Automatic gravity inversion for simultaneous estimation of model parameters and regional gravity background: an application to 2D pull-apart basins

Abstract

An automatic inversion using Marquardt algorithm has been developed to simultaneously estimate the parameters of 2D pull-apart basins and regional gravity background from the observed gravity anomalies. The density contrast is assumed to vary continuously with depth based on a parabolic equation. The salient feature of this inversion is that the initial parameters of the basin are not a required input. The applicability and efficacy of the inversion is demonstrated with a synthetic model of a pull-apart basin. The synthetic gravity anomalies have been analysed (i) solely due to the structure, (ii) in the presence of a regional gravity background, and (iii) in the presence of both random noise and regional gravity background. The analysis showed that the modelled structure closely mimics the true one irrespective of the degree of the polynomial that is used to describe the regional background. In the presence of both random noise and regional background, the estimated dips of the boundary faults and the regional background show modest variation from the assumed ones. Two real-field anomalies observed over a set of pull-apart basins associated with the suture zone of the Eastern Ghats Mobile Belt in Northeast India have been interpreted. The estimated depth in each case is consistent with the available borehole information.