Final Report Summary - GEISER (Geothermal Engineering Integrating Mitigation of Induced Seismicity in Reservoirs)
Executive Summary:
GEISER worked specifically at developing a better understanding of the key parameters that control induced seismicity in response to an injection. This improved understanding serves as input for the necessary seismic hazard and risk assessment of future EGS projects. As a basis for all analyses the GEISER research team has analyzed a series of past stimulation experiments from projects such as Soultz-sous-Forêts (France), Basel (Switzerland), Berlìn (El Salvador) and Gross Schönebeck (Germany) as well as from numerous Icelandic fields, where injections caused seismicity. In addition, datasets from The Geysers in California and from Cooper Basin (Australia) were provided to GEISER.
Analyses of the temporal and spatial evolution of the average size distribution and stress drop of the thousands of seismic events observed during the injections at the various sites revealed a pattern with the largest seismic events often occurring in the hours and days after the stop of injection. The relation between injected hydraulic energy and released seismic energy seems to be characteristic for many projects. This observation provides clues to the development of mitigation strategies and soft stimulation techniques such as a cycling of the injection pressure.
The relative size distribution of micro-earthquakes at sedimentary seems to be different from sites accessing reservoirs in or near the granitic bedrock in the Rhine valley. In some cases, a lower seismic hazard is evident in sedimentary rocks.
From the joint research efforts, GEISER presents a number of results and recommendations of major relevance for geothermal operations:
For the first time, ground motion prediction equations have been derived specifically for geothermal sites. At geothermal sites small magnitudes, high frequencies and short distances are much more relevant than for normal tectonic hazard assessment. Ground motion prediction equations forecast the level of shaking expected at the surface for a given seismic event. They are a key ingredient in any seismic hazard assessment.
A comprehensive framework for the assessment of seismic hazard and risk was developed. The framework is probabilistic in nature, allowing the representation of the uncertainty in numbers and in our understanding of the processes to be captured. GEISER also provides guidelines for safe and reliable EGS operations. Key is a dynamic –forewarning- advanced traffic light system that can be used in all phases of a project. This approach allows adjustment the operational conditions to mitigate unsolicited effects and to improve system performance. It incorporates site specific parameters, derived from the geological studies, from drilling as well as logging.
A strategy is proposed to enhance public support for EGS projects, based on lessons learned from past projects. A cost-benefit balance for the stakeholders throughout the entire exploration and production workflow is important, capable of identifying and properly addressing different interests and risks regarding a specific EGS project. Special attention has to be paid to risk perception. Nuisance and trivial damage should be addressed with care and considered as a significant project risk. For non-structural damage, a pre-agreed procedure is needed to evaluate and compensate the costs.
The GEISER research efforts and guidelines are important steps to enable the efficient and safe use of deep geothermal energy resources throughout Europe.
The GEISER results also point to future challenges. Especially, the modeling approaches and software developed for the assessment of induced seismicity now needs to be validated against a wider range of experimental data, including future injections. Further improvements depend on our ability to image larger and potentially tectonically loaded faults and to incorporate empirical experience. However our means to forecast a distinct seismic response in the underground will remain limited.
Project Context and Objectives:
Geothermal operations require production and injection of fluids. In the case of Enhanced Geothermal System (EGS) operations are usually related to engineering of hot, low permeability rock. Similar operations are common, and will be applied in the future to a large extent in conventional fracture dominated geothermal systems. The reason for this is that geothermal energy can only be extracted from a small volume of hot rock surrounding the fracture systems. Permeability enhancement in boreholes and creation of artificial fracture for production, therefore, are considered crucial for future development of the reservoirs. The same applies to the re-injection wells in fracture-dominated reservoirs. The fluid injected must go either into open fractures that are a part of a tectonic fault system, or into new fractures that must be created. In general, the basic knowledge on design and safe operation of fluid injection is based on a geologic map with fault/fracture distribution, tectonic movements, stress field, and natural seismicity with estimates of the moment magnitude of the largest seismic event to be expected in the area of interest.
The need to inject fluids to enhance productivity of a geothermal well often induces seismicity. Earthquakes induced by human activities occur in most mining-related operations, depletion of oil and gas reservoirs, fluid injection in the subsurface and dam impoundment and they often reduce public acceptance of such ventures. Promising geothermal projects jeopardized by this problem are at Soultz-sous-Forêts and at Basel. In the latter case, repeated seismic events, although not destructive, were felt by the local population and prompted the authorities to halt operations. To avoid these problems, action has to be taken in order to better understand and mitigate induced seismicity in the development of geothermal reservoirs.
An improved understanding of the causes and processes leading to induced seismicity in geothermal engineering and the development of appropriate mitigation strategies were the general objectives of the GEISER project. In particular, specific goals were defined
to understand why seismicity is induced in some cases but not in others
to determine the potential hazards depending on geological setting and geographical location
to work out licensing and monitoring guidelines for local authorities, which should include a definition of what level of ground motion is acceptable
to develop strategies for successful stimulation and improvement of the hydraulic properties of the geothermal reservoir without producing large magnitude events that pose a threat to buildings and disturb the public.
These objectives were approached in several work packages, addressing specific aspects of the general topic.
WP 2: Compilation of induced seismicity data from geothermal sites. The compilation served as a basis for all subsequent analyses. In this WP, data on induced seismicity from representative sites were collected, to provide an overview of lessons learned from previous experience. Data were checked and homogenized for comparison and further use in the other workpackages.
WP 3: Analysis of Induced Seismicity. Different seismological approaches were applied and further developed to analyse data sets of induced microseismicity from geothermal areas. Based on the evolution of induced seismic activity in space and time, the interrelation between the specific local geological settings, injection parameters and the occurrence of fluid-injection induced seismic events was addressed.
WP 4: Understanding the Geomechanical Causes and Processes of Induced Seismicity. Different modelling approaches and laboratory experiments were performed to investigate some of the key factors influencing induced seismicity: The goal was to come to a better understanding of the basic physical mechanisms that induce microseismicity and to deliver management and production scenarios with relative estimates of the stress-state changes for different geothermal settings.
WP 5: Seismic Hazard Assessment. On the basis of the results from WPs 3 and 4 and considering natural seismicity, WP5 assembled all the components of hazard assessment to be conducted before the selection and start of operations at an EGS site, to result in Guidelines for best practice. This included the analysis of the natural background seismicity, an estimate of the ground shaking produced by the microseismicity induced during the initial stimulation phase and the probability of triggering a large earthquake (M>4) ahead of its natural time of occurrence, either during the stimulation phase or during the long-term EGS operation.
WP 6: Strategies for EGS operations with respect to Induced Seismicity (Mitigation). In this workpackage strategies for operators and regulatory bodies were proposed to develop guidelines for the selection, licensing and long-term operation of EGS sites in different geological settings. Recommendations for proper monitoring of induced seismicity are given, and for a minimization of the risk associated with induced seismicity.
GEISER worked specifically at developing a better understanding of the key parameters that control induced seismicity in response to an injection. This improved understanding was part of the seismic analysis in WP3 and the mechanical considerations in WP4 and serves as input for the necessary seismic hazard and risk assessment of future EGS projects. For this purpose, the GEISER team has analyzed a series of past stimulation experiments from projects such as Soultz-sous-Forêts (France), Basel (Switzerland), Berlìn (El Salvador), various Icelandic sites where re-injection of thermal fluids caused seismicity and Gross Schönebeck (Germany). In addition, datasets of seismic events recorded at The Geysers in California were provided to GEISER as well as stimulation data including the corresponding seismicity from Cooper Basin (Australia).
As a result of the seismic analyses, GEISER provided Guidelines for techniques/methodologies for seismological investigations to be applied in future EGS operations, developed on the basis of successful analyses of past sequences (Deliverable D3.3). In these guidelines it is recommended that all observables (1) in-situ stress, 2) hypocenter locations, 3) seismic moments, 4) stress drop and 5) b-value) should be investigated during or immediately after stimulation as their combination allows direct estimates of the spatial and temporal evolution of pore pressure in the stimulated rock volume. Thus, semi-automatic methodologies need to be developed to determine these parameters in near-real-time. The last three points, if available in real-time, constitute also essential input to advanced, dynamic traffic-light systems that are based on probabilistic seismic hazard assessments during ongoing stimulation.
