[daip] UVCON/FITS problem
Eric Greisen
egreisen at nrao.edu
Fri Oct 5 16:18:09 EDT 2007
Vincent Fish writes:
> I'm having some problems running UVCON, and I was hoping you might have
> some insights as to what's going wrong.
>
> The intent is to simulate a UV data set from an array, given an input
> image and a few observational parameters. The task works, in that when
> I use an image of Cyg A from Rick Perley and put in, for instance, the
> VLBA (using AIPSTARS:VLBA_UVCON), I can generate a usable, realistic UV
> data set.
>
> Now, what I'm trying to do is to use an image I've generated from
> (somebody else's) theoretical modelling of a source. Their data were
> in the form x,y,intensity; I've created a FITS file from this. When I
> read it into AIPS, I can display the image properly on the TV (TVALL).
> But when I try to use it as the input image in UVCON, it dies with the
> following error messages:
>
> poisso> UVCON1: TABINI: REQUESTED CC FILE 0 DOES NOT EXIST
> poisso> UVCON1: PROBLEM WITH CC FILE(ERROR= 2), WILL USE THE IMAGE
> poisso> UVCON1: *** WARNING: OVERRIDING SPECIFIED COMPUTATION METHOD
> poisso> UVCON1: IMAGE SIZES => DFT IS MUCH BETTER THAN GRIDDED
> CONTINUING ANYWAY
> poisso> UVCON1: SETGDS: imaging done with one OFFSET tangent plane
> poisso> UVCON1: CHNDAT:COULD NOT FIND FREQ AXIS
> poisso> UVCON1: FRQTAB: CHNDAT ERROR 1
All of the messages up to the FREQ axis message make sense if you
told it to use a CC model w/o providing one. You can tell it to use
a CMODEL = 'IMAG' model and those messages will go away. I do not
understand the next one which - as you have deduced is critical -
since the UV header gets created early on, squirreled away and then
sent in to the place where this test is done and the UV data header
must have 5 axes including FREQ. At this stage, it appears that it is
not testing your model header but the self-created UV header and it is
failing. If you send me your model and your full set of adverbs, I
suppose I could try it in the debugger (Leonia does not seem to be
around just now).
Eric Greisen
>
> Ah ha, you say. I've generated a 2-dimensional image that has no FREQ
> axis, so of course UVCON (or CHNDAT, or AXEFND, or H2CHR) chokes. But
> it's more fundamental than that. When I generate a 3-axis FITS file in
> IDL (RA--SIN,DEC--SIN,FREQ) or a 4-axis one (...,STOKES), UVCON dies
> with the same error message. This happens no matter whether I set the
> FITS header keywords before FITLD'ing the images or afterward (using
> PUTHEAD).
>
> Next, I tried FITLD'ing the 2-axis FITS file generated by IDL and using
> the AIPS verb AXDEFINE to create a FREQ axis (explicitly called such).
> It's clearly there. Here's the IMHEADER output (most of the keywords
> are stolen from the Cyg A file):
> AIPS 1: Image=Undefine (MA) Filename=05_200_2AXES.MODEL . 1
> AIPS 1: Telescope=MWA-LFD Receiver=MWA-LFD
> AIPS 1: Observer= User #= 101
> AIPS 1: Observ. date=31-MAY-2002 Map date=05-OCT-2007
> AIPS 1: Minimum= 0.00000000E+00 Maximum= 7.19929289E-04 JY/BEAM
> AIPS 1: ----------------------------------------------------------------
> AIPS 1: Type Pixels Coord value at Pixel Coord incr Rotat
> AIPS 1: RA---SIN 128 10 26 36.288 1024.00 -0.020000 0.00
> AIPS 1: DEC--SIN 128 26 00 00.000 1025.00 0.020000 0.00
> AIPS 1: FREQ 1 2.0000000E+11 1.00 1.0000000E+00 0.00
> AIPS 1: STOKES 1 1.0000000E+00 1.00 1.0000000E+00 0.00
> AIPS 1: ----------------------------------------------------------------
> AIPS 1: Coordinate equinox 2000.00
> AIPS 1: Map type=NORMAL Number of iterations= 3000
> AIPS 1: Conv size= 0.24522 X 0.23557 Position angle= 2.97
> AIPS 1: Observed RA 10 26 36.300 DEC 26 00 00.00
> AIPS 1: Maximum version number of extension files of type HI is 1
> AIPS 1: Keyword = 'WTNOISE ' value = 1.000007E+00
> AIPS 1: Keyword = 'CCFLUX ' value = 6.537626E+03
> AIPS 1: Keyword = 'CCTOTAL ' value = 6.537626E+03
> AIPS 1: Keyword = 'PARANGLE' value = -4.800507E+01
> AIPS 1: Keyword = 'ZENANGLE' value = 9.515759E+01
>
> But yet again, UVCON/CHNDAT can't find the frequency axis. For the
> record, I'm not sure why it even needs to. One specifies an observing
> frequency via APARM(1), a bandwidth via APARM(8) (problem arises for
> positive, negative, and zero values of APARM(8)), and a number of
> frequency channels via APARM(9).
