Detection of overpressure zones and a statistical model for pore pressure estimation from well logs in the Krishna‐Godavari Basin, India

Abstract Abnormally high pressures, measured by repeat formation tester (RFT) and detected by well log data from 10 wells in the Krishna‐Godavari (K‐G) Basin, occur in the Vadaparru Shale of Miocene and Raghavapuram Shale of Early Cretaceous age. Overpressures generated by disequilibrium compaction, and pore pressures have been estimated using the conventional Eaton sonic equation with an exponent of 3.0. The observed abnormal pore pressure gradient ranges from 11.85 to 13.10 MPa/km, whereas fracture pressure gradient varies from 17.40 to 19.78 MPa/km. The magnitude of vertical stress (S v ) has a gradient from 21.00 to 23.10 MPa/km. The minimum horizontal principal stress (S h ) magnitude is found to vary from 64 to 77% of the S v in normally pressured to overpressured sediments. A multiple linear regression model with a squared multiple correlation coefficient ( R 2 ) of 0.94 is proposed for pore pressure prediction from gamma ray, density and sonic logs to focus on efficient drilling operations and to prevent borehole instability. The statistical model has been calibrated with the RFT data from five wells covering about 3400 sq. km area of the onshore K‐G Basin. The model predicted pore pressure values are in close agreement with the actual RFT data for another four wells including a well in the offshore K‐G Basin. Hence, the proposed regression model may be useful for predicting pore pressure from other well logs in the K‐G Basin.