3.2 Source Catalog

SCHED uses catalogs to get source information such as names, positions, and velocities. These catalogs are in keyin format. Entries are terminated with a “/”. Some or all of the catalog may be included in a section in the SCHED input file. SCHED will look there first for sources, and then go to the two possible external catalogs to find the rest (or all). Please note that the source catalog, along with the locations catalog of station positions, is updated approximately annually to new solutions. All source positions will change, although likely by small amounts. But relative astrometry projects should either use their own source positions to keep them constant over long times, or include processing steps that account for any changes in the catalog positions (the preferred solution).

As of Feb. 2015, there are two catalogs (plus older versions) included with the SCHED distribution. One contains a recent astrometry solution from the Goddard Space Flight Center geodetic group. There is a symbolic link to it called sources.gsfc in $SCHED/catalogs. Another contains a recent solution provided by Leonid Petrov. The symbolic link for the latest of these is sources.petrov. The Petrov file typically has very nearly all of the sources in the Goddard solution (they are working from the same data sets) plus a significant number more — over 9000 in total. There used to be a third file, sources.vlba that contained all sources in either of the above plus around 500 others from JVAS and an old VLA calibrator list. As the first two lists got large, the remaining extra sources in sources.vlba become more suspect as VLBI targets. Plus maintenance of the combined list was somewhat problematic, so it was dropped in Feb. 2015. If there is a sources.vlba in the distribution, it is either out of date, or is actually a symbolic link to one of the main files, put there to try not to break old schedules.

The sources in the catalogs from the geodetic solutions are based on the International Celestial Reference Frame version 2 or ICRF2 (these have CALCODE=V). These have positional errors of a fraction of a milliarcsecond to a few milliarcseconds. They include sources from the standard geodetic observing program, sources from the VLBA calibrator survey, the Australian LBA survey, and various sources specially observed in the RDV project on the VLBA.

Users of accurate positions from the geodetic catalog are encouraged to reference the original papers, some of which are listed in the catalog headers. Information about, and data from, the GSFC solutions can be found at http://vlbi.gsfc.nasa.gov/dataresults_main.htm. Catalogs and much other useful information provided by Leonid Petrov can be found at http://astrogeo.org/.

SCHED input parameters SRCFILE and SRCFILE2 are used to point to any desired external catalogs. A file name of up to 80 characters can be specified. The default source file is:


On unix systems with the environment variable SCHED properly defined, this is the standard catalog distributed with SCHED. Users of the html version of the manual can read the catalog by clicking GSFC or Petrov.

Source catalog information can be given in the main SCHED keyin file. If the keyword SRCCAT appears, all input after the next “/” is assumed to consist of source catalog entries until a line containing the keyword ENDCAT and a “/” is encountered (don’t combine this keyword with a catalog entry). Such “in-line” catalogs may appear anywhere in a SCHED keyin file, although it is probably best to put them near the beginning. SCHED will look in the external catalog for any source not found in an in-line catalog. If you wish to prevent SCHED from looking in external catalogs, specify SRCFILE=NONE (SRCFILE2 is not used at all unless specified.). It is common to provide in-line source catalog entries for unique program sources, but let SCHED find the fringe finders and other calibrators in the standard catalog. SCHED will look in the in-line catalog first, then in SRCFILE, then in SRCFILE2. It will use data from the first place a source is found.

The parameters for each entry in the source catalog are below. Lower case letters in this list are optional. All catalog parameters except EQUINOX and EPOCH revert to a default unless specified for a source. EQUINOX and EPOCH default to the last specified.

Note that the plotting capabilities of SCHED, specifically the RD plots, can be used to examine the distribution of sources in the catalog and to find catalog near any desired location. This is useful, for example, for finding phase calibrators near target sources.

SCHED will use whichever equinox coordinates are provided in the catalog and determine J2000, B1950, and DATE coordinates. All of these coordinates are listed in the .sum file. Various stations require different equinox coordinates. For example, the VLBA only understands J2000 coordinates while some other sites only understand B1950, although those may all be gone by now. Sched uses the SLALIB routines to make the coordinate conversions.

