3D object-oriented image analysis in 3D geophysical modelling: Analysing the central part of the East African Rift System

• Implementation of 3D object oriented analysis in 3D geophysical modelling.
• Implementation of object-based inversion approach to estimate density values.
• Comparing the combined satellite and surface data gravity models with satellite only gravity models for subsurface modelling.

Non-uniqueness of satellite gravity interpretation has traditionally been reduced by using a priori information from seismic tomography models. This reduction in the non-uniqueness has been based on velocity–density conversion formulas or user interpretation of the 3D subsurface structures (objects) based on the seismic tomography models and then forward modelling these objects. However, this form of object-based approach has been done without a standardized methodology on how to extract the subsurface structures from the 3D models. In this research, a 3D object-oriented image analysis (3D OOA) approach was implemented to extract the 3D subsurface structures from geophysical data. The approach was applied on a 3D shear wave seismic tomography model of the central part of the East African Rift System. Subsequently, the extracted 3D objects from the tomography model were reconstructed in the 3D interactive modelling environment IGMAS+, and their density contrast values were calculated using an object-based inversion technique to calculate the forward signal of the objects and compare it with the measured satellite gravity. Thus, a new object-based approach was implemented to interpret and extract the 3D subsurface objects from 3D geophysical data. We also introduce a new approach to constrain the interpretation of the satellite gravity measurements that can be applied using any 3D geophysical model.