Data-Driven Analytics for Integrated Fracture Characterization in a Lower Cretaceous Reservoir, Offshore Abu Dhabi

Abstract Fractures are crucial factors that affect the productivity and safety of hydrocarbon reservoirs. However, conventional methods of fracture detection and characterization are often limited by data availability and quality. In this study, we present a data-driven analytics approach that integrates four different seismic methods to predict and characterize fractures in a Lower Cretaceous Reservoir in an offshore oilfield in Abu Dhabi. The four methods are: Velocity Variation with Azimuth (VVAZ), Amplitude Variation with Azimuth (AVAZ), Amplitude Fracture Estimate (AFE), and Fault Likelihood. We apply these methods to 3D seismic volumes and verify their results with well data, drilling records, and well test data. We integrate seismic data, drilling records, and well test data to evaluate the results in characterizing the fracture zone within the Lower Cretaceous Reservoir, which exhibits strong anisotropy, significant time shifts on seismic data, and notable mud loss during drilling. The VVAZ and AVAZ methods estimate fracture orientation and density from azimuthal variations of velocity and amplitude, respectively. The AFE method measures fracture intensity from azimuthal fractional energy. The Fault Likelihood method detects fracture zones from seismic discontinuities related to faults. The results show that each method provides different aspects of fracture information that are complementary to each other. The VVAZ, AVAZ, VVAZ & AVAZ joint inversion methods reveal a dominant fracture orientation of NE-SW, consistent with regional stress directions, and a high fracture density in the central crest part of the field where mud loss was observed during drilling and high permeability zone was detected by well pressure test. The AFE method and the Fault Likelihood method identify several fault-related fracture zones that crosscut the field, some of which coincide with mud loss and pressure anomalies. This study demonstrates an integrated approach that combines the four fracture prediction methods and can provide a comprehensive and reliable fracture characterization in the Lower Cretaceous Reservoir. This is essential for optimizing drilling and production strategies, ensuring safe operations, and maximizing hydrocarbon recovery in line with ADNOC's strategic objectives for its offshore assets.