Abstract

This paper presents a robust workflow for reconstructing water saturation in the Bassein Formation using a field-wide, regression-based J-function approach. Conventional upscaled saturation led to non-physical oil cut in most wells due to misrepresentation of capillary behavior. By deriving J-function exponents from well log data and mapping them spatially, a more realistic and zone-specific saturation distribution was achieved. The revised model aligned closely with log-derived saturation and resolved initialization anomalies. This methodology enhanced dynamic model reliability and supports more accurate water injection planning, offering a scalable solution for heterogeneous carbonate reservoirs with limited core data.