As of Fri Oct 19 12:17:44 2018

COMAP: Procedure to CLEAN and Self-Calibrate UVDATA


INNAME                             Input UV data (name)
INCLASS                            Input UV data (class)
INSEQ                              Input UV data (seq. #)
INDISK                             Input UV data disk drive #
IN2NAME                            Model for First Self-Cal
IN2CLASS                           Model image (class)
IN2SEQ                             Model image (seq #)
IN2DISK                            Model image disk drive #
OUTDISK                            Output image disk drive #
CELLSIZE         0.0     9999.9    (X,Y) size of grid in asec
IMSIZE           0.0     4096.0    Image size
NITER                              Number of clean iterations
REFANT                             Reference ant. for Self-cal.
SOLINT                             Min. Solution interval (min)
NMAPS                              Number of Map-Selfcal loops
DOPOL                              >0: Produce Polarization maps
DOCENTER                           >0: Move Source to map center
DOALL                              >0: Do amplitude self-cal
CUTOFF                             Clip Flux Limit. <0: No Clip
UVTAPER                            Taper to apply to UV data
UVRANGE                            Range of UV data to include
UVWTFN                             Weighting function for data
GAIN               0.0         2.0 Clean Gain
NBOXES             0.0        50.0 Number of boxes for CLEAN
CLBOX              0.0      4096.0 Four coordinates for each box
BMAJ     *                         FWHM(asec) maj. axis beam
BMIN     *                         FWHM(asec) min. axis beam
BPA      *      -360.0       360.0 CLEAN beam position angle
BADDISK         -1.0     1000.0    Disks to avoid for scratch.


Type:  Procedure
 Use:  COMAP is a procedure which maps and self-calibrates
       UVDATA.  Makes many simplifing assumptions, sets input
       values for MX and CALIB.  Uses UVSUB to divide by the
       source model before running PHASE to coherence time
       average the input UV data.


  INNAME.....Input UV data file (name).       Standard defaults.
  INCLASS....Input UV data file (class).      Standard defaults.
  INSEQ......Input UV data file (seq. #).     0 => highest.
  INDISK.....Input UV data file disk drive #. 0 => any.
  IN2NAME....Input Model Image for first self-calibration.
             if INNAME='POINT', a point source model is used.
  IN2CLASS...Model Image class
  IN2SEQ.....Model Image sequence number.
  IN2DISK....Model Image disk number
  OUTDISK....The disk drive # of output images.  0 => highest
             with space (note: map and Beam go on same disk.
  CELLSIZE...(X,Y) pixel separation in asec.
  IMSIZE.....(X,Y) The minimum desired size of the fields.
  NITER......CLEAN iteration limit. 0 => 1000
  REFANT.....Reference Antenna for Self-Calibration
  SOLINT.....Minimum Self-Cal solution interval (min.).
             0 => 1 minute
  NMAPS......Number of Map and Self-Cal Iterations
  DOPOL......Produce Polarization images.
  DOCENTER...Shift source peak to center of the map.
  DOALL......Attempt Amplitude Self-calibration.
  CUTOFF.....Maximum Flux allowed in UV-Data.  UV-data are
             allowed with brightness between 0 and CUTOFF.
             <0: No Clipping.  =0: Zero spacing flux + Noise
  UVTAPER....(U,V) gaussian taper (kilolambda) at 30 percent level
  UVRANGE....(Minimum,Maximum) baseline (kilolambda) in map.
  UVWTFN.....Weighting function of (u-v) place.
             blank=>Uniform; 'NA'=>Natural
  GAIN.......The CLEAN loop gain.  0 => 0.10.
  NBOXES.....Number (<=50) of rectangular search boxes
             To search on the first field; the clean window
             in all other fields is given by  FLDSIZE.
             0 => use FLDSIZE to determine windows on field 1.
  CLBOX......A 4x50 array with the BLC and TRC of each box.
             0 => use window specified in FLDSIZE.
  BMAJ.......The FWHM (asec) major axis of the restoring beam.
             If 0; value obtained from fitting to the beam.
             If <0; output will contain the residual image.
  BMIN.......The FWHM (asec) minor axis of the restoring beam.
  BPA........The position angle in the unrotated image of BMAJ.
  BADDISK....This array contains the numbers of disks on which
             it is desired that scratch files not be located.
             BADDISK has no effect on input and output maps.


