picture of CEAE webpage

Numerical Modeling of Induced Seismicity: Background, Code Development, and New Results for the Porosity-Pressure Constiutive Relationship

Mr. Danny Birdsell

Ph.D Candidate, Dept. of Civil, Environmental & Architectural Engineering, University of Colorado, Boulder

11:00 am Wednesday October 18, 2017 

Engineering Center ECCE 1B41

 

Abstract: Human-caused earthquakes are becoming more common in Oklahoma and Colorado with the rise of wastewater
injection deep underground.  Pore fluid pressure, in-situ stress state, fault orientation, and the properties of rocks and fractures all help dictate if induced seismicity will occur.  In the first part of the talk, we discuss how the code PFLOTRAN can be used to faithfully model key aspects of induced seismicity such as the coupled nature of fluid pressure and geomechanics, the ability for hydraulic diffusivity to change as a function of effective stress and shear failure history, and the importance of thousands of fractures and faults both mechanically and hydrologically in induced seismicity. PFLOTRAN is a massively-parallelized, coupled poromechanical flow code, but incorporation of fracture
capability must be futher developed.  Once implemented, this tool should offer insights into the spatiotemporal patterns of induced seismicity, seismogenic permeability, and the effects of background stress state and fault orientation on induced seismicity likelihood.
In part two of the talk, we examine mathematically and numerically how subsurface flow codes that can accept a number of constitutive relations for porosity, fluid density, and fluid flux can be used to simulate saturated flow.  Examples of these codes include PFLOTRAN, FEHM, STOMP, TOUGH2, and the oil and gas simulator BOAST.  We show that the governing equations of these codes often differ from the groundwater flow equation by not properly accounting for the motion of liquid relative to solid.  We
propose an alternate porosity-pressure relationship that brings the governing equations of these subsurface flow codes into agreement with the groundwater flow equation.  This alternative porosity model results in much better agreement with the Thesis solution, and it should be incorporated into these codes.