Search

SUMMARY We describe a 3-D discrete fracture hybrid model (DFHM) that returns forecasts of both induced seismicity and of power generation in an Enhanced Geothermal System (EGS). Our model considers pore-pressure increase as the mechanism driving induced seismicity, similarly to other hybrid models, but it uses discrete fracture modelling for flow and heat that allows accurate and realistic transient solutions of pore pressure and temperature in fractured reservoirs. Earthquakes and flow are thus considered as closely coupled processes. In the DFHM model, the creation phase of an EGS is described as a Markovian process with a transitional probability that encapsulates the irreducible uncertainty with regards to induced seismicity. We conditioned this transitional probability on field observations from the 2006 EGS project in Basel, achieving a good match with observations of seismicity evolution. Specifically, our model effectively reproduces and explains the observed long-term exponential decay of seismicity after the well was shut in, suggesting that pore pressure diffusion in a critically stressed fractured reservoir is sufficient to explain long-lasting post-injection seismic activity as observed in Basel. We then investigate alternative injection scenarios, using Monte Carlo simulations to capture the uncertainties in fault locations and stressing conditions. We show that the number of induced events depends not only on the total injected volume but also on the injection strategy. We demonstrate that multistage injection schemes are superior to single-stage ones, since the former are associated with less seismic risk and can generate at least the same revenue in the long term.

Reykjavik is almost entirely heated by geothermal energy. Yet, recent growth of the city significantly increased the heat demand. Past experiences in Iceland's capital region showed that hydraulic stimulation of existing geothermal wells is suited to improve hydraulic performance and energy supply. However, fluid injection may also trigger felt or even damaging earthquakes, which are of concern in populated areas and pose a significant risk to stimulation operations. Consequently, soft stimulation concepts have been developed to increase geothermal well performance while minimizing environmental effects such as induced seismicity. In a demonstration project of hydraulic soft stimulation in October 2019, more than 20.000 m³ of water were injected into well RV-43 in Reykjavik in multiple stages and with different injection schemes. The hydraulic performance of the well was improved without inducing felt seismicity. An a priori seismic risk assessment was conducted and for the first time the risk was continuously updated by an adaptive traffic light system supported by a sophisticated realtime microseismic monitoring. Our results confirm that it is possible to improve the performance of geothermal wells in Reykjavik and worldwide with acceptable technical, economic, and environmental risks. Here we provide an overview of the entire stimulation project including site description, stimulation design, zonal isolation, logging, seismic risk assessment and mitigation measures, realtime seismic, hydraulic and chemical monitoring, and stimulation results and challenges., Geothermics, 96, ISSN:0375-6505

The rapid increase in energy demand in the city of Reykjavik has posed the need for an additional supply of deep geothermal energy. The deep-hydraulic (re-)stimulation of well RV-43 on the peninsula of Geldinganes (north of Reykjavik) is an essential component of the plan implemented by Reykjavik Energy to meet this energy target. Hydraulic stimulation is often associated with fluid-induced seismicity, most of which is not felt on the surface but which, in rare cases, can be a nuisance to the population and even damage the nearby building stock. This study presents a first-of-its-kind pre-drilling probabilistic induced seismic hazard and risk analysis for the site of interest. Specifically, we provide probabilistic estimates of peak ground acceleration, European microseismicity intensity, probability of light damage (damage risk), and individual risk. The results of the risk assessment indicate that the individual risk within a radius of 2 km around the injection point is below 0.1 micromorts, and damage risk is below 10−2, for the total duration of the project. However, these results are affected by several orders of magnitude of variability due to the deep uncertainties present at all levels of the analysis, indicating a critical need in updating this risk assessment with in situ data collected during the stimulation. Therefore, it is important to stress that this a priori study represents a baseline model and starting point to be updated and refined after the start of the project.

At Geldinganes Island, Reykjavik, Iceland a hydraulic stimulation was recently conducted to enhance the productivity of an existing hydrothermal well. An experimental cyclic soft stimulation concept was applied. Seismic risk was assessed with an appropriate monitoring network which was set up and operated before, during, and for some time after the stimulation activities. An advanced traffic light system was developed and operated for the first time in this setup.A crucial element in such traffic light systems is the real-time monitoring of background and induced seismicity. During the experiment, real-time seismograms from the monitoring network were streamed over the internet to three different institutions (ISOR, ETHZ and GFZ), where they were analysed independently, with different combinations and setups of automatic, semi-automatic and manual methods. Both, classic pick based approaches and modern full-waveform methods were applied. Locations, magnitudes, and centroid moment tensor solutions were determined.Many things can go wrong in real-time or near-real-time processing of seismic data. Sensor failures, transmission failures, timing issues, processing hardware failures, computational limitations, software bugs and human error, just to name a few. In a temporary network the challenges are additionally salted by the need to validate sensor responses, orientations, gain factors and site conditions in a short time frame between station setup and beginning of the experiment. Furthermore, tuning of advanced analysis methods can be difficult without example events at hand.In this contribution, we would like to share our lessons learned in near-real-time processing of data from a heterogeneous temporary seismic network.

Author(s): Broccardo, Marco; Mignan, Arnaud; Grigoli, Francesco; Karvounis, Dimitrios; Rinaldi, Antonio Pio; Danciu, Laurentiu; Hofmann, Hannes; Milkereit, Claus; Dahm, Torsten; Zimmermann, Gunter; Hjorleifsdottir, Vala; Wiemer, Stefan

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-For\^ets 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., Comment: http://onlinelibrary.wiley.com/doi/10.1002/2016JB013236/abstract

© 2016 The Authors. This work investigates the injection-induced seismic response of a heterogeneous fault plane, featuring low-permeability asperities embedded into a high-permeability damage zone. We simulate the pressure evolution with a hydrogeological simulator, accounting for the heterogeneous fault plane. Seismicity occurs then on the asperities, represented as unstable patches reactivating by means of the Mohr-Coulomb criterion. The hydrological and seismic modules are implicitly coupled to account for effects of asperity reactivation on the permeability. Results show that permeability changes may cause at a later time a change in seismicity propagation. We also investigated such effects by varying the density of asperities.

The seismic risk associated with deep fluid injection in Enhanced Geothermal Systems can be mitigated by stopping reservoir stimulation when the seismic risk becomes unacceptable or by reducing production flow rates when seismicity occurs during the operational phase. So far, none of these mitigation measures have been included in the Levelized Cost Of Electricity. A meta-model is introduced that estimates the optimal price of electricity, based on an analytical geothermal energy model, and updates this cost to include the outlay for mandatory seismic risk mitigation measures. The proposed energy model computes both electricity production and heat credit. The costs added during reservoir stimulation are based on the probability of abandoning an injection well, based on a traffic-light system, defined as the ratio of scenarios that exceed a given seismic safety threshold in the risk space. In the production phase, the net energy generated is reduced by clipping the production flow rate so that the reservoir's overpressure does not exceed the regional minimum effective stress. Based on a generic geothermal triplet, we investigate the trade-off between heat credit and seismic risk mitigation cost. The added cost, mostly due to financial risk aversion, shifts the optimal site for a plant from between a few kilometres to tens of kilometres away from populated areas, for increasingly vulnerable building stocks. Finally, using a simple yet realistic optimisation strategy, we study the role that a seismic safety standard plays for determining the number of EGS plants that can be sited in a given region. ISSN:0306-2619 ISSN:1872-9118