[ Basic Info | References | User Guide ]

Basic Information on uvcal


Task: uvcal
Purpose: Apply special processing options to uv-data.
Categories: uv analysis

UVCAL is a MIRIAD task which applies special processing options
to uv-data. By default, UVCAL applies the calibration files 
before it processes the uv-data into the output file.

Key: vis
The names of the input uv data sets. Multiple names can be given,
separated by commas. At least one name must be given.

Key: select
The normal uv selection commands. See the Users manual for details.
The default is to process all data. Note that window selection can
not be used with options=hanning, passband, contsub, or wide.

Key: radec
Source right ascension and declination. These can be given in
hh:mm:ss,dd:mm:ss format, or as decimal hours and decimal
degrees. Setting RA and DEC will change the phase center to
the RA and DEC specified. The default leaves the data unchanged.

Key: options
This gives extra processing options. Several options can be given,
each separated by commas. They may be abbreivated to the minimum
needed to avoid ambiguity. Possible options are:
'nocal'      Do not apply the gains file.
'nopass'     Do not apply bandpass corrections.
'nopol'      Do not apply the polarization leakage file.
'nowide'     Do not copy across wide-band channels.
'nochannel'  Do not copy across spectral channels.
'unflagged'  Copy only those records where there are some
             unflagged visibilites.
'contsub'    Continuum subtraction using a linear fit to the real
             and imaginary parts of each spectral window.
             Exclude badchan and endchan specified below.
'conjugate'  Conjugate visibilities
'fxcal'      Calibrate FX correlator data
             by dividing the cross correlations by the 
             geometric mean of the autocorrelations.
'hanning'    Hanning smooth each spectral window.

'linecal'    Remove phase slope across passband due to line length
             changes. This option also corrects the phase difference
             between the wide channels due to line length.
             options=linecal,avechan remakes the wideband from
             the channel data after the phase slope is removed. 
'atmcal'     ATM phase monitoring testing - work in progress
             It will normalize on freqatm, and if missing will attempt
             to use lo1. Use puthd to fake a new freqatm variable.

'parang'     Multiply LR by expi(2*chi) and RL by expi(-2*chi)
             where chi is the parallactic angle for Alt-Az antennas.
'passband'   Fit passband to lsb and correct the uv-data.
             Can be used when the source is strong. e.g. for
             planets. The passband is estimated from the lsb,
             smoothed within each spectral window, and applied
             to the lsb data. The complex conjugate is applied
             to the usb. This corrects for time variable IF
             passband errors. The uv-data must contain the same
             spectral windows in both sidebands of LO1.
'noisecal'   Copy conjugate of LSB spectral windows into USB.
         Can be used to copy the CARMA noise data into the USB.
         MFCAL can then fit the IF bandpass in both sidebands.
         This option assumes that the data contain the same 
         spectral windows in LSB and USB.
 'uvrotate'  Rotate uv-coordinates from current to standard epoch.
             The standard epoch, ra, dec can be changed using PUTHD.
 'avechan'   Average unflagged spectral channels. Make new wideband.
            excluding badchan and endchan specified below.
 'slope'    Fit phase slope for each spectral window. Make new wideband.
            wideband phase is slope in units radians/GHz, and can
            be used to fit delay and antenna positions using BEE. 
            Exclude badchan and endchan.
 'holo'      
Enable other MIRIAD tasks to analyze interferometer pointing data.
replace [u,v] with [dazim,delev] for ant1 in each baseline.
if(dazim(ant1).eq.0  .and.  delev(ant1).eq.0) then replace [u,v] with
[dazim,delev] for ant2 in each baseline. If both ant1 and ant2 have
zero dazim and delev, then [u,v] = [0,0]. [u,v] are in arcsec units. 
The data can be calibrated using select=uvnrange. N.B. not "uvrange".
e.g. uvcal options=holo vis='infile' out='outfile'
   selfcal vis='outfile' 'select=uvnrange(0,1)' options=amp flux=1
Beam profiles can then be analysed using standard programs. e.g. UVPLT.
Holography images of antenna surfaces are made using INVERT and IMHOL.
INVERT makes a real image; options=imaginary makes the imaginary part.
IMHOL combines the real and imaginary images into amplitude and phase
images of the antenna surface.
Related tasks: VARMAP makes images versus [dazim,delev] directly.

NOTE:   Options=nochannel cannot be used with "hanning", "passband"
and "contsub". For these 3 options no processing is performed on the
wideband data.

