The combined interpretation of full tensor gravity gradients and seismic data

Geophysical data joint inversion can effectively reduce the uncertainty of inversion results. In this paper, a joint inversion workflow is presented with gravity field, its gradients and seismic data sets in order to resolve geological structures details. It is well known that gravity gradient data is far more sensitive to sudden change of the interface and relative boundaries, which can identify interface mutation and reveal the shape characteristic of subsurface horizons correctly, leading to building geology-geophysics model effectively. Exclusively, traditional processing methods have been updated to cope with complicated models in terms of the property parameters as well as accuracy of resolved model, based on benefit from the combined inversion. Firstly, event identification technique is used to interpret seismic data and to create the layer distribution of the model. Secondly, new potential field interpretation method is applied to obtain parameters of the salt dome (depth, density and so on), layers and shallow sources, which result in building geology-geophysics model by means of the created parameter. Thirdly, seismic stochastic inversion is applied to estimate the impedance, as well as density distribution. The uncertainty analysis technique is also applied to estimate accuracy of the final results via evaluation of the inversion process. Finally, the gravity, magnetic fields and gravity gradient data sets are used simultaneously to estimate the layers of the model by applying the constraint of seismic data and log data sets. Based on the geology-geophysics model requests and the above information used, a fast simulated annealing (FSA) workflow is proposed to consolidate the joint inversion of gravity, gravity gradient and seismic data, and get the final geology-geophysics model. The theoretical test and practical application results show that the FSA method is useful to enhance the joint inversion of gravity field, gravity gradient and seismic data effectively, which can generate higher-resolution details for the building model and assist to provide more reliable information.