Spatio-temporal evolution of hypocenters and moment tensors derived from time-reverse imaging

• DOI: 10.5194/egusphere-egu24-14828
• Publication Date: 2024-03-09T02:09:04Z
• Authors: Claudia Finger; Katinka Tuinstra; Peter Niemz; Peidong Shi; Laura Ermert; Federica Lanza

The location and focal mechanism of microseismicity induced during fluid injection experiments in geothermal wells can be used to infer the extent fracturing orientation local stress field. Passive seismic instrumentation is typically deployed at surface boreholes around site monitor microseismic activity. Recent methodological advancements enable locating often thousands events a timely fashion. However, determination mechanisms limited selected number larger-magnitude events.   Time-Reverse Imaging (TRI) exploits time-invariancy elastic wavefield propagate backwards time from stations through an adequate velocity model. Under ideal conditions, will converge on initial source origin time. excellent accuracy for with signal-to-noise ratios smaller than one capability determining moment tensor each locatable event has been demonstrated controlled synthetic real studies. Numerous improvements adaptations have proposed augment resulting image volume identify individual TRI promising one-stop solution analyzing but two major disadvantages: (1) computation needed simulate high-frequency prohibits analysis continuous hours or days recordings, (2) identification volumes susceptible overestimating due noisy images. Alterations concept based pre-computed Green’s functions exist provide near-real require compromises terms minimal ratio. cannot identified yet these types methods. Thus, main challenge applying reducing computational time, while retaining most beneficial capabilities. This balancing act requires careful possible scaling simulation parameters. Here, we apply recordings created sensor setup experiment April 2022 UtahFORGE test site. A combined model complex geology network including locations fiber optic cables, deep shallow borehole sensors, nodal sensors. data demonstrate gain using multiple sensors speed characteristic applied prior back propagation. Accuracy are compared cases optimize their trade-off. Finally, instead individually picking events, usefulness interpreting spatio-temporal evolution hypocenters tensors directly results.