CO2 Storage Site Assessment from Hydrocarbon Field Revision Using a High-resolution Stratigraphic Framework and Machine Learning

Summary Saline aquifers are broadly considered as best candidates for CO2 storage projects, and Relative Geological Time ‘RGT’ modeling ( Pauget et al., 2009 ) has already been integrated with Machine Learning (ML) techniques to evaluate such reservoirs’ potential (SNS Vision project, Reiser et al., 2024 ; Northern Lights JV project, Legeay et al., 2024 ). This method furthermore provides a wide range of tailored workflows to rejuvenate matured and abandoned hydrocarbon fields, broadening perspectives from seismic data portfolio revision, green field exploration and experimental analogue modeling. Results of subsurface characterization are shown for a series of challenging tectono-stratigraphic settings, providing an overview of the potential and perspectives of this combined RGT-ML approach for CO2 storage site assessment.