Latitude observations at Jozefoslaw are made since 1959 with a visual zenith telescope (Zeiss No. 17224), using the Horrebow-Talcott method. A 12-group program, which has been revised twice in the past years, is used in routine observations. On average, about 150 group results are obtained per year, and the precision of a group result is on the order of 0.20''. To monitor the stability of the plumb line, a meridian gravimetric base-line of 7 points was established in 1976, with which the variations of the meridian component of the plumb line are measured 3-4 times a year basing on the measurements of differences in gravity accelerations. Scintrex, Sodin and Worden-Master precise quartz gravimeters had been used in the early stages of the gravimetric observations, with precision of the PLV on the order of 0.002''. Lacoste-Romberg's gravimeter - Model G No. 432 has been used since September 1987, and the precision has been improved (Barlik & Rogowski 1989).
The data used in this study are shown in Table 1 (click here), in which N is the number of the observations.
| Code | Type | Time Interval | N |
| LO | Latitude observations | 1976-1995 | 3028 |
| PL | Plumb line variations | 1976-1995 | 77 |
The reduction method for obtaining a latitude residual (LR) is the same as the one used in the previous work (Li 1996), but the polar motion results used in obtaining an LR time series are those from space techniques (Gross 1995).
To compare the LR and the PL with the same dates, a weighted mean of LR is first calculated at 30-day intervals. The new LR time series is smoothed(e = -11) (Vondrak 1969, 1977) and then interpolated at the same 77 dates of the PL time series. The 77-point time series of the LR(77) is then compared with the PL time series (see Fig. 1).
The band filters used and the calculating of the cross-correlation function between the LR(77) and the PL time series are the same as those used in the previous paper (Li 1996).
Figure 1: The PL time series and the processing of
the LR time series