Monitoring microseismicity of the Hengill Geothermal Field in Iceland

Article Geochemistry & Geophysics

A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland

Nima Nooshiri et al.

Summary: Despite the complexity and time-consuming nature of current inversion and modeling techniques, this paper presents a novel approach based on deep learning for rapid source-parameter estimation of microseismic earthquakes. The approach employs convolutional neural networks and a fully connected neural network to obtain moment tensor and spatial location information. The algorithm can be trained using synthetic seismic data only and is robust against perturbations such as observational noise and data incompleteness. The method achieves excellent performance and shows good agreement with standard methodology, making it a viable option for real-time estimation of source parameters in microseismicity.

GEOPHYSICAL JOURNAL INTERNATIONAL (2022)

Add to Collection

Article Multidisciplinary Sciences

A risk-based approach for managing hydraulic fracturing-induced seismicity

Ryan Schultz et al.

Summary: The study developed a risk-informed strategy for selecting red-light thresholds for immediate well shut-in during hydraulic fracturing operations in the Eagle Ford shale in southern Texas. By simulating impacts, it was found that impacts were greater in the northeast and smaller in the southwest of the area, driven by concentrations of population density. The method provides guidance for managing induced seismicity risks through traffic light protocols.

SCIENCE (2021)

Add to Collection

Article Geosciences, Multidisciplinary