To investigate the effects of changes in pore pressure, temperature and the effect of pre-existing fault segments, laboratory experiments and several modeling activities with complementary approaches were performed throughout WP4. Different numerical tools have been developed, or adapted, in order to study the coupled mechanisms that induce seismicity during EGS operations. The efforts put in WP4 also include the development of tools to simulate catalogs of synthetic events, based on rock mass physical features and on injection rates, that are useful in order to assess the seismic hazard associated to EGS operations.
Another ingredient to understand and mitigate induced seismicity was provided by WP5 which provided a General probabilistic hazard and risk assessment framework for hazard associated with natural seismicity and to the seismicity induced and triggered during and after EGS operations. Within this workpackage, ground motion prediction equations have been derived specifically for geothermal sites am. At geothermal sites small magnitudes, high frequencies and short distances are much more relevant than for normal tectonic hazard assessment. Ground motion prediction equations forecast the level of shaking expected at the surface for a given seismic event. They are a key ingredient in any seismic hazard assessment. In addition, hazard models for specific test areas were developed. These models are based on a) statistical forecast approaches such as the Gutenberg-Richter model, and on b) physical considerations including geomechanical aspects, so called deterministic approaches.
The two types of proposed forecast models, statistical and physical, are complementary. At the present time, statistical models appear as a reasonable choice to forecast induced seismicity in a prospective way for decision support (e.g. traffic light systems). It has been shown that they fit the data well and that observed variations between the best models (Shapiro-type and ETAS modified - used in the logic tree of Fig. 1) have a low impact on the overall uncertainty (Mignan et al., in revision). Physical models can also well reproduce observed induced seismicity sequences (e.g. Gischig and Wiemer, 2013). While statistical models may outperform physical models due to a lower number of parameters and a faster computation, physical models are crucial for a better understanding of the evolution of induced seismicity over longer time horizons, of b-value changes, or of Mmax (to only cite a few). However at the present time, there is no consensus on which physical model best describes induced seismicity.
Based on the innovative hazard and risk assessment framework described above, it was possible to provide some recommendations to stakeholders (Deliverable 5.6 Recommendations for induced seismicity hazard and risk management).
On the basis of recommendations derived in WPs 3,4 and 5, guidelines for safe and reliable EGS operations were proposed in WP6. These guidelines are presented in project deliverables
D6.1 Description of the effect of different stimulation techniques on seismicity and strategies to mitigate induced seismicity
D6.2 Description of optimized monitoring strategy, both for the permanent receiver network and the network specifically installed during stimulation treatments
D6.3 Report summarizing the results of the tasks in form of input and boundary conditions for regulatory guidelines
D6.4 Report on development of public awareness and acceptance
In previous projects, a so-called traffic light system had been proposed to better control induced seismicity. This approach was based on a reaction to increases in reservoir pressure during injection. GEISER also provides guidelines for safe and reliable EGS operations. Key is a dynamic – forewarning – advanced traffic light system that can be used in all phases of a project. It requires, however, the determination of a maximum acceptable seismic magnitude and its accepted probability of occurrence. The reliability of the dynamic model is based on the availability of rock physics and seismic data, with models updated from real-time monitoring. This approach allows adjusting operational conditions to mitigate unsolicited effects and to improve system performance. It incorporates site specific parameters, derived from the geological studies, from drilling as well as logging. The efforts in induced seismic hazard and risk assessment have to mimic the potential damage.
The GEISER research efforts and guidelines are important steps to enable the efficient and safe use of deep geothermal energy resources throughout Europe.
This section must be of suitable quality to enable direct publication by the Commission and should preferably not exceed 40 pages. This report should address a wide audience, including the general public.
Project Results:
WP3 Analysis of Induced Seismicity in Geothermal Reservoirs
The objectives of the work package were to apply different seismological approaches to analyse data sets of induced micro-seismicity from geothermal areas. Induced seismic activity in space and time was examined with a view to establishing any systematic correlation with local geologic setting and injection parameters. These points were addressed in tasks defined to investigate three key aspects in detail
Spatiotemporal characteristics of fluid-injection induced microseismicity
Occurrence of large magnitude events (LME) in geothermal reservoirs
Role of pore pressure and crustal stress in induced seismicity
For this purpose, the datasets collected in WP 2 were used for detailed analysis.
The research activities in this WP were used directly for recommendations developed in WPs 5 and 6. Three deliverables resulted from the work performed.
D3.1 Evaluation of systematic relations between the seismic response to fluid injection with particular regard to the generation of LME, the depth of injection, lithology, injection pressure, crustal stress state, and local structural geology
D3.2 Source characterisation of LME and their occurrence in time and space
D3.3 Guidelines for techniques/methodologies for seismological investigations to be applied in future EGS operations, developed on the basis of successful analyses of past sequences.
In particular, D3.3 was designed to provide practical guidelines for future geothermal energy developments. Deliverables D3.1 and D3.2 were used to provide input for the guidelines provided in D3.3.
Summary of the results and recommendations
Enhanced Geothermal System (EGS) operations are usually related to engineering of hot, low permeability rock. Similar operations are common, and will be applied in the future to a large extent in conventional fracture dominated geothermal systems. The reason for this is that geothermal energy can only be extracted from a small volume of hot rock surrounding the fracture systems. Permeability enhancement in boreholes and creation of artificial fracture for production, therefore, are considered crucial for future development of the reservoirs. The same applies to the re-injection wells in fracture-dominated reservoirs. The fluid injected must go either into open fractures that are a part of a tectonic fault system, or into new fractures that must be created. In general, the basic knowledge on design and safe operation of fluid injection is based on a geologic map with fault/fracture distribution, tectonic movements, stress field, and natural seismicity with estimates of the moment magnitude of the largest seismic event to be expected in the area of interest.
Interface issues
In case no seismic catalogue is available for the geothermal site of interest, we advice to use synthetic catalogues based on physical model approaches (WP4). Different stimulation scenarios for one and the same reservoir and fracture geometry should be used in order to find safer injection strategies. In this context, we recommend to connect these models to probabilistic seismic hazard assessment (WP5). In particular we advise to use the method of Forward Induced Seismic Hazard assessment (FISHA) based on zero a priori seismicity information as opposed to the conventional probabilistic seismic hazard assessment (PSHA) applied to induced seismic events obtained real-time.
Conclusion
As the combination of all observables ((1) in-situ stress, (2) hypocenter locations, (3) seismic moments, (4) stress drop and (5) b-value) allows direct estimates of the spatial and temporal evolution of pore pressure in the stimulated rock volume, every effort should be made to perform such investigations during or immediately after stimulation. Thus, semi-automatic methodologies need to be developed to determine these parameters in near-real-time. The last three points, if available in real-time, constitute also essential input to advanced, dynamic traffic-light systems that are based on probabilistic seismic hazard assessments during ongoing stimulation.
Partner contributions
AMRA devoted their attention to two geothermal fields: The Geysers in California (USA) and Campi Flegrei caldera (Italy). They analyzed a seismic waveform dataset of more than 15000 events (1.0 <ML <4.5) recorded at The Geysers geothermal field by a dense surface seismic array during August 2007 to October 2011. The obtained results provided interesting issues concerning the investigated area characteristics, but above all, they highlighted very important general indications about the importance in geothermic contexts of seismological analyses for characterizing the medium mechanical properties.
The experience of The Geysers study showed that the strong variation in P-wave velocity and Vp/Vs (Fig. 1) are closely related to a variation of the attenuation. In turn, all these parameters have been found consistently with the subdivision of The Geyser field in two sub-regions characterized by significantly different medium and fracture source properties.
A further important issue related to the availability of high resolution 3D velocity models is that they would be very important for improving the earthquakes location, and thus for obtaining a better reconstruction of the fractures orientation and dimension.
