Mechanical Parameters Controlling Sandstone Intrusion Emplacement and Geometry

Abstract

Post-depositional remobilization and injection of sand are now recognized as common and important processes in deep-water clastic systems. Like igneous intrusions, sandstone intrusions exhibit geometries such as dykes, sills, cone- and saucer-shaped or wings. To form, these features require hydraulic fracturing of the overburden, fluidization and injection of depositional sand into the contemporaneous fracture network. Understanding mechanisms governing their overall architecture is essential because they provide conduits for fluids through impermeable overburden and they can act as hydrocarbon reservoirs (intrusive traps). The reservoir modelling of large-scale sandstone intrusion reservoirs is challenging due to imperfect characterisation of the intrusions, deficiencies in existing reservoir modelling approaches, and lack of geostatistics and associated modelling rules. The scaled physical modelling primarily addresses this problem by providing analogues which can be used to populate reservoir models when subsurface data are inadequate. Here, we use physical experiments to simulate sandstone intrusion emplacement. By progressively increasing (200Pa/min) the porewater pressure in a reservoir consisting of glass microspheres, a glass microsphere/water mixture was eventually injected into a contemporary hydraulic fracture network through low permeability host rock (a sand-gelatine mixture). The materials were scaled to allow comparison with large-scale geometries observed in the subsurface. We tested the influence of three parameters on the geometry of the intrusion complex: (1) reservoir geometry, (2) reservoir depth and (3) overburden cohesion. The experiments show that dome-shaped reservoirs favoured “wing” development at shallow depth, whereas narrow reservoirs favoured the formation of vertical dykes. Increasing depth caused wings to be initiated closer to the reservoir central axis up to a critical depth where dykes formed at the top of dome-shaped reservoirs. Physical experiments are a powerful tool to assess the dynamic of emplacement during intrusion processes. Fracture propagation velocity increased with decreasing overburden cohesion, with increasing depth and with increasing fracture dip. These experiments constrain intrusion architecture function of overburden behaviour and function of shape, depth and lateral extent of the overpressure source. These results may be helpful for 3D seismic intrusion recognition and interpretation.

AAPG Datapages/Search and Discovery Article #90216 ©2015 AAPG Annual Convention and Exhibition, Denver, CO., May 31 - June 3, 2015