I/325 Radio-optical reference frame link (Zacharias+ 2013)
Radio-optical reference frame link using the USNO astrograph and deep CCD
imaging (optical position results)
Zacharias M.I., Zacharias N.
<U.S. Naval Observatory, Washington DC (2013)>
=2013yCat.1325....0Z 2013yCat.1325....0Z
=2014AJ....147...95Z 2014AJ....147...95Z
ADC_Keywords: Fundamental catalog ; Positional data ; VLBI
Keywords: astrometry ; reference systems ; QSO
Abstract:
Between 1997 and 2004 several observing runs were conducted mainly
with the CTIO 0.9m to image ICRF counterparts (mostly QSOs) in order
to determine accurate optical positions. Contemporary to these deep
CCD images the same fields were observed with the USNO astrograph in
the same bandpass. These provide accurate positions on the
Hipparcos/Tycho-2 system for stars in the 10 to 16 magnitude range
used as reference stars for the deep CCD imaging data. Here we present
final optical position results of 413 sources based on astrograph data
reductions using UCAC2 and UCAC4 type reference stars. These optical
positions are compared to radio VLBI positions. The current optical
system is not perfectly aligned to the ICRF radio system with rigid
body rotation angles of 3 to 5 mas (3σ) found between them for
all 3 axes. Furthermore, statistically, the optical-radio position
differences are found to exceed the total, combined, known errors in
the observations. Systematic errors in the optical reference stars as
well as physical offsets between the centers of optical and radio
emissions are both identified as likely causes. A detrimental,
astrophysical, random noise (DARN) component is postulated to be on
about the 10mas level. If confirmed by future observations, this could
severely limit the Gaia to ICRF reference frame alignment accuracy to
an error of about 0.5mas per coordinate axis with the current number
of sources envisioned to provide the link. A list of 36 ICRF sources
without the detection of an optical counterpart to a limiting
magnitude of about R=22 is provided as well.
Description:
The following tables provide the observed, optical positions of 413
ICRF sources on the Hipparcos Catalogue Reference System (HCRS), as
well as position differences to the ICRF radio positions and auxiliary
data. Details of the project are described in the Astronomical Journal
paper with the same title.
Most targets were observed with the Cerro Tololo 0.9m telescope, some
with the Kitt Peak 2.1 m telescope. All targets are between about -86
and +32 degree declination spanning all 24 hours in RA. The link to
the primary optical coordinate system is provided by USNO astrograph
observations performed within weeks of the deep CCD imaging. Primary
reference stars are drawn from the Tycho-2 Catalogue, excluding stars
fainter than about V = 11. The astrograph data provides secondary
reference stars in the 10 to 16 mag range used to reduce the deep CCD
exposures. For both the deep and the astrograph data the same bandpass
(579 to 643 nm) was used with custom filters. Excluding outliers the
RMS radio-optical position differences are about 25 mas per coordinate.
Thus these data constitute the most precise of its kind so far.
The astrograph data were reduced with 2 different procedures,
following the pixel reduction schemes of the UCAC2 and UCAC4 type
reductions, respectively. The main differences are:
- UCAC2-type reductions utilize Gauss function image profile fits,
followed by empirical corrections of x,y data due to pixel phase
errors, while UCAC4-type reductions use modified Lorentz profiles for
stellar image fits, matching the observed PSF very well.
- UCAC2-type reductions use linear corrections as a function of
various parameters to mitigate systematic position errors caused by
the poor charge-transfer efficiency of the astrograph CCD, while
UCAC4-type reductions utilize bin-by-bin empirical corrections.