Fig. 1 Map view showing the Vp/Vs ratio inferred from the 3-D tomography, at 0, 1, 1.5 and 2 km depth, at The Geysers.
NORSAR has focused on the analysis of data from two sites, (1) Paralana, in South Australia, a new enhanced geothermal system (EGS) where the first main stimulation occurred in July 2011 and (2) Basel, in Switzerland, an EGS which was stopped during the stimulation phase in 2003 due to the occurrence of a large magnitude event (LME).
As EGS operations normally produce a large number of microearthquakes, automatic processing is usually required. However, there is an uncertainty of automatic picks, if for example the data are noisy or if several events occur at the same time. A careful check of the picks, based for example on location error and residual values, is always a first step for improving the data base. Another step to refine phase picks is to perform waveform cross-correlations for pairs of earthquakes recorded at the same station (Rowe et al., 2002). The approach at Paralana was first to process the data automatically with MIMO (Oye and Roth, 2003), to assess the quality of the automatic picks using a reference subset of events picked manually, and to cross-check the database manually (for events with magnitude larger than one and for the large residuals). Then, we improved the travel-time differences with waveform cross-correlation (Fig. 2) and relocated the events using the double-difference algorithm (Waldhauser and Ellsworth, 2000)
Figure 2: Seismograms of two events (1 and 2, blue and red, respectively) recorded at a surface station. Waveforms are aligned according to the automatic P-picks (blue star): wrong P-picks for event 2 is then highlighted by the gap between the waveforms (blue line and red dashed line). After waveform cross-correlation, the waveform of the event 2 is correctly aligned (red line) with event 1 and the travel-time difference can be corrected.
ÍSOR
Seismicity data from three Icelandic geothermal fields, namely Krafla, Hengill and the Reykjanes peninsula, have been analysed. Data from Krafla that were obtained during the drilling of the IDDP well in 2009 have been analysed in some detail. The well was cased with a steel casing down to 1950 m and with a slotted liner around an aquifer at 1950-2080 depth at a top of a molten or partially molten magmatic intrusion. Low pressure stimulation with cold water during completion of the drilling and subsequent tests induced seismicity of magnitudes up to 1 in local magnitude. The epicentres initially followed the top of the magmatic layer horizontally away from the wellbore until it met an inclined fracture, most likely a pre-existing one. The epicentres then followed this fracture upwards showing the connection between the heat-mining zone at the top of the magmatic layer and the active fault systems.
The data from Hengill contain seismic and pressure recordings obtained during drilling of an injection well in February 2011 for the 303 MWe Hellisheidi power plant in the Western part of the Hengill geothermal area. The injection well was drilled into a complex system of normal NE trending faults belonging to the axial rift zone of the Mid Atlantic Ridge in Iceland and N-S trending right lateral strike slip faults with character of the South Iceland transform zone. During the drilling, the well entered an open fracture at 1320m and total loss of circulation (40 L/s) was observed. A swarm of earthquakes was immediately initiated with magnitudes up to ML 2.2. The earthquakes were clearly felt in the neighbourhood. These data were analysed and used to investigate the interaction between injection and pre-existing fractures. In September 2011 a full scale injection of 550 L/s started into the fissure swarm resulting in high level of induced seismicity. About 3000 earthquakes have been located in the area in 2011. The earthquakes came in intensive swarms with quiet intervals in between. The seismicity culminated on October 15th when several events of magnitude more than 3.0 were measured, the largest one exceeding 3.8. This earthquake swarm occurred in conjunction with major disturbance of the injection rate. This is among the largest quakes that have been triggered by geothermal re-injection in the world. It caused serious inconvenience in a nearby village as people were not prepared.
Figure 3: Location of induced events around the IDDP well at Krafla during low pressure stimulation with cold water.
Figure 4: Induced seismicity during circulation loss of cold water while drilling an injection well.
The data from these latest events at Hengill were not supposed to be included and analysed within the GEISER project as they were collected late in the year 2011, the second year of GEISER. However, as this dataset is of high importance in understanding the earthquake triggering effect of large scale re-injection, some of the obtained were used for the GEISER work. The power company that owns the data agreed to release them to the scientific community.
Data from Reykjanes peninsula include data from two production fields, Svartsengi and Reykjanes. In Svartsengi, production has been ongoing since 1977 and re-injection started in 1984. Reykjanes entered production of 100 MWe in 2006 but without re-injection until 2009. In addition to the national seismic network in Iceland the University of Iceland and the University of Wisconsin operated a local seismic network in Reykjanes from December 2008 until May 2009. Around 320 earthquakes have been located during that period. The seismic dataset consists of continuous waveforms from 11 seismic stations at the tip of the peninsula. The seismic activity was analysed, with relocation of earthquakes using the double-difference algorithm, evaluation of focal mechanisms of the earthquakes, and investigation of the relationship with injection and production data and tectonic structures.
ETH Zürich focused their analyses on the vast data set acquired during the stimulation of the Basel EGS in 2006, which induced about 3500 seismic events. These occurred during the injection phase and the following months and were locatable by the six-station borehole network (Häring et al., 2008). Close to 200 of the strongest events were also recorded by various surface networks (Deichmann and Ernst, 2009; Deichmann and Giardini, 2009). Preliminary hypocenter locations and magnitudes were obtained from manually picked arrival-times and amplitudes. This information was sufficient to map the overall orientation and dimension of the stimulated rock volume in near real-time. More detailed insights into the physical processes were obtained only through systematic re-evaluation of the data set by various groups of scientists over the following years: hypocenter locations were refined in several stages by the application of various methods; magnitudes were re-evaluated and seismic moments were calculated by different groups; focal mechanisms and moment tensors were determined; b-values and stress-drops were mapped in space and time; pore pressure evolution was estimated.
Hypocenter locations are of fundamental importance for all subsequent analyses, providing the basis for seismic velocity models. Poorly calibrated velocity models and arrival-time errors lead to a more or less amorphous appearance of the so-called seismic cloud, which in addition can be systematically mislocated and misshaped (Kraft and Deichmann, 2013). In fact, a large component of the observed scatter of the initial hypocenter locations was due to faulty or inconsistent arrival time picks made under time pressure by inexperienced analysts.
A well-calibrated velocity model, while ensuring the proper location of the seismic cloud as a whole relative to the injection well, is not sufficient to resolve individual faults activated by the stimulation. The detailed analysis of the sequences associated with the larger magnitude events (Mw > 2) induced during the stimulation of the Basel EGS showed that the activated faults have dimensions on the order of several 100 m and are often oriented obliquely to the overall orientation of the microseismic cloud (Deichmann et al., 2013). These results reveal a complex internal structure of the flow paths in the rock volume stimulated by the water injection and imply that geo-mechanical models consisting of a single throughgoing structure are too simplistic.
Although the dimensions of the faults activated by the stimulation of the Basel EGS were estimated on the assumption of a constant stress-drop (Deichmann et al., 2013), the analysis by Goertz-Allmann et al. (2011) shows that stress-drop varies over a wide range and tends to increase systematically with distance from the injection point, which suggests that stress drop correlates with pore pressure perturbations due to the injection. This hypothesis was tested by calculating the injection-related pore pressure perturbation based on a simple linear pore pressure diffusion model and find a good correlation of the expected pore pressure perturbation with the estimated stress drops. These results were reported in more detail in periodic report 2 (2012).
An alternative approach to obtain direct observational estimates of pore pressure within the stimulated reservoir was taken by Terakawa et al. (2012). Using all available observations (orientation of stress field, stress field ratio, friction values determined in the laboratory) and some modelling assumptions (homogeneity of the stress field over the stimulated volume) it was possible to fully characterize the average stress field and to calculate the shear stress on each fault and the fluid pressure necessary to induce failure. Although overall the fluid pressures decrease towards the periphery of the stimulated volume, the pressure field within this volume is characterized by several patches of higher and lower pressures. More important is, however, that according to the results of this analysis the fluid pressures needed to trigger many of the observed earthquakes are in the range of 10 to 20 MPa even out to several 100 m from the injection well. This result is obtained totally independently of the applied injection pressures but is entirely consistent with these pressure values.
