Digital Outcrop Modelling of a Fractured Crystalline Basement: A Potential Unconventional Reservoir Outcrop Analogue

Abstract

Hydrocarbon exploration is typically limited to clastic or carbonate basins and lithologies. However, production from crystalline basement rocks is becoming more viable with the increasing price of oil and depletion of conventional plays. Indeed, the third largest oil field in the USA is the Wilmington Field in South California with reserve estimates predicted at three hundred million barrels of oil remaining from the initial three billion barrels. The field includes a production well in the fractured basement schist that has yielded more than twenty two million barrels of oil. Fracture characterisation in these rocks is important due to the propensity of fracture driven hydraulic regimes in basement rocks where porosity is often very low. This project utilises terrestrial lidar scanning to model fracture networks on a reservoir-scale basement unit in the north-west of England: The Borrowdale Volcanic Group (BVG). The BVG is a laterally extensive, highly heterogeneous 6 km thick unit, comprising mainly igneous rocks with associated sedimentary and metamorphic rocks. The rocks, which have a complex tectonic history, were formed by explosive caldera-type eruptions in the Late Ordovician (Caradoc) period, and include lavas, ignimbrites, sills and dykes. The model will be used as an analogue to create a fluid flow model. The region has existing data used in the abandoned plans to site a nuclear disposal facility in the area in the 1990's. Using the lidar data collected with a Riegl LMS-Z420i terrestrial laser scanner, a Digital Outcrop Model (DOM) has been constructed and the fracture distributions mapped using a combination of manual and automated approaches. The resulting fracture data are analysed to give statistical information on fracture dimension, distribution and spacing. Stochastically generated Discrete Fracture Networks (DFN's) have then been modelled from these fractures statistical data. Uncertainties in the data include biases due to exposure orientation and extent, resolution of lidar data and whether fractures are open or closed at depth (which may be estimated from the current stress regime). Multiple realisations of the DFN are generated to help take into account uncertainty in the fracture-derived statistics. The DFN is then used to calculate porosity and permeability values for the fracture network, which in conjunction with matrix porosity and permeability values form the basis for flow simulation of fluid movement through these rocks.

AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014