Abstract

Hydrocarbon exploration involves significant risks, necessitating methodologies that improve target identification confidence and operational discovery accuracy. Low-Frequency Passive Seismic (LFPS) has emerged as an innovative complementary solution, which uses ambient seismic noise rather than the traditional active sources, making it cost-effective and sustainable. This paper highlights the successful application of LFPS in Akholjuni field of Cambay Basin, India, achieving a remarkable correlation accuracy in hydrocarbon prediction, thus serving as a robust de-risking tool. The LFPS process begins by deploying very highly-sensitive broadband seismometers to the survey area & acquiring continuous passive recordings (≈15–20 hours per station). Given the uncontrolled nature of ambient noise, comprehensive quality-control (QC) steps are applied. Using an in-house processing flow, the data are corrected for instrument response, despiked, cross-correlated, and analyzed for bandlimited V/H amplitude ratios (β). Selected spectral components are decomposed via SVD and weighted using supervised learning to produce probabilistic anomaly maps that are validated against well information and other operational studies.