Intra- and Inter-Facies Variability of Multi-Physics Data in Carbonates. New Insights from Database of ALBION R&D Project

Abstract In carbonates, the geological facies is a key driver for populating reservoir models with petrophysical properties. Conventionnal core analysis mainly contributes to establish relationships between facies, petrophysics and geophysics. However, populating gridblocks reservoir models with petrophysics requires parsimonious facies classifications and effective relationships at larger scales that field studies rarely investigate. Studying outcrop analogues helps filling the gap between lab measurements and effective upscaled properties of models, and considerably improves the modelling workflows. The ALBION R&D project developed an innovative framework for multi-physics and multi-scales characterization of Barremian-Aptian carbonates from south-eastern France. These outcropping rudist-rich limestones constitute an analogue of Middle-East reservoirs. Petrophysical and geophysical properties were measured on plugs from cores and outcrops but also at larger scales thanks to original experiments on cores, in and between boreholes. Indeed the analogue includes several experimental areas, where hydraulic tests in sealed wells sections and tomographies between very close boreholes allowed investigating petrophysical and geophysical rock properties at intermediate decimetric to decametric scales. Thanks to the resulting database, this paper aims quantifying the variability of multi-physics data (e.g. porosity, permeability, and P-wave velocity) at different scales in regards of an updated and unified facies classification. The latter is only based on sedimentary origin and fabrics. Other available properties affecting petrophysics are used to cluster facies associations in sub-classes. Consequently the facies classification does not allow discriminating the distributions of porosity, permeability, nor p-wave velocity. For the rudist facies, that is the most sampled, texture subclasses do not help this work. Reversely, the place of sampling, that is likely a proxy of diagenesis and age, cluster the petrophysical distributions. The results remind us that a proper facies definition should consider both sedimentary origin, fabrics, texture, diagenesis and tectonics. They also point out the relative importance of each characteristics in regards of the scale of interest and the difficulty to infer upscaled relationships between rock properties from CCAL because the representative elementary volume of carbonates is usually higher than the plug and even the core volumes.