Up: Astrometry of satellites I 1990-1991
Table 4 shows that the errors in the present observations
arise mostly from errors in the star positions. Near opposition Uranus
regresses by roughly one chip width per night so that each night's
observations are measured relative to a nearly independent set of stars
from the
nights before and after. This was confirmed by breaking down the
observations into separate nights which revealed significant night to night
changes in scale and position angle.
Fundamental Right Ascensions and Declinations are not required for the
satellites because their theory can be improved simply from observations of
one satellite relative to another.
In Table 7 (only available in electronic form at the CDS
via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via
http://cdsweb.u-strasbg.fr/Abstract.html)
the satellite positions are presented as
standard co-ordinates
relative to a reference satellite
which is usually Oberon, but otherwise the next satellite observed nearer
the planet. In reducing the data the scale and rotation of position angle
with hour angle were fixed for the whole series and separate zero points
of position angle (J2000.0) were fixed for each year. The
constants were found from weighted means over the separate nights.
The formulae used were:
|  |
(1) |
|  |
(2) |
|  |
(3) |
|  |
(4) |
where
are measured in pixels on the chip
relative to the reference satellite, and
is the hour angle in hours. The standard errors are in units
of the least significant digit.
The standard co-ordinates were
corrected for refraction under a pressure of 760 mBar; the results
are presented in Table 7.
Up: Astrometry of satellites I 1990-1991
Copyright The European Southern Observatory (ESO)