To map possible flow paths that have been opened during stimulation, and thus to constrain geo-mechanical models of permeability enhancement, estimates of the dimensions of the activated faults are also necessary. This requires reliable measures of seismic moment or moment magnitude (not just local magnitude) and, if possible, stress drops (Bethmann et al., 2011; Goertz-Allmann et al., 2011). The detailed analysis of the sequences associated with the larger magnitude events (Mw > 2) induced during the stimulation of the Basel EGS showed that the activated faults have dimensions on the order of several 100 m and are often oriented obliquely to the overall orientation of the microseismic cloud (Deichmann et al., 2013). These results reveal a complex internal structure of the flow paths in the rock volume stimulated by the water injection and imply that geo-mechanical models consisting of a single throughgoing structure are too simplistic.
Bachmann et al. (2012) apply high-resolution b-value mapping to induced seismicity, in order to obtain information on the stress regime and possibly the pore-pressure evolution in space and time inside the stimulated rock volume.
Figure 5 (a) Overview of the Basel induced seismicity showing the depth of the 3560 located events (circles) and the location of the seismic stations (triangles). The red plane marks the top of the crystalline basement, within which all events occurred. The color scale indicates the recording completeness ranging from Mw 0.7 to 1 (Mc). (b) Close-up of the events with the overall b-value distribution based on all events. While values range from Mw 0.8 to 3.5 the color bar is limited from 1 to 2 for a clearer visibility. (From Bachmann et al., 2012).
The key findings of Bachmann et al. (2012) are summarized in Figure 5, which shows the strong and highly systematic spatial heterogeneity of the b-values for the seismicity induced by the Basel EGS. Unusually high b-values shown in red are found near the injection point and earlier in the sequence; further out, b-values tend to be closer to the normal tectonic average of around 1.0. Bachman et al. (2012) and in even more detail Goertz-Allmann and Wiemer (2012) have developed a geo-mechanical model that explains these observations and that offers a framework for improved forecasting of induced seismicity. The fact that the largest events in induced seismicity often – but not always – occur after shut-in is well-explained by this model.
EOST
The EOST crew took advantage of the large amount of data gathered by the Soultz EGS scientific project to develop a seismological methodology and obtained a reliable knowledge of the deep structures developing reliable methods that can be applied at various scales and different moments of the development of geothermal projects. The focus was on studying the induced seismicity, seismic noise correlation for exploration and monitoring of geothermal reservoirs, and on seismic tomography using VSP data.
Specific focus of the investigation was on the temporal variations of the elastic parameters through 4D seismic tomography, which is very useful for the understanding of the mechanical behaviour of the geothermal reservoirs, when used with rigorous assessment tests and post-processing procedures (e.g. the Weighted Average Model method; Calò et al., 2009, 2011, 2013a) to avoid misunderstanding due to the presence of bias in the models.
Results of these studies show that injection tests performed in regions initially poorly connected to large faults are characterized by a low anomaly of the P-wave velocity mainly located around the zone where microseismic activity develops. In some specific periods (i.e. when the injected flow rate was suddenly increased) the velocity anomaly disappears suggesting that the velocity variations within the reservoir (and consequently the related variations of effective stress) are not associated with simple water diffusion from the injection well, but rather reflect the occurrence of large-scale aseismic events in the reservoir. In regions where pre-existing faults are well documented, the accumulation of effective stresses close to the well is avoided probably because the structures represent the main paths of the injected water. This results in a lack of large low Vp anomalies during the stimulation and in the occurrence of the induced seismicity located along the major structures.
The noise cross correlation tomography method was tested to image 3D structures of the Soultz-sous-Forêts EGS at small scale (about 1x1 km2). This method proved favourable for application in densely populated regions characterized by high geothermal potential, as is the case for the upper Rhine valley (Calò et al., 2013b). To image the geometrical features of the Soultz-sous-Forêts EGS Rayleigh waves were reconstructed from cross-correlations of 15 months of ambient seismic noise recorded by 23 seismological stations installed around the geothermal power plant. The reconstructed waveforms were used to measure group velocity dispersion curves at periods between 1.0 and 5.0 s. The obtained measurements were inverted for two-dimensional group velocity maps and finally for a 3-D S-wave velocity model of the Soultz region from 0 to 5.2 km depth.
The procedure applied at Soultz can be used as a guideline for the reliable 3D imaging of
geothermal reservoirs (Calò et al., 2013b). Furthermore cross-correlation of ambient noise could be applied to develop geothermal monitoring systems.
Analysis of the VSP data collected in 2008 to better understand permeable fractures serving as main circulation paths of the reservoir gave promising results. The main new insight is the evidence of a previously not identified structure clearly underlined by reflections and a well-marked high velocity anomaly. This structure is located between the wells GPK3 and GPK4 and could explain the low connection between these two wells already observed by hydrologists.
GFZ
The research group at GFZ German Research Center for Geosciences was focused on analysis of seismic data from Berlín Geothermal Field, El Salvador and The Geysers geothermal field in California, USA. The studies performed at these sites were related to the high resolution reservoir characterization using induced seismicity data and state of the art waveform processing techniques.
Three state-of-the-art algorithms, namely the Double-Difference (hypoDD) re-location technique, the Spectral Ratio (SR) technique and the Stress Inversion (SI) method were used to analyse IS generated by fluid injection and steam production. hypoDD significantly improves the precision of hypocenters allowing imaging of the fluid path and propagation in response to multiple injections with unprecedented detail (Kwiatek et al., 2013). In addition, the application of the SR technique provides refined source parameters that can later be used to interpret the subtle interactions between pressure perturbations, fluid flow and fracture (re-) activation within the reservoir. Finally, we show how the SI technique can contribute towards monitoring geomechanical processes occurring in the reservoir in response to short-term injection and long-term production activities (Martínez-Garzón et al., 2013).
Figure 6. Berlìn Geothermal Field (BGF): Spatial distribution of 581 seismic events recorded between Oct-2002 and Jan-2004 (locations provided by BGF operator). Thin grey lines display shifts in earthquake locations towards the centre of the BGF due to the application of DD relocation technique. (Kwiatek et al., 2013)
The techniques described above to characterize IS and geomechanical processes at the two geothermal reservoirs represent selected case studies. These methods, however, can be applied to any data set of IS with a reasonably good quality of the recorded waveforms obtained from a sufficient monitoring network.
WP4
The aim of the development of a geothermal site by hydraulic and/or chemical stimulation is to irreversibly increase the injectivity/productivity of a well by a local increase of the transmissivity of the fractures and/or faults intersecting the well and an improvement of the connectivity of the well to fracture and/or fault network. The generation of microseismicity is inherent in these methods of development of wells. However, if a large number of microearthquakes is inevitable, major events (magnitude between 2 and 3) must be studied carefully. Understanding these events, may lead to propose scenarios for the stimulation reducing or eliminating them while maintaining the effectiveness of the stimulation relative to the injectivity. The WP4 work package is dedicated to understanding the geomechanical causes and processes of induced seismicity in various contexts and at various scales. This work was carried out in three sub-tasks : sub-task 4.1 related to the role of pore pressure changes; sub-task 4.2 related to the role of temperature changes; and sub-task 4.3 related to the role of faults and fractures. The following provides an overview of the S&T results obtained within WP4.
