3 Precision estimation of SIPA1 and comparison with meridian catalogs

The rms differences of SIPA1 and Hipparcos catalogs are $\pm 0\hbox{$.\!\!^{\prime\prime}$}118$ and $\pm 0\hbox{$.\!\!^{\prime\prime}$}095$ for $\Delta\alpha\cos\delta$ and $\Delta\delta$ respectively. It can be seen from Fig. 1a and Fig. 3 that the large rms difference for $\Delta\alpha\cos\delta$ is mainly caused by those stars brighter than 3th magnitude, and that the right ascension system of SIPA1 coincides with the Hipparcos system very well, except for those bright stars. The amplitude of systematic errors in declination, $\Delta\delta_\delta$, are about $-0\hbox{$.\!\!^{\prime\prime}$}03$ to +0$.\!\!^{\prime\prime}$03. Removing the systematic differences from SIPA1 using Eq. (2) and Eq. (3), the remain differences are shown in Figs. 4a,b. And the rms differences are reduced to $\pm 0\hbox{$.\!\!^{\prime\prime}$}070$ and $\pm 0\hbox{$.\!\!^{\prime\prime}$}072$ for $\Delta\alpha\cos\delta$ and $\Delta\delta$ respectively.

To judge the level of Photoelectric Astrolabe catalog, we compared SIPA1 with the most productive automatic meridian circles, Carlsberg, Bordeaux and Tokyo. First, the CAMC7 (CAMC Consortium 1992) and TMC89 [, (Yoshizawa et al. 1993)] are compared with the Hipparcos catalog respectively. After removing systematic differences, the rms differences of $\Delta\alpha\cos\delta$ and $\Delta\delta$are calculated. For the Bordeaux meridian circle, the comparison between Hipparcos and 173 radio stars observed with the Bordeaux meridian circle is used [, (Requième et al. 1995)]. The results are given in Table 1.

  

Figure 4: a,b) The difference between SIPA1 and Hipparcos catalog after removing the systematic errors. a) $\Delta\alpha\cos\delta$ vs. $\delta$.b) $\Delta\delta$ vs. $\delta$