TITLE: Reservoir Simulation
Auth1: Matthias Imhof * mgi@vt.edu
Auth2: Roger Turpening
Auth3:
Auth4:
Auth5:
Auth6:

TEXT: (or see attached pdf file)

Oil, gas and water are produced daily in large quantities from subsurface reservoirs whose properties may not be well understood. The situation also occurs in reverse with the injection of fluids or gases for stimulation or waste burial. Clastic reservoirs are fairly well understood, but carbonate or hardrock environments where fractures dominate petrophysics are poorly understood. The major problem is the scale dependence of reservoir properties. One single feature just outside the examined rock volume may dominate reservoir properties. The measured properties become effectively dependent on the size of the sample. Moreover, ground truthing observations at reservoir scales may be prohibited by costs. We propose to simulate oil or gas reservoirs by injecting fluids into blocks of rocks ranging from cubic meters (pilot studies) to cubic kilometers. After injection, we will begin production of these fluids. In a later stage, we will apply methods of production enhancement such as fracturing. All stages will be monitored remotely with geophysical methods and selective drilling to determine the dominant petrophysical features. The last step will be drilling and mine back to examine which features controlled injection and production and whether they were observed geophysically. Seismic experiments can be performed over a broad range of frequencies (10 - 3000 Hz) which corresponds to wavelengths between 500 and 1 meters. The fundamental question is what features are actually captured at these scales? How large does the fracture density need to be before it has an observable effect? How do the different fracture scales interact with the different experimental scales? The proposed experiments should also be combined with geobiological ones because the fluid injection constitutes a controlled change of the biological environment, and the resulting biological communities may again affect the petrophysical properties. Small experiments may last for weeks, while the cubic kilometer scale experiments may last for fifty years. A completely characterized long term facility is necessary to perform such experiments.