>
> What am I doing wrong? Inputs are below. (Yes, I know the VLBA doesn't
> work at 200 GHz; etc.)
>
> Thanks for your help.
> Vincent
>
> AIPS 1: UVCON Generates UV data for a given array and model
> AIPS 1: Adverbs Values Comments
> AIPS 1: ----------------------------------------------------------------
> AIPS 1: INFILE 'AIPSTARS:VLBA_UVCON' Antenna location file name
> AIPS 1: IN2FILE *all ' ' TSYS and Efficiency
> AIPS 1: BLANK=>INFILE's data are used
> AIPS 1: for TSYS and Efficiency
> AIPS 1: OUTFILE *all ' ' Antenna location file name
> AIPS 1: Positions are in equatorial
> AIPS 1: coordinate system.
> AIPS 1: IN2NAME '05_200_2AXES' Model image name (name)
> AIPS 1: The both dimensions of the
> AIPS 1: model must be 2**N pixels
> AIPS 1: IN2CLASS 'MODEL' Model image name (class)
> AIPS 1: IN2SEQ 1 Model image name (seq. #)
> AIPS 1: IN2DISK 1 Model image disk unit #
> AIPS 1: INVERS 0 CC file version #.
> AIPS 1: OUTNAME ' ' Output UV file name (name)
> AIPS 1: OUTCLASS ' ' Output UV file name (class)
> AIPS 1: OUTSEQ 0 Output UV file name (seq. #)
> AIPS 1: OUTDISK 0 Output UV file disk unit #.
> AIPS 1: relatively of the initial
> AIPS 1: declinat., given at APARM(3)
> AIPS 1: per field (asec)
> AIPS 1: APARM 200 0 Control information:
> AIPS 1: -29 -12 1: Frequency of chan. 1, GHz
> AIPS 1: 12 0 2: Wavelength of ch 1, cm
> AIPS 1: 60 -8 IF both .LE. 0 then
> AIPS 1: 0 0 wavelength = 0.1 cm
> AIPS 1: 3: Source declination, deg
> AIPS 1: 4: Min hour angle, hours
> AIPS 1: 5: Max hour angle, hours
> AIPS 1: The hour angles are for
> AIPS 1: the given array center
> AIPS 1: 6: Min antenna elevation, deg
> AIPS 1: 7: Integration time, sec
> AIPS 1: 0 => 1.D6
> AIPS 1: to simmulate snapshot
> AIPS 1: 8: Bandwidth(increment) of
> AIPS 1: the freq. channel, MHz
> AIPS 1: >=0 => it is increment to
> AIPS 1: simulate multi chann. data
> AIPS 1: =0 => Bandwidth = 1MHz
> AIPS 1: <0 => one channel data to
> AIPS 1: simulate multi frequency
> AIPS 1: UV coverage
> AIPS 1: 9: Number of freq. channels
> AIPS 1: 0 => 1
> AIPS 1: 10: Max blockage allowed
> AIPS 1: 0 => 1 no blockage
> AIPS 1: BPARM *all 0 Control information:
> AIPS 1: 1: Multiplier of the calcul.
