Gravity anomalies of pull-apart basins having finite strike length with depth dependent density: A ridge regression inversion

Abstract

An inversion using ridge regression to estimate simultaneously the parameters of pull-apart basins having finite strike length (2.5D) and regional gravity background from observed gravity anomalies is presented. A parabolic function is used to describe the density contrast variation with depth within the structure. The algorithm begins with initializing both the regional background and parameters of the basin and subsequently improves them iteratively until the modelled gravity anomalies mimic the observed ones. The applicability and efficacy of the inversion is demonstrated with a set of synthetic gravity anomalies 1) attributable entirely due to a theoretical model, 2) in the presence of pseudorandom noise and 3) in the presence of both pseudorandom noise and regional gravity background. It was found from the analysis of synthetic gravity anomalies that the modelled parameters of the structure closely mimic the true ones even when the gravity anomalies are corrupted with pseudorandom noise. In the presence of both random noise and regional background the estimated parameters deviate only modestly from the assumed ones. Furthermore, the applicability of the algorithm is exemplified with a derived density-depth model to analyse the Bouguer gravity anomalies observed over the Ranigunj basin, India. The estimated depth of the basin is consistent with the available borehole information. The interpretation of the basin supports the hypothesis that this basin might have been formed as a result of both E-W kinematics and orthogonal extension rather than simple local extensional tectonics. © 2009 European Association of Geoscientists & Engineers.