Seismic Avo Attributes and Machine Learning Technique to Characterize A Distributed Carbonate Build Up Deposit System in Salawati Basin Eastern Indonesia

Abstract

PT. RH Petrogas is the operator for acreage in the Salawati Basin in Eastern Indonesia and is interested in exploring this area further, particularly for the Miocene Carbonate or Limestone reservoir, one of the most challenging reservoirs that producing hydrocarbons. This Miocene carbonate build-up has proven to be a prolific oil reservoir in the Salawati Basin. The seismic data suggests the extension of this play to the north of this area, stratigraphically, without confirmation from a successful well penetrations yet. The integration of Seismic attributes, coherency, spectral decomposition, and seismic AVO attributes are seen as key tools in this exploration phase. However, a constraint on this workflow is the paucity of existing well data in the immediate area and discovery wells; adequate seismic interpretation tools are needed, especially with the help of the latest technology, to find optimal locations for increased drilling success. Moreover, to help increase oil production by exploring and exploiting hydrocarbon in this area. The study area incorporates some adjacent oil fields in the Salawati Basin covered by a single 3D seismic to understand the reservoir characteristics of the Miocene Carbonate and predict its distribution using the machine learning (ML) technique. The ML uses seismic AVO attributes data to generate a facies volume classification that can delineate and characterize a complex regional channel deposition system in a carbonate reservoir distribution in Miocene. Using the facies volume output from ML generated a sculping horizon-based target facies of the reservoir with rendering and geobody techniques can help to reveal and characterize a carbonate deposition channeling system over the Miocene Carbonate. The study suggests ML facies volume visualization techniques can help determine the characteristics model of a distributed carbonate clastic deposit system in Salawati Basin, Eastern Indonesia. The model shows some exciting lateral variations in reservoir intervals and reveals new insight and can help find the optimal location for new drilling targets more effectively, increase drilling confidence success, reduce cost and risk for optimizing exploration hydrocarbon strategy in the Miocene Carbonate of Salawati Basin.