marees terrestres bulletin d'informations - Université de la Polynésie ...

The instrumentation of the tidal station has improved significantly thanks to the
superconducting gravimeter (SG) OSG-050, installed in February 2007. The Pecný station as the
core station of ECGN (European Combined Geodetic Network) is equipped with permanent GNSS
station, absolute gravimeter (AG) FG5#215 and the superconducting gravimeter OSG-050. These
high quality instrumentations in the field of gravimetry allow to monitor wide range of gravity
variations of geophysical origin from Earth’s free oscillations to secular gravity variations.
Possibility of frequent simultaneous AG and SG observations at the station allow to solve main
problems of both instruments: drift and calibration of the OSG-050 and the offset variations of the
FG5#215.
2. Calibration and drift
The calibration factor and drift of the SG has been determined using simultaneous
observations with the AG (Hinderer et al. 1998). Altogether 15, typically one-day absolute
measurements with FG5#215 has been carried out from April 2007 to June 2008. For the purpose of
SG calibration, five AG campaigns has been extended to three-days observations. These extended
observations were carried out during tidal variations at least 230 Gal. The precision of all
individual determination of scale factor was better than 0.07%. However, the final calibration factor
and corresponding accuracy should be computed from results of repeated calibrations. The
dispersion of individual results with corresponding error bars in Fig. 2 show necessity of such
repeated measurements. The final calibration factor of the OSG-050 has been determined as average
of all calibrations . From dispersion of individual result we can assume that accuracy of the final
calibration factor is of about 0.06%.
Fig. 2. Calibration factor of the OSG-050 determined from 5 simultaneous three-days observations with
FG5#215. Error bars represents precision of individual calibrations.
The SG drift has been determined from the comparison between SG and AG observations
(see, Fig. 3 and Fig. 4) and described by linear term 1.7 0.4 Gal/year. We can assume, that after
removing drift from SG time-series, the rest of differences between AG and SG is caused mainly by
random and systematic errors of the AG. This approach can help to detect variations in AG offsets.
In our case (see, Fig. 4) all differences are within expected error bars of the AG (1.1 Gal, Niebauer
et al. 1995). Comparison of both techniques has been used for the evaluation of FG5#215
repeatability (precision) by such experimental way. The standard deviation of individual absolute
gravity measurements of 0.6 Gal respect to the OSG-50 observations describe the precision of the
FG5#215. It is necessary to say that the FG5#215 is not installed permanently at the Pecný station
and the precision of 0.6 Gal includes error of instrumental set-up, meter alignments etc.
11820