Laboratory experiments (GFZ) coupled with acoustic emission (AE) monitoring during fluid injection have been conducted in dry and partially saturated sandstone specimens, with a pre-existing laboratory induced fracture plane. The experiments show that the re-activation of a fracture through fluid pressure perturbations into saturated specimen is characterized by fewer high amplitude events than into dry specimens. Furthermore, the reactivation occurred via collapse and shear cracks in places of the pre-existing fracture plane and shear and tensile cracks in regions far from it. Other laboratory experiments have also been conducted with specimens which had a predefined saw-cut plane oriented at different angles to the direction of the major imposed principal stress (Figure 1). Stick-slip test have been conducted, and the results show that slip along the saw-cut plane was accompanied by an increased number of AE events, all located in different patches along the saw-cut plane. New fractures, far from the saw-cut planes were not induced. Mohr-Coulomb envelopes reached sliding lines with different coefficients of static friction, which indicates a change in the roughness of the saw-cut planes (higher coefficients of static frictions, thus, more rough surfaces due to wearing). Shear cracks were dominant during the sliding, while the number of tensile cracks was not negligible. The latter were related to local pore pressure perturbations and local dilatancy taking place during the sliding
Figure 1: View of rock specimen with saw cut plane – Location of AE during pore pressure increase
Partner INGV was mainly involved in sub-task 4.1. Their contribution dealt with modeling of time-dependent Coulomb stress changes due to fluid injection and fluid withdrawal in wells by using a thermo-fluid dynamical approach in porous media with a coupling of the two codes THOUGH2 and COMSOL. The approach consists in a two-step procedure: in the first step, injection of withdrawal of water is simulated by TOUGH2, by computing pressure and temperature changes in the volume; in a second step, the incremental stress tensor is computed by COMSOL. According to the models, stress changes can attain considerable values, able to significantly re-orient the mechanisms of most stressed fractures, with respect to the orientation due to the regional stress field alone. Also, the results highlighted that the main causes of induced seismicity during stimulation are due to the Coulomb stress changes generated by water injection, rather than by the effect of the temperature of the injected fluid.
Changes of Coulomb stress due to induced seismicity during stimulation have also been analyzed by partner GEOWATT. For this purpose an efficient method has been developed to calculate coseismic stress changes from an elliptical slip distribution on a circular fracture using superposition of rectangular sources. The method has been applied on a dataset of 715 focal mechanisms derived from seismic recordings of the Soutz-sous-Forêts GPK2 2000 stimulation to calculate temporal evolution of change of Coulomb stress (Figure 2). It comes out that the distribution of stress changes at the hypocenters reveals no evidence for events to be triggered from previous induced seismicity. However, about 5% of the events occur in areas where the stress is increased by more than 1 MPa and are thus candidates for triggering by previous events.
Figure 2: 3D views of coseismic stress changes in the Soultz-sous-Forêts reservoir. The wells GPK2 (red), GPK3 (green) and GPK4 (blue) are displayed with bold open hole section. The microseismic events used for the computation are represented by sphere symbols which scale with the magnitude, not with the geometry
Partner TNO was involved in the three sub-tasks of WP4. TNO has shown within sub-task 4.1 that for creating an EGS in compartmentalized sedimentary reservoirs, both the effects of pressure changes in the rock matrix and the impact of large scale heterogeneities should be taken into account. It has been shown within sub-task 4.2 that for simple systems dominated be few large fractures, poro-elasticity processes were most likely to control the overall thermo-mechanical process, by increasing the shear strength resistance of the fracture and balancing local traction and fracture opening effects, thus preventing large shear failure to develop. Within sub-task 4.3 partner TNO developed a specific coupled continuous approach with FLAC3D in order to study the role of pore pressure changes causing fault zone reactivation and inducing seismicity during the EGS operations. A sensitivity analysis has been performed on key parameters such as the in-situ stress regime, the fracture or fault strength and fault frictional behavior after the onset of failure.Results of the FLAC3D model were compared to results of the Block-Spring model as proposed by Baisch. The models were also applied on the GPK3 stimulation case that was performed in Soultz-sous-Forêts in 2003 (Figure 3). The basic features of the observed seismicity were reproduced. The 3D physical approach and the Block-Spring model showed their specific advantages: the Block-Spring model, if calibrated, establishes a fast modeling tool for sensitivity analyses; the FLAC3D implementation allows better understanding as it is based on actual physics.
Figure 3: Stimulation of Soultz-sous-Forêts GPK3 well with 3D continuous approach – Distribution of pore pressure (a), fracture aperture (b) and permeability (c)
Partner ARMINES was also involved in the three sub-tasks of WP4. ARMINES studied with a Discrete Fractured Network approach the impact of the thermal effect during a long term flow circulation test based on a forward thermo-hydro-mechanical modeling of the in-situ observations made at Rosemanoves site (UK). It has been shown that when injection pressure is maintained at constant level over time, shear can be re-activated at places where a previous rupture occurred, as soon as thermal tractions accumulated with time reach a magnitude similar to the stress drops generated during reservoir stimulation. The delay of a new event to occur is controlled by heat diffusion phenomenon in the rock blocks adjacent to the fracture. Note that calculated seismic moments due to the long term cooling of the rock are one order of magnitude lower, but that a small increase in injection pressure can precipitate new seismic activity as soon as some thermal stress has been stored. It comes out that seismic activity during the life cycle of a reservoir is not independent of pre-stimulation phases during initial reservoir development. This is of prior interest for mitigation strategies. The role of fractures heterogeneities has also been undertaken by partner ARMINES by studying different reservoir configurations: a 3D multi fracture system and a single fault segment with heterogeneous properties under a given stress regime (Figure 4). According to the models, the development of shear is controlled by pressure diffusion, and therefore in both cases hydraulic properties, contrasts in properties of the various structures and boundary conditions are governing the mechanical process. The critical conditions with regard to the occurrence of large induced seismic events are obtained in the situation of some large fracture with unknown hydraulic properties exists at an intermediate distance of the well, and when the ‘bulk’ hydraulic diffusivity of the massif is limited. In the case this structure is also of poor natural hydraulic diffusivity, or locked by other fault segments, the risk of inducing there a significant event is higher because it can be pressurized over a large region. Therefore tracking by geophysical methods any of these structural surfaces of potential failure at distances up to some hundreds of meters from a well is crucial. Testing them separately, if cross cut by a well, even with short term hydraulic tests would produce most valuable data, that would be helpful to design the most reasonable stimulation scenario, with the constraint of a pre-identified admissible pressure in the fault. This recommendation should form part of the mitigation strategy. Drilling capabilities allows now the possibility to develop deviated wells, or wells with multiple branches in smaller diameters. Targeting such geological structures should be evaluated in case VSP had identified them for instance. Although not affordable at first glance, these investments would anyhow participate to the long-term survey of any EGS production field at industrial scale and have the potential of allowing re-entry in the future for further injection or production developments. It is therefore suggested that the deepest section of a well in a new site should be first used at survey and not be automatically designed to be included in the final system.
Figure 4: Left: Prescribed step-rate injection test and calculated hydraulic head at injection well. Right, corresponding shearing events occurring in time with their calculated moments in blue, with hydraulic head (green area) in the fault when shear starts to develop into it, after about 3 days of injection. No events are simulated during the late injection phase
Partner BRGM was involved in sub-task 4.3 related to the role of faults and fractures. BRGM focused its study on the coupled hydro-mechanical response of a fault network intersecting a deep geothermal well to a hydraulic stimulation. The stimulation zone around Soultz-sous-Forêts deep wells was modelled using a 3D Distinct Element Method. According to the stimulation phase and post-stimulation phase carried out here, the models reproduce the evolution of the Overpressure-Flowrate curves observed in-situ, even if the total increase of injectivity has not been reached yet. Contrary to what is traditionally done, looking at the orientation of the most favourably oriented fault zone in respect to the principal stress does not allow determining the most efficient fault network. On the basis on the study performed, the three kinds of behaviour can be met (Figure 5):
• Direct stimulation due to shearing in fault segment favourably oriented with respect to σH. But, all the favourably oriented with respect to σH are not stimulated;
• Little/no stimulation, even if favourably oriented with respect to σH;
• “Indirect stimulation” of fault segment, not favourably oriented with respect to σH, due to the connectivity in 3D of the fault network.