The J2000 and B1950 coordinates systems move relative to each other so an “observe” date is needed to make an accurate conversion. This is the date at which the coordinates of the reference sources were determined. For example, the VLA calibrator list coordinates were determined in 1979.9 originally, and all subsequent updates, based mainly on J2000 positions, were precessed to B1950 using an assumed “observe” date of 1979.9. A similar situation exists at MERLIN, where observing is done in B1950 coordinates. There, the calibrator positions (based again on positions provided originally in J2000) are precessed to B1950 using an “observe” date of 1950.0. Any B1950 source position determined on MERLIN should be precessed to other systems using that date. The user can specify the “observe” date using the main schedule parameter PRECDATE which defaults to 1979.9, the correct value for use with B1950 source positions determined on the VLA. For purposes of pointing antennas, the value of this date that is used is not very important. But if you are trying to phase reference, phase the VLA at a high frequency, or derive positions for correlation, this is an issue that should be dealt with properly.

The parameters of the source catalog are:

This is the name of the source. Up to 10 source names (increased from 5 on Sept. 2010) can be provided to handle aliases. The names can have up to 12 characters each, although, as noted below, some downstream software may not like that much. No software we know of requires less than 8 characters.

Be a bit careful about special characters included in source names. Some are benign. But, for example, the * and ? characters are used as wild cards in many programs so, if either ends up in a file name (as it will after SPLIT in AIPS), it could cause problems. The temptation to use * is high because of sources such as SgrA*. But the conservative approach is to avoid it. Also, do not include imbedded blanks in source names, as they cause problems at a variety of places in processing. Minus signs (dashes), plus signs, underscores, and parentheses should not cause problems and are the commonly desired characters.

Note that the number of characters you can have may be limited by downstream software. Snap files require source names of 10 characters or less, but are not normally written by SCHED any more. Finally, beware that the Mark III correlators and the geodetic software have 8 character source name limits.