COMAP:      Procedure to Map and Self-Calibrate UV data
DOCUMENTOR: G. Langston and W. Cotton (NRAO)
            Also see MX and CALIB documentation.
            and COMB.


     COMAP combines the functions of Mapping, CLEANing and
self-calibrating a UV data set.  The UV data must be in AIPS
single source format.

    The COMAP procedure first divides the UV data by the best
source model to create "pseudo-point source" UV-data.  The AIPS
capability unique to COMAP is the use PHASE task which reads in
the divided UV data and creates an intermediate "Coherence-Time-
Averaged" UV data set.  The data in this file are smoothed over
the coherence time of each baseline.  In this way, short
baselines which often have higher signal-to-noise ratio due to
more signal as well as lower baseline length dependent effects
are more significant in the Self-calibration solution.

    After the data are smoothed, the data are Self-calibated and
the Self-calibaration is applied to the original (unsmoothed)
UV data.  The Self-Calibrated UV data are then deconvolved and
the cycle repeated.

                   ADVERBS PECULIAR TO COMAP

     Most of the adverbs used by COMAP are duplicates of those
used by MX and CALIB.  However if the Adverbs are not specified,
(such as REFANT or CELLS) value is estimated from the UV data.

     Input UV-data used by COMAP to creat the self-calibrated
UV-data and output Image. COMAP is able to pick a REFerance
ANTenna, based on the signal to noise ratio of visablities.

     First model image used to self-calibrate the input UV-data.
The clean components for the image specified by IN2NAME,
IN2CLASS, IN2SEQ and IN2DISK are selected in the same manner
as subsequent self-calibrations.  If no image NAME is specified,
the first self-calibration follows the first deconvolution.
If IN2NAME is "POINT" then a point source model used used as the
starting point of self-calibraton.

     The minimum desired size for all of the fields.  The limits
are 32x32 to 4096x4096 and must be a power of 2.  The adverb
FLDSIZE define the region over which clean components are
searched for.

     For the first field, up to 50 CLEAN windows can be
specified via CLBOX as an alternate to FLDSIZE.  This allows
more flexibility than a single window centered on the phase
center.  If NBOXES is greater than 0 then the contents of CLBOX
is used to specify the input window.  Since these values are in
pixels care should be taken that they are determined from an
image made with the same cellsize and shift.
     NOTE: the values containes in CLBOX are not used to
determine the size of the image for field 1.  IMSIZE and/or
FLDSIZE must be used for this.  In the case that CLBOX and
NBOXES are used, this is the only use made of FLDSIZE for field
1.  Its use for higher numbered fields is unaffected.  If CLBOX
is 0's then the value of FLDSIZE (or its default) is used for
     NBOXES and CLBOX specify the size and location of the
rectangular boxes comprising the "CLEAN Window" area A.  You
make the best use of prior knowledge about the source to help
COMAP do its job when you restrict A as closely as possible to
the true boundaries of its emission. Recall that CLEAN attempts
to optimize F(n) as a representation of the sky, with zeroes
everywhere else.  The more information you provide about where
the "real" emission is, the more likely COMAP is to converge on
a unique, credible solution.

     When using COMAP to make several small maps over a large field
of view, use Natural weighting rather than Uniform weighting in
order to obtain the signal to noise and resolution which are
comparable to that obtained from one large map over the field of
view. For mapsizes of 256 or less, the loss of signal to noise
using Uniform weighting can be a factor of two or three.

