Bayesian estimation of scale factor for CG-5 relative gravimeter

Abstract

The terrestrial time-variable microgravity survey is an important means to monitor geodynamic processes underground. The Scintrex CG-5 relative gravimeter, which is the mainstream instrument for terrestrial relative gravity measurement at present, its scale factor (SF) is an important factor affecting the quality of the observation data. Unlike the conventional gravity adjustment procedure, this study takes account of the nonlinear drift rate of the CG-5 gravimeter, depending on the known absolute gravity (AG) in the network, and the scale factor is considered as one of the hyper-parameters to be estimated by means of Akaike’s Bayesian information Criterion (ABIC). In order to quantify the errors caused by the bias of scale factor, we applied this new approach to process actual gravity survey campaign data in Yunnan gravity network from 2018 to 2020, then analyzed the residuals of gravity differences (GD) between station pairs and the difference of GD between two relative gravimeters (CG5 #1169 and CG5 #1170). The cross-validation of the absolute gravity (AG) from the quasi-synchronous observation in the network is also used in the paper. It has shown that this new approach is effective to improve the accuracy of scale factor and adjustment results.