252
Vondrak & Ron
4.
Discussion
and
conclusions
First of all, one should notice rather big di erences among the trends (sine
columns, rst three rows) in longitude: 2.156(OA99/1), 1.516(OA99/2) and
2.990(VLBI) and in obliquity:
;
0
:
093(OA99/1),0.007(OA99/2)and 0.220(VLBI).
These di erences are de nitely much larger than their formal errors but, on the
other hand, the di erences between OA99/1 and VLBI are smaller than those of
our preceding solution OA97 (Vondrak et al., 1998b) that yielded 1.543mas/yr
in longitude and 0.131mas/yr in obliquity. The correction to precession is now
(solution OA99/1) much closer to the VLBI value than before. This change is
very probably due to our less tolerant approach to Hipparcos proper motions
(20% of them being corrected in OA99 in contrast to only 10% in OA97). Also
interesting is the obvious change of trend in longitude around 1956, noticed al-
ready in the OA97 solution by Bizouard et al. (1997). This e ect can probably
be also caused by imperfections of the Hipparcos proper motions. (Di erent
stars were observed in the second half of the century than in the rst one.)
Secondly, there is a discrepancy in the cosine (out-of-phase) term of the
principal nutation in longitude between both optical solutions and VLBI (they
di er in sign). This is the very term that is highly correlated with the trend
in the VLBI solution. This also holds for the cosine term in longitude with
a period of 3398 days (see also Yaya et al., (2000) for a similar discussion).
The largest di erences between optical astrometry and VLBI are however found
in the semiannual and annual terms. In addition, the values found from the
present solution OA99 di er substantially from OA97. It is clear that this is the
weakest point of optical astrometry. We suspect that this e ect is mostly due to
proper motions of some of the stars. Di erent stars at di erent right ascensions
are observed during the year (Only the `night' stars are observed.), and if their
positions di er from reality at a certain part of the sky they can produce an
annual wave in celestial pole o sets. All the other terms agree quite well, on the
level of their formal standard errors.
Acknowledgments.
This study was made possible thanks to the support through
the Key Project No. K1003601 nanced by the Academy of Sciences of the Czech
Republic.
References
Bizouard, Ch., Capitaine, N., Ron, C. & Vondrak, J., 1997, in Proc. IAU Coll.
165 Dynamics and astrometry of natural and arti cial celestial bodies, I.
Wytrzyszczak, J. Lieske, R.A. Feldman (eds.), Kluwer, 481{486.
Feissel, M. & Mignard, F., 1998, Astron. Astrophys.,
331
, L33{L36.
Kovalevsky, J., Lindegren, L., Perryman, M.A.C. et al., 1997, Astron. Astro-
phys.
,
323
, 620{633.
Lieske, J.H., Lederle, T., Fricke, W. & Morando B., 1977, Astron. Astrophys.,
58
, 1{16.
Seidelmann, P.K., 1982, Celest. Mech.,
27
, 79{106.
Vondrak, J., Feissel, M. & Essa , N., 1992, Astron. Astrophys.,
262
, 329{340.