     If OUTDISK = 0, the map and beam will be put on the same
disk.  The Q and U maps may be put on another disk.  It is best
to specify OUTDISK.

     For effective CLEANing of the maps, the number of pixels
per beam should be such that the pixel value immediately north
or east of the beam center is no less than about 50 percent of the
peak. However, if tapering is used, the outlying (u,v) points
may not have any significant weight in the map.
     Strong aliased sources should be CLEANed in separate fields
unless they are close to the object of interest.
     COMAP will make maps which have a power of two pixels on a
side; between 32 and 4096 on the X-axis and between 32 and 4096
on the Y-axis.  FLDSIZE defines the region to be searched for
CLEAN components.
     If for some reason it is desirable to map a region much
larger than the region being CLEANed, IMSIZE can specify the
minimum size of a map.  Components will be CLEANed from the
region specified by FLDSIZE but the output image size will be as
specified by IMSIZE.  Values in IMSIZE must be powers of 2.

     If you do not expect your source to show significant
circular polarization, as is normally the case with galactic
and extragalactic continuum sources, making a V map can be a
useful diagnostic for calibration problems, correlator offsets,
etc.  The V map should be a pure noise map close to the
theoretical sensitivity if your data base is well calibrated
and edited.

     The default uniform weighting option gives higher
resolution than natural weighting.  However, uniform weighting
gives a lower signal to noise ratio.  Natural weighting is
therefore preferable for detection experiments.
With uniform weighting the dirty beam size decreases slightly
with larger maps, other parameters remaining unchanged.

     The depth to which CLEAN carries out its deconvolution is
approximately measured by the product NITER*GAIN.  The first
CC extension file version corresponds to the first output
frequency channel.
     The value of NITER and the execution time needed
to reach a given CLEANing depth are minimized by setting GAIN =
1.0, but setting GAIN > 0.5 is recommended only when removing
the sidelobes of a single bright unresolved component from
surrounding fainter structure.

     The default values of 0 for these parameters invoke an
algorithm whereby the central portion of the dirty beam B is
fitted with an elliptical Gaussian function whose parameters are
then used to specify the Clean Beam H.  The algorithm can be
"fooled" by positive or negative sidelobes near the main
lobe of B, and has been known to prescribe unsatisfactory forms
for H, particularly for snapshot maps.

     Mapit can be used in BATCH mode.  Below is a sample AIPS
session which sets up the BATCH queue and runs COMAP:

 ... if mapit inputs look ok
 ... create a list of batch commands
 ... now start the batch queue

                      Hidden ADVERBS

      There are several hidden ADVERBS to COMAP which are not
usually changed by the user.  In an attempt to keep the inputs
to COMAP simple, these ADVERBS are not in the ADVERB list above.
However, there are situations where these other ADVERBS might
prove useful to the advanced user.   These adverbs are listed

DOINT:    SCALAR, default FALSE (-1)
     DO INTeractive COMAP; if true, before deconvolution the
user is requested to input clean boxes, marking the locations
were the image will be searched for clean components.  This
option allows spurious source components to be rejected allowing
MX to clean only in user specified regions.  This interactive
mode also alls the user to stop cleaning when satisfied with the
output.  This mode was inspired by the Jodrel Bank OLAF package

DOZAP:    SCALAR, default TRUE (1)
     If DOZAP is true, then all intermediate images and self-
calibration solutions are deleted.  This step is necessary to
conserve disk space.  If the user wishes to study COMAP progress
setting DOZAP false leaves all intermediate images and UV-data
around for examination.