> AIPS 1: noise. 0 => 1
> AIPS 1: -1 => 0 (no noise)
> AIPS 1: 2: Atmosphere noise at zenith
> AIPS 1: in degrees.
> AIPS 1: 3: RMS of pointing error,
> AIPS 1: random among all antennas
> AIPS 1: but constant in time,
> AIPS 1: in arcsec
> AIPS 1: If (BPARM(3).LT.0) then
> AIPS 1:
> AIPS 1: the phase and amplitude
> AIPS 1: errors of each antenna
> AIPS 1: are simulated instead
> AIPS 1: of pointing error.
> AIPS 1: ABS(BPARM(3)) is half
> AIPS 1: range of homogeneously
> AIPS 1: distributed phase, rad
> AIPS 1:
> AIPS 1: BPARM(4) is half
> AIPS 1: range of homogeneously
> AIPS 1: distributed natural LOG
> AIPS 1: of factor to amplitude
> AIPS 1:
> AIPS 1: BPARM(5):
> AIPS 1: 0 => only one (first)
> AIPS 1: clean component is
> AIPS 1: affected by the
> AIPS 1: phase/amp noise
> AIPS 1: 1 => all clean
> AIPS 1: components are affected
> AIPS 1:
> AIPS 1: 4: global pointing error,
> AIPS 1: constant in time for all
> AIPS 1: antennas,
> AIPS 1: in arcsec
> AIPS 1: 5: RMS of pointing error,
> AIPS 1: random among all antennas
> AIPS 1: and in time,
> AIPS 1: in arcsec
> AIPS 1: 6: Type of the primary beam
> AIPS 1: 1 => circular dish with
> AIPS 1: the flat illumination
> AIPS 1: 2 => illumination is 10dB
> AIPS 1: down at the dish edge
> AIPS 1: 3 => illumination is 15dB
> AIPS 1: down at the dish edge
> AIPS 1: The dish diameter is given
> AIPS 1: at the INFILE (antenna 1)
> AIPS 1: 4 => Gaussian beam with
> AIPS 1: given BMAJ, BMIN, BPA
> AIPS 1: in degrees
> AIPS 1: 5 => Gaussian beam with
> AIPS 1: variable BMAJ, BPA
> AIPS 1: depending on the time
> AIPS 1: .GT.0 =>Multiply the model
> AIPS 1: by the primary beam.
> AIPS 1: 0 => Not multiply the
> AIPS 1: model by the primary beam.
> AIPS 1: 7: Time tolerance, in minutes
> AIPS 1: 0 => 1
> AIPS 1: If the difference of the
> AIPS 1: current and previous time
> AIPS 1: is < the time tolerance,
> AIPS 1: the pointing error or the
> AIPS 1: phase of the antenna or
> AIPS 1: primary beam parameters
> AIPS 1: are not changed
> AIPS 1: 8: Shift the UV data by
> AIPS 1: RASHIFT, DECSHIFT?
> AIPS 1: 0 => yes shift
> AIPS 1: 1 => no shift
> AIPS 1: 9: Range of the primary beam
> AIPS 1: 0 => 2.5
> AIPS 1: 10: If OUTFILE.NE.BLANK then
> AIPS 1: 0 => calculate OUTFILE and
> AIPS 1: exit
> AIPS 1: 1 => calculate OUTFILE and
> AIPS 1: carry out the rest of
> AIPS 1: job
> AIPS 1: BMAJ 0 FWHM major axis of the
> AIPS 1: Gaussian primary beam, degree
> AIPS 1: See help for the variable
> AIPS 1: primary beam (BPARM(6)=5)
> AIPS 1: BMIN 0 FWHM minor axis of the
> AIPS 1: gaussian primary beam, degree
> AIPS 1: BPA 0 Position angle of Gaussian
> AIPS 1: primary beam, degree
> AIPS 1: DO3DIMAG -1 1 => use W term calculating
> AIPS 1: visibilities (only if
> AIPS 1: CMETHOD='DFT', and
> AIPS 1: CMODEL ='COMP')
> AIPS 1: 0 => no W term calculating
> AIPS 1: visibilities
More information about the Daip
mailing list