Joint Inversion of Teleseismic Delay Times and Gravity Anomaly Data for Regional Structures

Abstract

A new algorithm is presented for the joint inversion of teleseismic delay times and gravimetric data. It is assumed that an approximate linear relationship between density and velocity anomalies exists, the value of which is unknown and may vary in different regions of the model, especially with depth. Therefore, the correlation coefficient is treated as additional unknown parameter, leading to a non-linear (quasi Gaussian) inver- sion scheme. The algorithm uses a Bayesian approach with the possibility of including a priori information and variable degree of relaxation of the condition of linear correla- tion between density and velocity anomalies. The versatility of the approach is demon- strated with the inversion of synthetic models with variable degree of complexity, repre- senting the lithospheric structure of a typical continental rift.