Forecasting Fluid-Induced Seismicity Using Machine Learning Time-Series Prediction

Abstract Seismicity induced by fluid injection (enhanced geothermal systems [EGSs], waste water disposal, and CO2 storage) remains a significant risk to communities and industry because it may cause structural damage and economic losses. In some cases, this can lead to the immediate shutdown of the project, as observed at the Pohang geothermal site in South Korea (2017), after inducing a magnitude 5.5 earthquake. Consequently, it is crucial to develop new strategies to anticipate and mitigate the seismic risk posed by these operations. In this work, we present an innovative machine learning-based approach to forecast the seismicity induced by fluid injection into the ground. It leverages a time series of various injection parameters, such as injected volume, flow rate, and wellhead pressure, to capture their relationship with the seismicity rate. Once our model is trained, we forecast the future number of induced earthquakes on a fixed timescale of interest. We apply this strategy to two case studies that have caused induced seismicity at different spatial and temporal scales. First, we investigate the relationship between the massive wastewater injection in Oklahoma since 2006 and the increase in seismicity. Capturing this relationship with linear regression, it is then possible to forecast the seismicity caused by any future hypothetical injection scenario. Second, we study the seismicity induced during the hydraulic stimulation of the Soultz-sous-Forêts (France) EGS in 2005. Using a gradient boosting model, we predict the seismicity over the next 24 hr, relying solely on injection parameters. Our results show that, while model selection should account for site complexity, machine learning can effectively predict seismicity rates using injection parameters, even with limited training data. This work paves the way for real-time applications, such as predicting microseismicity during ongoing operations at the Rittershoffen EGS site, 6 km from Soultz-sous-Forêts.