RMSFACT:  SCALAR, default 1.2
     RMS FACtor is the rms noise factor used to stop bad
self-calibrations and deconvolutions within COMAP.  COMAP
continues the cycle of self-calibration and deconvolution
until the requested number of iterations are performed OR
the noise in the image produce by the current deconvolution is
worse than the previous deconvolution.   Setting RMSFAC
greater than one allows the solution to get worse, in the
hope that as the COMAP parameters change, the solution will
improve.  Note that the default allows the solution to get
20  percent worse before quiting.  Experience has found that the
transition from PHASE to AMP+PHASE self-calibration
causes significant changes to the RMS noise level IN EITHER
DIRECTION.  Few errors have snuck past the 20  percent level and
failed to improve.  If AMP solution is bad it is usually
EXTREAMLY bad and is detected.

CLIMIT:   SCALAR, default 0.04 Jy
     Clip LIMIT: the lowest source flux level which may be
cliped in the CLIP step.  Because of the current performace
of the VLA receivers, clipping at a lower level is certainly
cliping the NOISE due to the receiver temperature.  The
function of CLIP is to remove NON-GAUSIAN distributed noise
(ie interference).

CCFACT:   SCALAR, default 2
     CC FACTor is used to determine the number of clean
components to use in the next self-calibration cycle.  It is
assumed that all negative CCs are spurious.  If negative CCs
are spurious, then probably a few faint positive CCs are
sperious as well.  CCFACT is used to determine the positive
clean component limit.  All CCs brighter than CCFACT times the
absolute value of the brightness of the first negative CC are
used in the Self-cal model.

SOLFAC:  SCALAR ARRAY, defaults 1,  3/4, 2/3, 1/2, 1/3,
                               1/4, 1/5, 1/6, 1/8, 1/10
     SOLution interval FACtors, is an array of values used
to scale the input self-calibration solution interval.
Self-calibration is most successful if the time over which
the data is averaged is just longer than the time over
which ionosphere (atmosphere) changes the phase of the data.
However the longer the averaging time, the better the model
is.  The SOLFAC array is used to scale SOLINT from long
to short as a function of self-calibration loop.  The
first two self-cals use the input interval, the third self-
cal uses 2/3 SOLINT, the fourth, a half SOLINT, etc.

SCTYPE:     STRING*2, default '  ' (least square fit)
     Self-Calibration solution TYPE, SCTYPE, is the type of
fit used by CALIB to calculate the antenna gains which minimize
the difference between UV-data and the source model.  The least
linear fit (SCTYPE='L1' is less sensitive to a few spurious
data than the least squares (SCTYPE='') fit.  The least squares
fit is better if the data has errors which are gaussian
distributed.  The least linear fit is better if interference is

NOLDUV:     SCALAR, default 2
     NOLDUV is number of self-calibration cycles using the
original UV-data as input.  After NOLDUV self-cals, the
previous self-cal output is used as input to the next self-cal
executions.  By default, the first two self-cals use the
original UV-data.

                      Notest on Snapshot imaging

     A number of interesting observations have been made
concerning imaging VLA snapshots during the testing of COMAP.
One is the use of the MX adverb FACTOR to determine when the
MX minor cycle clean should stop.  (Also See MX HELP)  In
brief, MX, performs a minor cycle clean by reading in a large
number of bright pixels of an image.  MX finds the maximum
pixel in the input image and subtracts the beam from the
entire image, until the brightest residual is fainter than
the Beam's brightest sidelobe.  In this way no BEAM features are
confused with real sources.

     The point to note is that the VLA BEAM model assumes
perfect phase calibration, however for snap-shots, the phase
calibration is usually rare and the BEAM side-lobes are
usually large.  This causes sidelobe in the image to be
brighter than expected and can be turned into sperious features
due to self-calibration.  Fortunately there is an easy way
to avoid this problem, using the MX adverb FACTOR.  By setting
FACTOR to -1, the minor cycle clean stops after each clean
component is found, but is very slow.  The MX default is FACTOR
equals 0.  For FACTOR = 1, the Minor Cycle does not stop until
all NITERations are complete.

A reasonable value for FACTOR for snapshots is between -0.4 and
-0.1, so that cleaning stops before side-lobes are confused with

See HELP COMAP_DO and HELP COMAP_MX for further COMAP inputs.