 'line_1' (so that you remenber this is the line data from day 1).
SPLIT will generate a new UV file. You need to run INDXR on it to generate a new NX table. The defaults of INDXR are just fine for these purposes.
Unfortunately, you are going to have to do each of the following flagging/calibration steps for each file individually. Sorry, dudes.
Monday, May 21, 2007
UVRANGEs contradict Calibrator Quality Ratings
What does it mean if your calibrator has a good ("P") rating in the calibrator manual for your array configuration whose longest baseline is ~50 kilolambda--- but then there's a uvrange limitation on it with a uvmin of 100 kilolambda? I'm trying to use calibrator 0841+708 at C band in C array.
5/23/07:
Resolution: After talking to Miller Goss, I think that this must be a typo in the manual. The amplitudes look fine, for the most part, and there is no obvious confusing source in the field. So, I guess this is just a word of caution about the manual, that it can make mistakes some times!
5/24/07:
More Resolution: Rick Perley had this to say about the current state of the VLA calibrator manual:
Right you are! The 'P' in that column/row should be an 'S', I think. It's one of the 'core-dominated' sourceRead more!
types that make good high frequency, high resolution calibrators.
Nobody is keeping up the manual these days (Greg Taylor used to do this, but he's now gone to UNM). We'll
certainly get back on this after the EVLA settles in, and we'll be able to utilize far more sources as calibrators.
In the meantime, I'll make a note in my copy.
Thursday, May 17, 2007
WBR Step 3: Choosing a Reference Antenna
At several steps in your data calibration, AIPS is gonna ask you for a REFANT, a reference antenna. So we might as well choose one now!
First of all, run PRTAN to view your antenna setup. Really the only thing you need to do here is tell PRTAN to write to file (DOCRT = -1) and give it a file name (in OUTPRINT)
>tget prtan
>inp
AIPS 1: PRTAN: Task to print the Antenna (AN) extension of a uv file.
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: USERID 0 Image owner ID number
AIPS 1: INNAME 'MULTI4.8' Image name (name)
AIPS 1: INCLASS 'CH0' Image name (class)
AIPS 1: INSEQ 1 Image name (seq. #)
AIPS 1: INDISK 1 Disk drive #
AIPS 1: INVERS 0 AN file ver. #
AIPS 1: NPRINT 0 No. records to print 0 => all
AIPS 1: DOCRT -1 > 0 => use terminal instead
AIPS 1: > 72 => terminal width
AIPS 1: OUTPRINT 'ANT.TXT'
AIPS 1: Printer disk file to save
In the outputted file, you'll get a cute little map of your antenna locations that looks like this:
Location Of VLA Antennas
N18 ( 1)
N16 (26)
N14 (24)
N12 ( 5)
N8 ( 9)
N4 ( 7)
N2 (27)
N1 (25)
(20) W2 E2 ( 3)
( 8) W4 E4 (15)
(23) W6 E6 (12)
( 6) W8 E8 (10)
( ) E10 ( 2)
(21) W12 E12 (17)
( 4) W14 ( )
(11) W16 E16 (19)
(22) W18 E18 (28)
VLA:OUT (13)
VLA:OUT (14)
VLA:OUT (16)
VLA:OUT (18)
VPT:OUT (29)
(Yeah, this is data during the EVLA upgrade so there are quite a few dishes missing). For your reference antenna, you are going to want to pick a dish near the center of the array, but not exactly at the center (the centermost ones suffer most from RFI). You are also going to want to choose a dish that never crashed during the course of your observations. A good way to see what dishes went down during your observations is to download the observing logs. See if there are any notes from the telescope operator about the dish you are considering using as your REFANT.
Finally, you can take a quick look at the amplitudes of the data using the 'MATX' mode of LISTR. This will spit out amplitudes of your data (on each baseline) averaged over a given time increment. Sometimes you will see really anomalous amplitudes for a given antenna over some timerange (like, an order of magnitude off). If you see something really fishy for the dish you are considering for your REFANT, don't use it. Find a different REFANT that does not have crazy amplitudes.
So, to create such a file of amplitudes, run LISTR setting these parameters:
OPTYPE = 'MATX'
DOCAL = 2 (apply weights)
DPARM(1) = 0 (list amplitudes)
DPARM(4) = 0 (will list amplitudes averaged over the entire scan. You can change this if you want.)
