Validating induced seismicity forecast models—Induced Seismicity Test Bench

Abstract

Induced earthquakes often accompany fluid injection, and the seismic hazard they pose threatens various underground engineering projects. Models to monitor and control induced seismic hazard with traffic light systems should be probabilistic, forward‐looking, and updated as new data arrive. In this study, we propose an Induced Seismicity Test Bench to test and rank such models; this test bench can be used for model development, model selection, and ensemble model building. We apply the test bench to data from the Basel 2006 and Soultz‐sous‐Forêts 2004 geothermal stimulation projects, and we assess forecasts from two models: Shapiro and Smoothed Seismicity (SaSS) and Hydraulics and Seismics (HySei). These models incorporate a different mix of physics‐based elements and stochastic representation of the induced sequences. Our results show that neither model is fully superior to the other. Generally, HySei forecasts the seismicity rate better after shut‐in but is only mediocre at forecasting the spatial distribution. On the other hand, SaSS forecasts the spatial distribution better and gives better seismicity rate estimates before shut‐in. The shut‐in phase is a difficult moment for both models in both reservoirs: the models tend to underpredict the seismicity rate around, and shortly after, shut‐in.

Seismicity Induced by Simultaneous Abrupt Changes of Injection Rate and Well Pressure in Hutubi Gas Field

“…Detailed numerical modeling of stress, pore pressure, and fault failure process would provide quantitative assessments of various physical models for explaining the induced seismicity (Catalli et al, 2013;Gischig & Wiemer, 2013;Goebel et al, 2016;Guglielmi et al, 2015;Hornbach et al, 2015;Király-Proag et al, 2016;Samuelson & Spiers, 2012;Shapiro et al, 2010). However, such modeling would require not only the production/exploitation data, such as injection rates, injected volumes, and well pressures, but also the detailed geological and hydrogeological information of the area and mechanical properties of the faults that have not been characterized in Hutubi gas field.…”

Evidence for a transient hydromechanical and frictional faulting response during the 2011 Mw 5.6 Prague, Oklahoma earthquake sequence

“…However, the seismic hazard model did not relate any injection well operational parameters, such as injection pressure or injection rate, to changes in seismicity. Physics‐based models capable of assessing seismic hazard related to induced seismicity, such as those described by Norbeck and Horne [] and Király‐Proag et al [], require detailed information about the hydraulic and mechanical properties of the faults that exist in the model. It remains difficult to use traditional reservoir engineering approaches, for example, pressure transient analysis, to measure the in situ properties of basement faults that would be necessary to inform physics‐based models of induced seismicity.…”

Hierarchical Bayesian Modeling of Fluid‐Induced Seismicity

“…Moreover, once the project is under way and physical information becomes available, the Bayesian framework enables the computation of posterior distribution for the model's parameters, the formulation of predictive models for the Poissonian process and a robust forecasting strategy. Although we demonstrate that the proposed rate model fairly accurately describes the selected data sets, different models (e.g., based on geomechanical principles, Catalli et al, 2016;Gischig & Wiemer, 2013;Goertz-Allmann & Wiemer, 2012) or an ensemble of different models (Király-Proag et al, 2016) can be used without altering the structure of the proposed framework.…”