The adaption of a joint model, that takes into account progressive damage during shear, has allowed discriminating the coseismic part of the shear from the aseismic one and thus better estimating the event magnitude. It results that:
• In the directly stimulated fault segments, the maximum seismic moments reach a plateau (M < 2) due to the repartition of the shear in several distinct seismic areas, clearly lower than the maximum seismic moments calculated with the standard Mohr-Coulomb model used so far (M around 3);
• In the indirectly stimulated fault segments, there is barely any shear displacement at the origin of major seismic events.
Finally, with the developments undertaken by BRGM, tools are available to study the occurrence of potential major induced seismic events on the same mechanical basis as the evaluation of the efficiency of the hydraulic stimulation of a deep well. Moreover, from the point of view of the knowledge of EGS well, the results obtained show the importance of taking into account the 3D reality of the fault network around the well, for the efficiency of the stimulation as well as for the estimation of the magnitude of the major seismic events. In order to lead stimulation scenarios, limiting the maximum magnitude of seismic events, an improvement is still needed in choosing the parameters of the hydro-mechanical behaviour to reach the final value of the well injectivity.
Figure 5: Stimulation of Soultz-sous-Forêts GPK4 well 3D Distinct Element approach - Location of the nodes reaching the rupture, superimposed to the shear displacement for a fault segment directly stimulated by shearing (left) and another fault segment indirectly stimulated (right)
The implementation of the effect of fluid injection into a pre-existing Boundary Integral Element Method code modeling dynamic rupture has also been done by partner BRGM in order to build, based on physical parameters, catalogues of synthetic events (location, time, magnitude) that are necessary to evaluate the seismic hazard (WP5). With this approach, it is assumed that the evolution of fluid migration and induced seismicity takes place along a single planar fault, described as a finite permeable zone of variable width. A sensitivity analysis has been performed in order to study the parameter interdependencies. It turns out that seismicity and fluid migration are strongly influenced by the injection rate and the fault rheology heterogeneity. The simulated seismicity generally tends to rapidly evolve after triggering, independently of the injection history. This self-induced seismicity takes place in the case where shear rupturing on the planar fault becomes dominant over the fluid migration process. On the contrary, if healing processes take place in the fault, so that the fluid mass becomes trapped within the fault, rupturing occurs continuously during the injection period.
To conclude, laboratory experiments and several modeling activities with complementary approaches have been addressed throughout WP4. Different numerical tools have been developed, or adapted, in order to study the coupled mechanisms that induce seismicity during EGS operations. The efforts put in WP4 also include the development of tools (partners ARMINES, BRGM, TNO) to simulate catalogs of synthetic events, based on rock mass physical features and on injection rates, that are useful in order to assess the seismic hazard associated to EGS operations. A considerable progress has been therefore made in GEISER.
WP5: Seismic Hazard Assessment
The objectives of WP5 of the GEISER project were to (i) develop a comprehensive framework to assess the earthquake hazard associated with natural seismicity and to the seismicity induced and triggered during and after EGS operations, (ii) develop statistical methodologies to assess the stability or increase in seismic hazard associated with EGS operations and (iii) test modelling tools of incremental accuracy to assess shaking and where possible damage which could be produced by EGS induced and triggered seismicity.
This work led to the publication of 15 peer reviewed articles by the different groups involved in WP5 (Douglas and Jousset; 2011; Goertz-Allmann et al., 2011; Bachmann et al., 2012; Convertito et al., 2012; Douglas et al., 2013; Gischig and Wiemer, 2013; Aochi et al., 2012; Goertz-Allmann and Wiemer, 2013; Mena et al., 2013; Hakimhashemi et al., 2013; Douglas, 2013; Catalli et al., 2013; Edwards and Douglas, 2013a, b; Mignan et al., in revision; Schoenball et al., 2012), plus six GEISER deliverables. Several additional publications are work in progress.
The main results of WP5 are threefold:
1) General probabilistic hazard and risk assessment framework (D5.1 2, 4, 5)
Figure 1 summarizes the main characteristics of the proposed approach, in the form of a logic tree. It was first shown that the three types of seismicity (natural, induced and triggered) could be treated in a similar manner at the hazard and risk level, hence the proposal of a common framework. This framework (Mignan et al., in revision) combines state-of-the-art time-dependent earthquake forecasting methods (probabilistic part) with standard seismic hazard assessment (so-called hazard curves) and standard risk assessment (here the RISK-UE macroseismic method). This is the first time that such an approach has been applied to EGSs, based on lessons learned from a long tradition of tectonic event analysis.
The three types of seismicity behaviour have some unique specificities, which do not impact the structure of the framework but yield different input parameters to be taken into account. In particular, induced seismicity differs from natural seismicity in the fact that the rate of occurrence is correlated to the volume of injected fluids (e.g. Shapiro-type or modified Epidemic-Type Aftershock Sequence ETAS models - light blue in Fig. 1). Distinction between induced and triggered seismicity can be made based on the maximum magnitude Mmax parameter (dark blue in Fig. 1), the size of an induced event being controlled by the size of the reservoir while the size of a triggered event is controlled by the maturity of the existing fault network. Another characteristic of induced seismicity, compared to tectonic events, is the range of magnitudes considered. Induced seismicity can cause nuisance for magnitudes as low as m = 2 while standard hazard and risk modelling is defined for m = 5+ (damaging events). It means that extrapolation of hazard and risk results from m = 5+ to 2 ≤ m ≤ 4 inevitably leads to biases. This has been corrected at the hazard level with the definition of ground motion prediction equations (GMPEs) for induced seismicity data sets (Douglas et al., 2013; Edwards and Douglas; 2013; see inset in Fig. 1) and with the calibration of vulnerability curves to avoid overestimating losses (red in Fig. 1) due to induced events (based on the SERIANEX study).
The logic tree approach captures all of these specificities, but more generally all epistemic uncertainties to be considered for hazard and risk assessment in the EGS context. It also provides a simple way to incorporate results of statistical methodologies to assess changes in hazard before, during and after injections of fluids. In particular the likelihood method was used to compare different time-dependent forecasts, which gives the weights to be included in the logic tree (Mena et al., 2013). Finally, the logic tree can be updated following the testing of modelling tools of incremental accuracy to assess shaking (and damage). In particular, the 36 GMPEs proposed for induced seismicity represent a large range of possibilities, which can be refined if more data becomes available (Douglas et al., 2013). In Figure 1 (in yellow/orange), weights are based on improved knowledge on GMPE parameters for the case of Switzerland. If GMPEs are not available, one could instead use Intensity Prediction Equations (IPEs), based on global data or on regional data (e.g. Swiss IPEs in Fig. 1).
Figure 1: Logic tree for induced seismicity probabilistic hazard and risk assessment. Ground Motion Prediction Equations (GMPEs) are the ones proposed by Douglas et al. (2013). Weights are the same for most branches except for induced-seismicity based GMPEs for which weights are fixed for Switzerland. IPE refers to Intensity Prediction Equation and GMICE to Ground Motion-Intensity Conversion Equations. Source: Mignan et al. (in revision).
2) Hazard models for specific test areas (D5.3)
Within the scope of WP5, a series of hazard models have been described, based on different forecast models and hazard metrics and tested on various EGS induced seismicity sequences. Sequences include: Basel (Switzerland), Rosemanowes (United Kingdom), Soultz-sous-Forêts (France) and The Geisers (USA). Statistical forecast include: Gutenberg-Richter law, ETAS model modified with a flow-induced background rate and the Shapiro-type model. Physical models include: Geomechanical models, non-linear flow model and static stress transfer model. Hazard metrics considered are: event magnitude, European Macroseismic Scale (EMS) intensity and peak ground acceleration (PGA). Here is a synthesis of these results, which is also used as input for the development of guidelines for best practice (see next section).
The two types of proposed forecast models, statistical and physical, are complementary. At the present time, statistical models appear as a reasonable choice to forecast induced seismicity in a prospective way for decision support (e.g. traffic light systems). It has been shown that they fit the data well and that observed variations between the best models (Shapiro-type and ETAS modified - used in the logic tree of Fig. 1) have a low impact on the overall uncertainty (Mignan et al., in revision). Physical models can also well reproduce observed induced seismicity sequences (e.g. Gischig and Wiemer, 2013). While statistical models may outperform physical models due to a lower number of parameters and a faster computation, physical models are crucial for a better understanding of the evolution of induced seismicity over longer time horizons, of b-value changes, or of Mmax (to only cite a few). However at the present time, there is no consensus on which physical model best describes induced seismicity.
