Modelling the impact of depositional and diagenetic processes on reservoir properties of the crystal-shrub limestones in the ‘Pre-Salt’ Barra Velha Formation, Santos Basin, Brazil

Abstract

Abstract The Cretaceous Barra Velha Formation (‘Pre-Salt’) of the Santos Basin, Brazil, includes lacustrine crystal-shrub limestones which represent important hydrocarbon reservoirs. Neither single nor multi-phase flow behaviour of these unusual fabrics has been modelled to date. Here we quantify the impact of depositional and diagenetic processes on selected reservoir properties of the crystal-shrub limestones using highly flexible 3D synthetic pore-scale models that can explore parameter space. Rather than producing models validated to actual reservoir properties, this seeks to establish trends and identify emergent clustering behaviour. Shrubby aggregates nucleate from a surface to form upward and outward-growing, branching, fascicular optical calcite crystal fans that are diverse in size and form. The model allows exploration of the impact of primary growth parameters: number of bifurcations (or branching), shrub height, angle of cone (or fan) inflation as well as subsequent inter-shrub cementation and dissolution, on resultant porosity, permeability, and multiphase flow. Simulations show that depositional parameters, as well as two major diagenetic processes - cementation and dissolution - control porosity and permeability. In some cases, these trends diverge considerably from widely-referenced Kozeny-Carman (KC) porosity-permeability relationships. The angle of shrub inflation produces data clusters that traverse KC bands. By contrast, the number of bifurcations and the degree of dissolution follows KC trends, so providing guidance for potential rock type subdivision of similar lithologies. On this basis we propose that the crystal-shrub limestones in the Barra Velha Formation can be subdivided into three broad petrophysical reservoir rock types based on the degree of shrubby aggregate bifurcation and extent of dissolution. The trends explored here can be used to link conceptual and process-based geological models to data, so enabling more accurate prediction and upscaling of heterogeneous reservoir properties and their reservoir-scale distribution.