Key: badchan
Number of ranges of bad channels followed by list of up to 20 pairs
of numbers giving range of channels to exclude in options=contsub
and options=avechan. E.g. badchan=3,1,3,15,16,256,257 
excludes the 3 ranges   (1,3) (15,16) and (265,257)

Key: endchan
Number of channels to drop from the ends of each spectral
window in options=passband, contsub, and avechan. Default=0.

Key: nave
Number of channels to average in options=passband. Default=1.

Key: sigma
Rms noise level used to flag data in options=contsub and options=slope.
All channels are flagged if the rms noise level per channel is greater
than sigma after subtracting the continuum or slope.
Endchan and badchan are not included in computing the rms.
The default is sigma=0. which does not flag     any data.

Key: scale
Multiply the data and wideband by cmplx(scale1,scale2). Two values.
Default is no scaling. If only one value is given it is assumed to
be a real value. E.g. scale=0,1 multiplies the data by sqrt(-1)

Key: offset
Subtract a complex valued offset from the uv-data. Two values give
itude (after calibrations have been applied by uvcal) 
phase in degrees.  Default is none.

Key: model
Subtract a source model from the uv-data. Three values
give the flux density [Jy], scale [nanosecs], and index 
for a radial power law:
        model = flux * (uvdist/scale) ** index
The model visibility is subtracted from the uv-data.
Default model=0,0,0. i.e. no model is subtracted.

Key: onsource
Set the uvvariable "on" for each baseline to be true if both ant1
and ant2 have abs(dazim).LE.delta_az .AND. abs(delev).LE.delta_el, 
else set the uvvariable "on" = false.  The data are unchanged.
Antenna delta_az and delta_el used to designate an "on source"
observation. Two values given as dd:mm:ss, or as decimal degrees.
This enables other MIRIAD tasks to analyze interferometer pointing
or single dish data by designating "on" and "off" observations 
which can be selected using select=on.
Default onsource=0,0. Use non-zero values to process onsource.
options=holo (see above) allows us to use the standard grid-invert
imaging on planets, after we have done the calibration using 
delta_az and delta_el to select the "on source" calibrations.

Key: polcal
Subtract the source polarization from the uv-data. Two values
give polarized intensity, Ip, and position angle, 2*psi, in degrees.
Use same amplitude units as uv-data after calibration is applied..
subtract Ip*expi(2*psi)*expi(-2*chi) from LR, and
subtract Ip*expi(-2*psi)*expi(2*chi) from RL
where chi is the parallactic angle for Alt-Az antennas.
Default=0,0. is no correction.
After subtracting the source polarization, the averaged uv-data
gives the instrumental polarization for each baseline.

Key: polcode
Change the polarization code. Default polcode=0 makes no change.
Polarizations are  YX,XY,YY,XX,LR,RL,LL,RR,-,I,Q,U,V
Polarization codes -8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4
E.g.  polcode=4  changes YX,XY,YY,XX to LR,RL,LL,RR
E.g.  polcode=-4 changes LR,RL,LL,RR to YX,XY,YY,XX

Key: parot
Rotate uv coordinates and hence image by parot degrees.
Rotation is +ve to the E from N. i.e. increasing PA.
Default is no rotation.

Key: seeing
Amplitude correction for atmospheric coherence, specified by
the FWHM in arcsec, of a circular Gaussian fit to
the phase calibrator(s) for the target source.
Use uvfit object=gauss fix=xyc....
Default seeing=0. is no amplitude correction.

Key: fringe
Correct data for wrong on-line fringe rate calculation.
Apply a phase correction
7.29115e-5 * preamble(1) * freq * cos(obsdec) * fringe * (time - time0)
for each frequency channel,
where fringe is the fraction of the fringe rate calculated
from the values of preamble(1), freq, obsdec, and time in the uvdata.
time0 is an optional start time, given in the standard Miriad form 
and is converted to Julian date. 
time0 should be the same for all the target sources and 
phase calibrators to avoid phase offsets between targets and calibrators.
default: fringe=0,09MAY21:13:27:04.0.  i.e. no fringe rate correction
09MAY21:13:27:04.0 is Julian Day 2454973.060463
e.g. 
i) if fringe rate has not been applied on line, then set fringe=1
ii) data at 3040 MHz that has been fringe rotated for 3140, fringe=-(3140-3040)/

Key: out
The name of the output uv data set. No default.

User Guide References to uvcal

[ Basic Info | References | User Guide ]

Generated by smamiriad@cfa.harvard.edu on 09 Jul 2012