Groups affected by an EGS project are public authorities, local residents, the operator company and insurances. Depending on the group, a different metric may be required. The simplest metric is to use the probability of exceeding a given magnitude m (Bachmann et al., 2012; Goertz-Allmann and Wiemer, 2013; Mignan et al., in revision). Adding more information, such as Mmax and ground motion properties, while adding more parameters and thus more uncertainties, provides a better communication tool. The probability of exceeding a given intensity measure (e.g. EMS based on an IPE) will inform local residents and public authorities as to what events are likely to be felt or damaging (Gischig and Wiemer, 2013; Mena et al., 2013; Mignan et al., in revision). The probability of exceeding a given PGA value (Convertito et al., 2012), being based on an instrumental measure, can be useful to the EGS operator to follow specific thresholds.
3) Recommendations for induced seismicity hazard and risk management (D5.6)
Based on the innovative hazard and risk assessment framework described above, it was possible to provide some recommendations to stakeholders. This is described in the section on Task 5.5 below. Figure 2 shows a schematic example of a advanced traffic-light system.
Figure 2: Advanced Traffic Light System: Schematic overview of the foreseen software framework. Legend: W = weighting; GMPE = ground motion prediction equation; EGF = empirical green’s function; PSHA = probabilistic seismic hazard assessment.
For such a comprehensive hazard and risk assessment framework we provide the following guidelines:
(1) Use a deterministic risk approach prior to injection to estimate roughly the expected losses (some hazard metrics could also be used) in function of induced event expected magnitude;
(2) Use a probabilistic risk approach during injection by combining deterministic results to time-dependent induced seismicity forecasts;
(3) Distinct output metrics to communicate the risk to different interest groups (public authorities/regulators, operators, public, insurances);
(4) Include epistemic uncertainty using a logic-tree approach as well as realistic aleatory uncertainty (e.g. the seismic hazard "sigma") for an objective estimation of hazard and risk associated to EGS activities;
(5) Define a clear traffic-light system following well-established methodologies and regulatory standards.
4) Achievements by Tasks
Below more detailed outcomes of WP5 are given, broken down by each task:
Task 5.1: Assessment of seismic hazard associated to natural seismicity
A good definition of the background hazard and the probability of natural earthquakes is needed to assess the probability that an earthquake that occurred in the vicinity of an EGS site is causally related to the activity. This judgment may have substantial legal and financial consequences (in terms of insurance and liability); it also requires local and regional monitoring capable of resolving hypocenters with sufficient accuracy. Hazard assessment is a core responsibility of national authorities, and all nations in Europe maintain and update a probabilistic seismic hazard model. While the degree of complexity and sophistication varies, the basic methodology of probabilistic seismic hazard assessment (PSHA) is well established and accepted. We evaluated the available models for their suitability for EGS relevant estimation of the background hazard at a European level. We concluded that the SHARE seismic hazard model, to be published in 2013 as part of the EC FP7 project on Seismic Hazard Harmonization (www.share-eu.org/) is the ideal reference framework.
Task 5.2: Assessment of seismic hazard associated to EGS induced seismicity
We developed a probabilistic hazard assessment method for induced seismicity observed in EGSs. We first generated forecasts based on induced seismicity models available in the literature. Forecasts were then included in time-dependent seismic hazard assessment and combined to ground motion intensity prediction equations (IPEs) to compute induced seismicity hazard curves. Epistemic uncertainties due to the choice of the forecast model and IPE were captured by using a logic tree approach. The weight of the different forecast models was determined using the likelihood method on observed induced microseismicity (m < 4).
Task 5.3: Assessment of seismic hazard associated to EGS triggered seismicity
Triggered seismicity, in contrast to induced seismicity, corresponds to events that occur on existing faults, i.e. not produced during stimulation, but which are advanced in their seismic cycle due to increased stress during stimulation. The main criterion for the assessment of EGS triggered seismicity hazard is the maximum magnitude Mmax. We first presented a statistical approach, which investigates the role of Mmax in seismic hazard assessment. In that view, Mmax in the EGS regime can be the same Mmax considered in the tectonic regime (e.g. Mmax = 7.0) thus making no distinction between induced and triggered seismicity. The probability of occurrence of such event is very low but is a non-zero value (conservative probabilistic view). We then presented a physical approach, in which Mmax is estimated in a deterministic way. Distinction between induced and triggered seismicity as well as the debate on Mmax values remain open questions, as no consensus seems to exist. We concluded that the statistical approach should be preferred at the present state of knowledge, using the tectonic Mmax to remain conservative. However mapping of existing faults around EGS sites should help refining Mmax values, which has a significant role in hazard assessment compared to other input parameters for high hazard intensities.
Task 5.4: Shaking and damage scenarios from EGS induced and triggered events
Various ground-motion datasets of induced and natural seismicity (from Basel, Geysers, Hengill, Roswinkel, Soultz, and Voerendaal) were compiled and processed, and moment magnitudes for all events were recomputed homogeneously. These data were used to show that ground motions from induced and natural earthquakes cannot be statistically distinguished. Empirical GMPEs were derived from these data; and, although they have similar characteristics to recent GMPEs for natural and mining-related seismicity, the standard deviations are higher. To account for epistemic uncertainties, stochastic models subsequently were developed based on a single corner frequency and with parameters constrained by the available data. Predicted ground motions from these models were fitted with functional forms to obtain easy-to-use GMPEs. These are associated with standard deviations derived from the empirical data to characterize aleatory variability. As an example, we demonstrated the potential use of these models using data from Campi Flegrei.
Task 5.5: Guidelines for best practice in seismic hazard assessment for site selection and licensing
If geothermal energy from EGSs is to become a significant component in future energy policy, there is an urgent need to provide guidelines to help regulators in devising seismic hazard assessment specifications for the selection, licensing and long-term operation of EGS sites in different geological settings. Here we built upon previous results of WP5 on the assessment of natural, induced and triggered seismic hazard to develop a comprehensive framework for best practice. We provided the following guidelines:
(1) Use a deterministic risk approach prior to injection to estimate roughly the expected losses (some hazard metrics could also be used) in function of induced event expected magnitude;
(2) Use a probabilistic risk approach during injection by combining deterministic results to time-dependent induced seismicity forecasts;
(3) Include epistemic uncertainty using a logic-tree approach as well as realistic aleatory uncertainty (e.g. the seismic hazard "sigma") for an objective estimation of hazard and risk associated to EGS activities;
(4) Define sensible thresholds for the various stakeholders and implement these thresholds as part of an advanced traffic light systems, which is forward looking, validated abd dynamically updated on the fly as new data arrives.
The results of WP5 build upon work completed in WP3 and WP4, and are an integral input to WP6 and discussed there also. With respect to managing induced seismicity, we believe it is fair to say that GEISER WP5 has made substantial progress: We now have a framework for hazard and risk assessment that is accepted by a larger community and serves the need of a wide range of stakeholder. This framework will have to be expanded and validated in future pilot and demonstration projects.
WP6: Strategies for EGS operations with induced seismicity
WP6, -in close cooperation with WP5- provides guidelines for safe and reliable EGS Operations.
Task 6.1 soft stimulation investigated the influence of injection parameters to reservoir seismicity due to the stimulation treatment, reviewing past permeability enhancement experiences and drawing idealized scenarios.
At the moment there is no 3-dimensional fracture simulator simultaneously covering the reservoir development and the behaviour of the surrounding rock mass before, during, and after the stimulation treatment. Some of the capabilities that a complete tool to model the fracture creation/stimulation should possess have been addressed; a complete 3-dimensional simulator would prove useful in real-time operation to understand field data collected real-time, since new data allows refining and better definition of the stimulation treatment, not excluding the definition of a traffic light system capable of forecast based on injection parameters and not only reacting on recorded seismicity.