DOCRT = -1 (write to file)
OUTPRINT = 'CHK_ANT.TXT'
you could also specify a source, say your Phase Calibrator, if you want.
If an antenna passes all of these tests, than it's probably a decent REFANT!
Wednesday, May 16, 2007
Help with continuum subtraction?
I have 23GHz spectral line data with a weak continuum source, and need help making the continuum go away so I can measure the strength of the line emission... for a couple subtle reasons, UVLSF won't do the trick in this case. Any ideas for other ways to remove continuum? More details below.
The source itself is a weak point source. The problem with UVLSF is that spectral line structure is found across much of the spectral window, making it really hard to find line-free channels to average together to use as a model for the continuum. UVSUB uses clean component models, and honestly, I'm not sure how to use it or indeed if this is really what I want.
An additional twist-- I have a higher S/N image of the continuum source taken in an image of the same field at a slightly different central freq (~1.5 GHz offset). I have a feeling I can use this continuum emission as a better S/N image to subtract from the original image, but I'm not sure how to do this...
Ideas? Thanks...
Monday, April 2, 2007
Using Calibrator Models
Wisdom from Stefanie Muehle:
BTW, I now got a nice calibration of my spectal line data using the models as recommended in the AIPS cookbook. The thing you have to get your head around is that when you apply a model, you don't _expect_ the phases to be zero, because you are dropping the default assumption that the calibrator is a point source. The clean components and the visibility plots are not much help in checking whether your calibration is ok, either. So, here's the tip from Michael Rupen, one of the gurus at NRAO:
-------------------------------------------------------------------------------
The structure of the phases will depend in detail on the model. To check whether you're getting something reasonable, try the following:
* If you restrict the uv-range in UVPLT to match the original (no model) recommendations, you should still get flat (zero) phases. Do you?
* You can check the phases predicted by the model by using UVSUB with opcode 'MODL' (this may require splitting off 3C xxx to make single source files). Try that and see whether the predicted phases match what you observe.
If the post-calibration phases do NOT match the model, there is indeed something wrong, and we'll work with you to track it down. I hope and expect this is not the case though... :}
-------------------------------------------------------------------------------
So, you basically first split off the calibrators into single-source files (SPLIT), then you substitute in each file the observed phases with those derived from the model images provided by AIPS (UVSUB with optype = 'modl'). When you then plot the phases of these files, they should look like the UVPLT of your calibrated data, only without the noise. At U-band (2cm) the model phases of 0137+331 and 0542+498 are so close to zero that the difference between the model and a point source is lost in the noise. But for 1331+305, the effect was quite spectacular. The model predicts phases up to |\phi| = 60\degr! That doesn't look like one should "force" the phases down to around zero by assuming a point source.
The catch is that I don't know whether the calibration with a model affects the polarization calibration. PCAL should be ok, since we are using the phase calibrator, which is assumed to be a point source. But RLDIF requires the polarization calibrators 0521+166 and 1331+305, for which the phases (RCP and LCP) will be substantially different from zero.
Thursday, March 29, 2007
Smoothing in BPASS
What is up with all those warnings about smoothing in BPASS? It seems so fricking absurd that you can only smooth your bandpasses if you smooth your data.
(One of my data sets had very high spectral resolution, so each channel was quite narrow and typical integration times on bandpass calibrators led to very noisy bandpasses. Of course, I didn't want to add all this noise to my data, but I did want to calibrate the bandpasses. A good solution to this would be to simply smooth the bandpasses, so that you'd get the general shape of the bandpass without all the nasty channel-to-channel variations.) But is smoothing ok? I don't want to sacrifice the resolution of my data!
This advice in from Katie Devine at the VLA in NM:
Hey, I learned something today that you'll find interesting: the smooth warnings for smoothing the data itself can be ignored if you have an antenna with intrinsically nice bandpasses. The reasoning is such-- if you have a really crazy bandpass (like on the GMRT for example) that has large dips, and you smooth those out with out smoothing the data, then a line that may have fallen in a dip in the bandpass will not be calibrated properly, because that dip will have been smoothed out (make sense? I can clarify if you want).
On the other hand, the bandpass for the VLA just has a little bit of noise to it, but no other intrinsic shapes. So smoothing the bandpass without smoothing the data is ok in this case. Basically, your gut instinct was correct.
You'll want that bpass correction if you're citing line strengths, especially if you have more than one line in each band...
Subscribe to:
Posts (Atom)
If you have a question about AIPS that CANNOT be answered by reading the