Over the past decades, numerous fluid indicators have been introduced and proven to be sensitive to hydrocarbon presence. Mathematically, fluid indicators measure the deviation of hydrocarbon-saturated rocks from their background in a specific domain or space. The ideal indicator is defined as an attribute that responds to hydrocarbon presence only; at shale- or brine-saturated intervals, it shows weak or no noticeable variation. In this paper, we collect and review several common fluid indicators used in the oil industry. The collection includes common reflectivity-based indicators and impedance-based indicators from the late 1980s to present. We present an overview of these fluid indicators, their past and current uses, their limitations, and their integration using machine learning. We examine these indicators on 35 data samples collected from different world basins and analyze their performance using recent data analysis techniques. We then apply the attributes on real seismic data from offshore Australia and test their ability to detect hydrocarbon-saturated reservoirs. In closing, we invoke machine learning approaches to combine the attributes and produce a fluid probability attribute.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.