The UCAC2-type reductions show slightly smaller scatter in the optical
to radio position differences, while the UCAC4-type data lines up with
the radio coordinate system axes better.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
icrf-u2.dat 193 666 Results from UCAC2-type reductions of ICRF
optical counterparts (sample in table 4 of paper)
icrf-u4.dat 193 682 Results from UCAC4-type reductions of ICRF
optical counterparts (sample in table 5 of paper)
--------------------------------------------------------------------------------
See also:
I/289 : UCAC2 Catalogue (Zacharias+ 2004)
I/324 : UCAC4 Catalogue (Zacharias+ 2012)
VII/258 : Quasars and Active Galactic Nuclei (13th Ed.) (Veron+ 2010)
I/323 : International Celestial Reference Frame 2, ICRF2 (Ma+, 2009)
J/A+A/520/A113 : VLBI detection of 398 extragalactic radio sources (Bourda+, 2
J/A+A/526/A102 : VLBI imaging of 105 extragalactic radio sources (Bourda+, 201
J/A+A/553/A122 : Time stability of the ICRF2 axes (Lambert, 2013)
J/A+A/537/A99 : Large Quasar Astrometric Catalogue 2 (LQAC-2) (Souchay+ 2012)
http://www.obs.u-bordeaux1.fr/BVID/content.html : the Bordeaux VLBI Image
Database
Byte-by-byte Description of file: icrf-u2.dat icrf-u4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 2 I2 h RAh Right Ascension ICRS (hours)
4- 5 I2 min RAm Right Ascension ICRS (minutes)
7- 13 F7.4 s RAs Right Ascension ICRS (seconds)
15 A1 --- DE- Declination ICRS (sign)
16- 17 I2 deg DEd Declination ICRS (degrees)
19- 20 I2 arcmin DEm Declination ICRS (minutes)
22- 27 F6.3 arcsec DEs Declination ICRS (seconds)
30- 40 F11.8 h RAhour Right Ascension ICRS (decimals hours)
43- 53 F11.7 deg DEdeg Declination ICRS (decimals degrees)
55- 60 F6.1 mas Dist [0/1821] Distance between the optical and
radio positions
62- 68 F7.1 mas dRA* Distance(optical-radio) in RA*cos(Dec)
70- 76 F7.1 mas dDec Distance(optical-radio) in Dec
78- 81 F4.0 mas eRA [4/250] total, formal error of optical
position (RA*cos(Dec))
83- 86 F4.0 mas eDE [4/250] total, formal error of optical
position (Dec)
88- 92 F5.1 --- sRA ratio dRA*/eRA (position difference in RA
in σs)
94- 98 F5.1 --- sDE ratio dDE/eDE (position difference in Dec
in σs)
100-103 F4.1 mas eRad ?=99 error of radio position (per coordinate)
106-107 I2 --- Nccd number of deep CCD exposures used
109-112 I4 --- Nr* average number of reference stars per deep
exposure used
114-116 I3 --- Nrx average number of reference stars excluded
119-122 F4.1 --- err mean "plate" solution error of individual
deep CCD exposures
125-128 I4 --- S/N [1/1639] Approximate signal/noise ratio of
target on individual deep CCD
130-134 F5.2 mag mag ?=0.0 Optical brightness from LQAC-2
(J/A+A/537/A99)
136-140 F5.3 --- z [0/5]? Redshift from LQAC-2 (J/A+A/537/A99)
142 I1 --- T [0/3] Object type: 0=unknown, 1=AGN, 2=BL-Lac,
3=QSO (from Veron's 13th ed. (VII/258)
144-147 F4.2 --- S [0.8/7.2]?=9 Radio X-band (8-12GHz) structure
index from Bordeaux database: 1=very good,
2=good, 3=use with caution, 4=unacceptable
149 I1 --- U [2/4] UCAC-type reductions (2 or 4)
151-158 A8 --- Rname B1950 coordinate based name of target
160 A1 --- Run [abo-zJ] Code of deep CCD observing run (1)
162 A1 --- n_Run [rc] Type of target: r=regular target,
c=also used as calibration field
164 A1 --- Q [ufplgn] Quality letter of optical image (2)
167-193 A27 --- Notes Remarks; if a double the separation (pixel)
and flux ratio (source/companion) is given
--------------------------------------------------------------------------------
Note (1): the runs are assigned a 1-letter code, as follows:
o = CTIO 0.9m Dec 1997
p = CTIO 0.9m Feb 1998
q = CTIO 0.9m Jul 1998
r = CTIO 0.9m Aug 1998
s = CTIO 0.9m Jan 1999
t = CTIO 0.9m Mar 1999
u = CTIO 0.9m Jun 1999
v = CTIO 0.9m Sep 1999
w = CTIO 0.9m Dec 1999
x = CTIO 0.9m Mar 2000
y = CTIO 0.9m Jun 2000
z = CTIO 0.9m Sep 2000
a = CTIO 0.9m Jan 2001
b = KPNO 2.1m Jun 2001
J = CTIO 0.9m May 2004
Note (2): the quality is coded as follows:
u = unconfirmed, too faint for unique ID
f = faint optical source (S/N ≤ 5)
p = problem case, (≥ 5σ position difference) or blended image
l = large (opt.-radio) position difference (≥ 3σ)
g = good optical source
n = no ICRF2 radio position, instead following was used:
for 0447-439: Perlman et al. 1998AJ....115.1253P 1998AJ....115.1253P;
for 2316-423: Roopesh 2009, private communication
--------------------------------------------------------------------------------
History:
* 12-Nov-2013: the files were supplied by Norbert Zacharias
(files were named "fintab.all2" and "fintab.all4")
* 10-Dec-2013: comment the meaning of the 'n' quality letter.
(End) Norbert Zacharias [USNO], Francois Ochsenbein [CDS] 12-Nov-2013