Poststack inversion of Seismic data for Total Organic Carbon Estimation using neural network: A case study from the Barnett shale

The main goal of this paper is to use the multilayer perceptron neural network to suggest a 3D cube of the Total Organic Caron (TOC) in the lower Barnett shale gas reservoir. 3D Poststack seismic data recorded near two horizontal wells drilled in the lower Barnett formation are used as an input to train the neural machine, while the calculated acoustic impedance from sonic and density well-logs data for these wells are used as an output. The Multilayer perceptron neural machine is trained in a supervised mode and weights of connections are calculated, the whole 3D seismic data are then propagated through this machine and a cube of acoustic impedance is inverted. A cross-plot of the acoustic impedance versus the total organic carbon is used to provide a linear relationship between these two parameters; this relationship is used to suggest a 3D TOC cube. Obtained results are compared with the TOC of another horizontal well drilled in the lower Barnett, they shows the ability of the genetic inversion to enhance shale gas reservoirs characterization.