Data such as cores and borehole logging give a very detailed description of how the subsurface looks at the location of the well. However, heterogeneity in the subsurface – especially in fluvial sediments like we have here in the Netherlands – has a large influence of how subsurface can vary from one location to the next. Reservoir modelling tackles this problem and aims to provide a complete view of what the reservoir behaves like.
Designing geological models of subsurface reservoirs is a crucial aspect to estimate how much energy is available underground, to support reservoir management decisions, and to identify and mitigate operational risks. Good reservoir models have several requirements: (i) They need to capture the magnitude and scale of heterogeneity as well as the degree of reservoir connectivity and complexity of the flow paths; (ii) they need to adequately account for geological uncertainties by considering different geological scenarios; (iii) they need to be agile so they can be continuously updated as new data become available; (iv) they need to provide reliable uncertainty bounds for production forecasts that allow us to make reservoir management decisions that optimise production and minimise risks; (v) they need to be built with a purpose and allow us to challenge our understanding of the subsurface geology and its impact on the dynamic reservoir behaviour.