Optical observations of astronomical time and latitude are based on the local plumb line (vertical) at the observational site. Variations of the plumb line are considered as errors in astronomical observations. On the contrary, plumb line variation (PLV) related to the deformation of the geoid is an important scientific object capable of being measured in practice, with which geodynamics and some other sciences may be studied in a better way. Thus, both the astronomical and geodetical communities have long been interesting in measuring the PLV. Due to the difficulties in separating it from the other sources, such as star catalog error, observational error caused by atmospheric refraction, and the inaccuracy of the ERP determination, the problem remains unsolved. The high accuracy of ERP determination by space techniques since 1976, the promising prospect of a high-precision catalog after the Hipparcos Space Mission, the possibility of estimating atmospheric refraction by the NASA global atmospheric data (Schubert et al. 1993), all of these progresses in science and technology will soon provide us a new foundation in studying PLV.
New efforts have already been put into practice by the IAU and IAG since the beginning of the 1990's. As a first step, studies concentrate on the possibility of deriving long-term variations in the deflection of the vertical from astrometric observations (IAU Comm. 19 1991). Other than the efforts already made recently (Ishii et al. 1990; Li & Gambis 1994; Li 1995, 1996), one of the best ways to confirm the actual ability of astrometric techniques in measuring the PLV is to compare their results directly to the corresponding results derived from another technique. Fortunately, the Astronomical-Geodetic Observatory at Jozefoslaw is the place where regular latitude observations have been carried out since 1961, and gravimetric measurements have been used to calculate the PLV along the meridian there since 1976 (Rogowski & Barlik 1993). Thus, there are already 20 years of parallel observations of the two techniques concerned with the determination of PLV. The purpose of this paper is to compare the PLV results of the two techniques, from which a conclusion might be drawn on the reality of PLV results derived from the two techniques.
We restrict ourself here only to the interannual part of the PLV, mainly due to two points: the existence of star catalog errors in the astrometric observations, which are an order greater than the PLV we are going to detect; the sparse distribution of the gravimetric measurements in the case of Jozefoslaw observatory.
With the restriction of the data used in the paper, it is difficult for us to solve all of the problems in one paper. We expect that the evidence shown in the paper and an understanding of the errors of astrometric observation of the PLV at the site of an observatory, their existence in latitude or time observations, their possible influnce on astronomical studies already done in the past, and the ability of astrometric techniques in measuring them at interannual time scales, may be recognized.