- Lists of baseline corrections
- The basics
- Format and use of lists of baseline corrections
To access the Karl G. Jansky Very Large Array (VLA) baseline corrections for a given year, select
the year below and press "View".
To access the pre-upgrade VLA baseline corrections for a given year, select
the year below and press "View".
NRAO monitors the positions of the VLA antennas on a regular
basis -- roughly monthly, and more often during reconfigurations.
This involves observing a number of calibrators spread around the
sky, solving for the antenna-based gains, and from their behavior
deriving the three-dimensional positions of those antennas (see
e.g. chapter 12 of Thompson, Moran, and Swenson 1991,
Interferometry and Synthesis in Radio Astronomy, for a
fuller discussion). These corrections (generally less than a
centimeter) are then inserted into the correlator model in the
For observers the main point is that the positions used during
correlation may not be correct, leading to phase errors and
incorrect gridding of the uv-data (see Section 4.4.4 of the AIPS Cookbook).
More importantly, better positions may have become available
between the time your data was taken, and the time you reduce them.
This is particularly likely if your observations took place during
or shortly after array reconfigurations. The files given here list
the NRAO determinations of the corrections to the antenna
positions, the date on which those corrections were found, and the
date on which they were entered into the system.
Applying corrections in CASA
Use the CASA task gencal, with caltype='antpos'
for VLA data (after March 1, 2010), and
caltype='antposvla' for pre-upgrade VLA data (before January 12,
Applying corrections in AIPS
The AIPS task VLANT will do the full operation for you.
A small section of the baseline corrections file might look like this:
1997 BASELINE CORRECTIONS (IN METER)
MOVED OBSDATE Put_In_ MC(IAT) ANT PAD Bx By Bz
MAY28 JUN10 JUN12 19:50 7 W16 0.0000 0.0000 -0.0144
MAY28 JUN10 JUN12 19:50 10 W10 0.0036 0.0000 0.0000
MAY27 JUN10 JUN12 19:50 21 E2 0.0030 0.0000 0.0000
JUN10 JUN12 19:50 25 W14 0.0000 0.0025 0.0000
JUN10 JUN12 19:50 26 W12 0.0032 0.0000 0.0000
JUN23 JUN24 JUN25 14:54 2 N18 0.0033 0.0152 -0.0060
MAY27 JUN24 JUN25 14:54 6 E14 0.0049 0.0000 0.0000
JUN24 JUN25 14:54 7 W16 0.0061 0.0000 0.0000
JUN24 JUN25 14:54 11 N16 0.0041 0.0000 0.0000
JUN23 JUN24 JUN25 14:54 15 N14 -0.0117 -0.0150 -0.0096
JUN24 JUN25 14:54 25 W14 0.0040 0.0000 0.0000
JUN24 JUN25 14:54 28 E16 0.0048 0.0000 0.0000
JUN27 JUN28 00:46 1 N12 0.0030 0.0000 -0.0019
JUN24 JUN27 JUN28 00:46 3 N10 -0.0051 -0.0110 0.0000
JUN27 JUN28 00:46 7 W16 0.0000 0.0022 0.0000
JUN24 JUN27 JUN28 00:46 14 E18 0.0072 -0.0082 0.0059
The meaning of these columns is as follows:
- ANT MOVED
- the date on which the antenna was moved
to the indicated pad. A blank in this field does NOT mean
"ditto" -- it means "not moved". Corrections in these entries
have been re-determined relative to the previous entry for that
antenna, i.e. these are corrections to the previous position.
- the date on which the observations used to
determine these corrections were taken.
- the date on which these corrections were added
to the positions in the ModComps (the computers which run the VLA).
- the IAT at which these corrections were entered
into the ModComps.
- the antenna number, as given e.g. in the AN table
- the pad on which that antenna was sitting when the
observations were made (as listed e.g. by the AIPS task PRTAN).
- Bx, By, Bz
XYZ are in a right-handed Cartesian system. X and Y lie in a
plane parallel to the equator. X points toward the intersection of the
local meridian and the celestial equator, Y points East, Z points to
the North pole. (reference: Thomson, Moran & Swenson, 2nd edition
2001, p86, chap. 4.1)
These corrections are ADDITIVE. To determine
the appropriate correction for your observations, you should add up all
corrections measured between the MOVE date and the next move after your
observing run, excluding any corrections entered into the ModComps before
your observing began.
EXAMPLE: Based on the above listing, antenna 7 was moved to
pad W16 on 28 May 1997. The 10 June 1997 baseline run gave a -0.0144m
correction to its z-coordinate, which was entered into the on-line system on
12 June 1998 at 19:50 IAT. Further corrections were determined on the 24th
and the 27th. Let's assume this is the whole story, i.e. no further
measurements were made before the telescope was moved to a new pad. If you
observed between 28 May and 12 June, you would have to enter the sum of all
three corrections; between 12 and 25 June, the sum of the 24 and 27 June
corrections; beteen 25 and 28 June, only the 27 June correction; thereafter,
no correction is needed.
In general, one should use the sum of all available corrections. However,
antennas may indeed shift slightly over time, even when staying on the same
pad. The test for an individual data set is either to solve for the
antenna positions yourself, using those data; or to see what happens to your
phases before and after these corrections, and judge for yourself whether
the corrections have improved matters.
The procedure for applying these corrections inside AIPS is discussed in
section 4.4.4 of the
If you have any additional questions regarding baseline corrections, please
contact us through the NRAO help desk.