Defining the basic rules that describe long-term shoreface dynamics: A process-mimicking approach for reservoir modelling

Producing geologically reasonable subsurface reservoir models which also honor wellbore and seismic data is still a major challenge for geoscientists at a range of energy and mining industries. Models based purely on geostatistics usually don’t satisfy the expected representation of geological concepts, while process-based models can hardly be constrained by subsurface data. Our research project GEOPARD aims to add a new tool in this context by providing a process-mimicking approach. In this method, sedimentary units are placed in 3D space constrained by subsurface data while simultaneously following a set of geological rules. This contribution aims to share and discuss the methodological workflow under development, from the geological domain, to test this method on the shoreface accumulation system. The first major step of the workflow is to specify the geological unit(s) to be modelled, which must be of interest as a control upon reservoir heterogeneity and of an adequate scale for representation. Afterwards, geological rules can be defined. These can be classified as (i) element geometry, scale, and infill specifications, (ii) basic element-set rules, which define the expected relationship between a set of geological units, and (iii) advanced element-set rules, which define how these relationship between elements might change over a sedimentary sequence. Each set of rules is linked to appropriate references for justification and exemplified from virtual outcrops of the Blackhawk Formation at Central Utah, which is one of the most studied and best-preserved examples of this type of depositional system. Feedback from the modelling domain is itself incorporated into refining the geological rules, resulting in an iterative effort to improve the methodology. Testing the method in these type of accumulation systems, that can be said to follow relatively straight-forward rules, are key before advancing into systems which typically exhibit more complex sedimentary architecture. We expect that the development of this modelling method can be a powerful tool to produce more useful models for oilfield assessment and development, CO2 storage projects, hydrocarbon exploration, among many others.