Rapid ML and semi-automated methods for large-scale subsurface data interpretation and reservoir modelling: Applications for offshore freshened groundwater

With extreme climatic events and population growth predicted to continue increasing over the coming century, water stress across Europe (and elsewhere around the globe) is soon predicted to become critical. Offshore Freshened Groundwater (OFG) is being increasingly identified within continental margin sedimentary sequences worldwide, and has potential to be used as an industrial, agricultural, or potable resource, especially for draught mitigation during extreme climatic events. As part of an international effort under the Horizon Europe Water4All project RESCUE (RESources in Coastal groundwater Under hydroclimatic Extremes), we explore new methodologies to allow for the flexible and rapid identification, assessment, and modelling of OFG systems on a large scale. In this presentation, we share the preliminary work of the RESCUE project in exploring methods for the rapid and semi-automated detection of OFG from well logs, machine learning driven interpretation of seismic reflection data, and integrated porosity/permeability determinations from seismic attributes and well logs. The ultimate goal of this work will be to allow for the rapid generation of large-scale reservoir models, facilitating the dynamic modelling of high resolution and massive OFG systems with Parallel MODFLOW6.