We discuss acquisition of seismic velocity models needed to use passive seismic data to locate drilling targets in convective geothermal flow systems. The requisite models are derived by applying Fast Marching computation to observed travel time data between wellbore sources and surface sensors. It is a truth universally acknowledged [J Austen 1813], that well drilling uncertainty and risk is stalling growth of geothermal power production. The high flow demand of turbines combined with high spatial erratics in crustal flow systems creates high initial risks that paralyse geothermal project decision-making. As a remedy, we propose that multi-channel seismic data acquisition and processing proven for shale formation O/G production can image the erratic flow structures of convective geothermal systems in volcanic terrains. We assess the current impasse by first noting that crustal reservoir flow systems are spatially erratic at all scale lengths. Spatially-correlated crustal porosity φ(x,y,z) leads to poro-connectivity percolation clustering at all scales to generate the permeability empiric κ(x,y,z) ~ exp(αφ(x,y,z)). The value of the constant α, fixed by the observation that αφ ~ 3-4 for a wide range of crustal flow systems, guarantees that crustal well productivity is lognormally distributed. Lognormal distributions are consistent with flow observations for active and fossil geological flow systems worldwide. While hydrocarbon production pay can cover drilling costs across the entire lognormal productivity range, only the few high-flow wells can do so for convective geothermal systems. All other geothermal wells are sunk costs. On a well-by-well basis, geothermal well pay is given by resource temperature T and flow velocity structure V, Q ≡ ρCTV. The smoothly varying temperature field T can be remotely estimated to spatial resolution of 500m-1km. To date, however, there is little or no ability to remotely estimate flow structures V at 500m-1km spatial resolution, let alone achieve a nominal 50m spatial resolution needed to enable cost-effective production well drilling. It is thus essential for future convective geothermal development to apply new technologies to locate high fluid flow clusters at ~ 50m resolution in advance of costly exploration drilling. Proven shale formation flow imaging technology at 25m resolution proceeds via acquisition of standard surface seismic reflectivity velocity models. Adapting the demonstrated flow imaging technology to convective geothermal systems reduces to acquiring an adequate seismic velocity model of the target crustal volume. Velocity model data acquisition for a 3km x 3km x 2km volume of volcanic terrain can proceed via seismic refraction travel-times from downhole sources to ~ 1000 surface sensors. Downhole seismic source energy can be safely and reliably provided by deflagration burns of propellant charges of 30MJ (energy equivalent to a 2000 cubic inch marine airgun). Embedding the observed travel-time data in a 500 x 500 x 500 node target volume model at 6m ≡ 2ms resolution allows accurate computation of travel-times for a large range of nominal target volume source points to surface array sensors. As demonstrated by active flow imaging for shale formations, such a travel-time table accurate to 5m/2ms applied to convective geothermal flow systems is sufficient to process ambient seismic listening data from surface sensor arrays into reliable images of convective geothermal flow structures for production well targeting.

Press the Back button in your browser, or search again.

Copyright 2020, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.