Right ascension. Remember that the VLA and Westerbork need sub-arcsecond positions for phasing.
The coordinate error in the RA direction in but in milliarcseconds. Note that for RA, this is 15 times the coordinate value error. It is the coordinate error in the sense that, to get the position error in angle on the sky, one would need to divide by cos(Dec). Prior to Dec 2014, this manual description indicated that RAERR was the angle on the sky, but in practice for some time before that (forever?), it had been the coordinate error in milliarcseconds.
The angular error in the declination direction in milliarcseconds. For dec, this is the same as the coordinate error.
Epoch of the observations. Allowed values are B1950, J2000, and DATE.
Single-character calcode used for VLBA control files. It is used by postprocessing packages to identify types of sources, usually types of calibrators. The CALCODE is optional for SCHED. CALCODE=’G’ is reserved for pulsar observations. This calcode will trigger the use of the pulsar gate on the VLBA correlator. Think of G as “Gate”. CALCODE=’L’ triggers on-line/off-line channel differences to be used for pointing data analysis. CALCODE=’Z’ indicates that the source is a satellite and that ephemeris data are needed for correlation (SCHED will set this one for any satellite).
Velocity (km/s) of a spectral line source. Up to 8 (16?) values can be given, one per base band channel. The value for the first base band channel will be used for any other channels for which no velocity is specified. The reference frame (eg LSR, Heliocentric...) and definition (radio, optical or redshift) are given by VREF and VDEF.
The reference frame for velocity calculations. Supported options so far are ’L’ for LSR, ’H’ for heliocentric and ’G’ for geocentric. LSR is the default.
The definition of the velocity. Supported options are ’R’ for radio definition (V = c(ν0 - ν)∕ν0), ’O’ for optical definition ( V = c(ν0 - ν)∕ν = c(λ - λ0)∕λ0 = zc ), and ’Z’ if the redshift is given directly. The default is the radio definition.
For planetary motion, the offset from the RA and DEC are assumed to be zero at this time. The format is PMEPOCH = yyyy,mm,dd,hh:mm:ss; for example, PMEPOCH = 1995,12,7,15:0:0 for 15 hours UT on 1995 December 7. I appologize for any confusion caused by PMEPOCH not being the zero time for PMRA and PMDEC instead of DRA and DDEC. The SCHEDusage for planetary motion was established long before the proper motions were added to the program and I did not want to change the meaning of PMEPOCH. Also EPOCH is the standard throughout astronomy for the zero time of proper motion, which is why it’s meaning was changed from what is now EQUINOX.
Rate of change of RA in seconds of time per UT day. This acts in coordinates of date. Note that this is the rate of change of the coordinate value, not the angular rate which would be smaller by cos(dec). The zero offset time is given by PMEPOCH. Use DRA and DDEC for fast moving objects like planets. Use PMRA and PMDEC for slow moving objects like stars.
Rate of change of DEC in arcseconds per UT day. Note that the azimuth and elevation reported by SCHED will not take into account DRA and DDEC. This acts in coordinates of date. The zero offset time is given by PMEPOCH.
The proper motion in RA in milli arcseconds per year. This is an angular value and is 1/cos(dec) times the change in coordinate value. The zero offset time is given by EPOCH. When PMRA, PMDEC, or PARALLAX is used, the source coordinates are shifted from the time specified by EPOCH to a time that is the stop time of the first scan or, if it is given, PMEPOCH. PMRA, PMDEC are added to DRA and DDEC respectively so that small motions during the observation of fast moving stars can be accommodated. The proper motion and planetary motion parameters can both be used together, if one is mad enough to want to do so.
The proper motion in Dec in milli arcseconds per year. The zero offset time is given by EPOCH.
The zero point for proper motion expressed as a fractional year such as 1993.6.
The parallax in milli arcseconds. Synonym PARALAX for backward compatibility (yes, I misspelled it the first time so backwards compatibility is an issue).
Flux densities at up to 10 frequencies for the source. The arguments (up to 30) are in triplets with the first being the frequency in GHz, the second is the total flux density in some image, and the third is the unresolved (peak) flux density — what you are likely to see on long baselines. These values are typically taken as a byproduct of geodetic observations and should be taken as approximate. All compact source vary to some degree — most by a lot — so the source may well prove to have a different strength when used than is shown. But these flux densities should help distinguish the strong from the weak calibrators.
A text string that indicated where the flux density information came from. It is likely to match one of the catalog names from which positions of some sources were obtained. It may not match the position catalog for an individual source. A typical case, at the time this parameter was added, is that the flux densities come from one of Petrov’s catalogs while many of the sources use positions from a GSFC solution.

Below is a sample of a few sources from the March 2005 source catalog, slightly reformatted to fit the page width.

     RA=16:29:03.0298580 DEC= -20:26:55.100570  
     RAERR=  29.510 DECERR=  13.450 CALCODE=’V’  
     REMARKS=’VLBA Calib Survey - GSFC sols. - VCS2 - created 2004’ /  
     RA=16:30:11.2308370 DEC=  21:31:34.310260  
     RAERR=   4.370 DECERR=   5.380 CALCODE=’V’  
     REMARKS=’VLBA Calib Survey - GSFC sols. - VCS2 - created 2004’ /  
     RA=16:31:18.7777000 DEC=  10:52:02.460000  
     RAERR=   4.370 DECERR=   5.380 CALCODE=’M’  
     REMARKS=’JVAS-Browne etal. 1998, mnras, 293, 257; S8.4GHz= 72 mJy’ /  
     RA=16:31:16.5398860 DEC=  49:27:39.515680  
     RAERR=   0.520 DECERR=   0.630 CALCODE=’V’  
     REMARKS=’VLBA Calib Survey - GSFC sols. - VCS2 - created 2004’ /