| Abstract |
Standard treatments of EGS microseismicity (Meqs) stimulated by fluid injection into the ambient crust assume a downsized version of the rapid planar dislocation slip mechanics developed for fault-zone earthquakes. Generic to these EGS Meq treatments are three assumptions: (i) spatial variations in crustal fluid-rock interactions are uncorrelated; (ii) spatially averaged crustal rock-fluid interactions equate to a quasi-uniform elastic continuum that ruptures via rapidly propagating stress singularities in response to local stress orientations; (iii) the resultant collection of spatially uncorrelated planar stimulation dislocation slip surfaces form “cubic-law†flow channels connecting EGS injector/producer well pairs. We note here large volumes of data that dispute these assumptions. First, well-log, well-core, and well-flow empirics show that spatially correlated “pink noise†porosity distributions φ(x,y,z) control ambient crust rock-fluid interaction as given by the poro-permeability empiric κ(x,y,z) ~ exp(αφ(x,y,z)). Second, we show here through the seismic emission waveforms emitted by the dislocation slip mechanics of Haskell (1969) that Meq waveforms recorded at 2.5km depth in crystalline basement rock arising fluid injected at 6km depth have spectral distributions that radically differ from spectra expected from rapid dislocation slip processes. Third, we further note that the stimulation Meq waveform spectra detailed here are consistent with slow spatially erratic rupture propagation within pre-existing ambient crust permeability structures given by κ(x,y,z) ~ exp(αφ(x,y,z). In view of ambient crust poro-permeability empiric κ(x,y,z) ~ exp(αφ(x,y,z)) as supported by detailed Haskell Meq waveform spectral systematics, it is difficult to credit the standard hypothetical EGS stimulation process based on uncorrelated crustal rock-fluid interactions. |