A combination of different numerical tools has been employed to define the response of the reservoir to different injection parameters (temperature, pressure, fluid volume variation in time and space), integrating results obtained in WP4.
ARMINES extended the outcome of the work done in WP 4.4 investigating the effect of injected fluid temperature and thermal stress build up due to stimulation treatment. The numerical model developed was validated against the Rosemanowes (UK) test site data. A significant result comes out from the modelling exercise: seismic activity during the life cycle of a reservoir is not independent of pre-stimulation phases during initial reservoir development. This is of prior interest for mitigation strategies, since the thermal stress magnitude depends on temperature and volume of injected fluid.
GEOWATT AG contributed to subtask 6.1 by evaluating the role of existing (or expected) large scale faults in proximity of the engineered reservoir. Faults with different orientations, but with the same mechanical properties have been investigated. An important outcome of the study is that the highest danger of large magnitude events might not come from fractures that are optimally oriented to fail, because of the anisotropy of the pressure front and stress field heterogeneities in the model. This result is in agreement with field observations in Soultz (FR) and in Basel (CH).
GFZ investigated the reservoir response in term of pore pressure and stress change due to a tensile fracture opening, relating the fracture volume with the influenced reservoir volume. Some of the characteristics that a 3-dimensional fracture simulator should possess have been addressed. Stress shadow effects from the fracture and delayed thermal stress effect have been evaluated and compared with observations from Groß Schönebeck (DE) geothermal test site.
Stimulation treatment design is a critical process in engineering a geothermal system, however the knowledge of the target reservoir before completing the actual field operations is quite limited. Therefore, predicting the seismic behaviour of the reservoir can be difficult. The assumption behind the designed stimulation treatment may have to be updated on the fly, during the field operations, especially with respect to fault orientation and mechanic properties.
It is possible, however, to perform sensitivity analysis on the relevant mechanical and hydraulical parameters of the structure that may be reactivated. A range of stimulation scenario can then be defined and eventually integrated into a traffic light system based on injection parameters, to provide a more robust forecast capability, with respect to a system based only on recorded microseismicity.
The results of Task 6.1 have been included in deliverable D6.1
Task 6.2 seismic monitoring developed guidelines for seismic monitoring at EGS projects. The goal is mitigation of induced seismicity, in particular the prevention of large magnitude events (LME), i.e. seismic events that may cause nuisance and/or damage to local population, industry and/or infrastructure. The guidelines as delivered in deliverable D6.2 are aimed at both regulators and operators and address two import aspects of seismic monitoring: (1) monitoring requirements, defined in terms of data quality, temporal/spatial coverage, resolution, etc., and (2) monitoring network design and optimization, which is presented in a number of case studies.
For the monitoring requirements (1) we choose to follow the recent recommendations by the FKPE (Forschungskollegiums Physik des Erdkörpers) working group on “Induced seismicity” for seismic monitoring in the general context of geotechnical facilities. The FKPE recommendation is intended for reliable observation of earthquakes one magnitude unit below the human perception threshold. We adopt the FKPE recommendations for what we refer to as (a) basic seismic monitoring. However, the FKPE explicitly notes that their recommendation is not sufficient when the monitoring goal is to gain insight into the seismogenic processes. In the context of EGS and mitigation of induced seismicity the understanding of the seismogenic processes is of critical importance. We therefore extend the FKPE recommendations down towards lower magnitudes for what we refer to as (b) reservoir seismic monitoring. In our requirements we identify three main themes that are developed for both basic (a) and reservoir (b) seismic monitoring: (i) data quality: defined in terms of signal-to-noise ratios, frequency content and time-stamping accuracy; (ii) spatio-temporal coverage: defining both spatial and temporal intervals for the monitoring efforts; and (iii) data policy: where a general recommendation is to be as open as possible.
For the monitoring network design and optimization (2), we note that monitoring at least requires a network of continuously operating, time-synchronized seismic sensors. The network should be designed in such a way that it satisfies the monitoring requirements discussed above. In a surface network the sensitivity of a usually goes down to magnitudes around M=1 (at typical EGS reservoir depths), which is roughly one to two magnitude units below the perception threshold (i.e. the level that can be felt by humans at the surface). To extend the sensitivity of the network further down the magnitude scale it is usually necessary to install sensors in (deep) boreholes. To optimize the network design various approaches are possible. With (partial) contribution of GEISER a number of case studies on network design and optimization have been performed and included in deliverable D6.2. These studies are provided as illustrations of proposed techniques for the design and optimization of seismic networks.
Task 6.3 Real-time tools to monitor the evolution of induced microseismicity and Task 6.5 Provide boundary conditions for regulatory guidelines have built input and boundary conditions for regulatory guidelines included in deliverable D6.3
Activities in these tasks focussed on building a shared conceptual framework for this deliverable in close feedback with other Work package activities providing input. To this end a number of workshops have been organised between WP 5 and WP 6, including a workshop in Utrecht in March 2011, one in Zürich in November 2011 and various work meetings. From a regulatory perspective the boundary conditions need to be closely connected to conditions and evaluation criteria for licensing of activities. As an outcome of joint work package meetings during the general assembly in 2011 it became apparent that we needed to further synergize the insights of WP 3 and WP4 into WP 5 and WP6. To this end in Q1 2012 a joint workshop between WP3,4,5,6 has been devoted to the outline of key parameters which control induced seismicity and to discuss strategies how a priori assessment, monitoring and validations of these can be incorporated in a regulatory framework and technical guidelines. As an outcome of the Q1 2012 workshop it has been decided to organize a follow up workshop Q3 2012 for the technical guidelines serving as input for regulatory quidelines.
These technical guidelines have been described in detail in the following deliverables of Geiser:.
1) Input Guidelines for best practice for seismic hazard assessment (D5.6)
2) effect of different stimulation techniques on the seismicity and strategies to mitigate induced seismicity (D6.1)
3) Technical best practices for monitoring (D6.2)
In addition a separate deliverable of Geiser deals with Socio-economic best practices(D6.4) , which is also of great relevance to providing regulatory guidelines
Activities in these tasks dealt largely with fitting the fore mentioned technical and socio-economic guidelines in existing regulatory frameworks (France, Germany, Iceland, Italy, Netherlands, Switzerland). To this end we introduced a conceptual development framework for EGS, identifying main toll-gates for regulatory guidelines. Next we outlined the main features of various national mining laws and its practical regulatory implementation. For each of these an outline is given how the law and regulatory framework are or could be applied to EGS development.
The following recommendations for European regulatory guidelines have been included in Deliverable 6.3 to prevent unsolicited effects of induced seismicity:
1. to set a maximum level of acceptable magnitude (Mtol), and associated treshold probability (Ptol) for earthquakes caused by stimulation and production. The project should be halted if the expected probability of Mtol is higher than Ptol and no operational adjustment can lower this probability.
2. to request the project developer for an assessment of expected probability of Mtol prior to stimulation and to set-up an advanced traffic light system to safeguard that the expected probability remains lower than Ptol, based on a physics based approach (see deliverable 5.6 of GEISER).
3. to request the project developer to set-up a seismic monitoring network which follows GEISER recommendations (deliverable 6.2)
4. to provide incentives for project strategies promoting public acceptance (cf deliverable 6.4 of GEISER). One such incentive can be a differentiation in exploration licensing, allowing for a desk study phase, prior to the drilling and stimulation license. The desk study phase allows to build an outreach program.
Task 6.4 Public awareness and sensitivity to ground shaking due to injection and/or during EGS operations
In 2011 the activities within task 6.4 have been focused on the development of a strategy for creating public acceptance of EGS-projects. First, we did a literature review, based on empirical literature about current/past EGS-projects and theoretical literature about public acceptance. Next, we developed a draft strategy for creating public acceptance for EGS-projects. The results of the research activities of task 6.4 in 2011 have been collected in a progress report task 6.4. The research activities in In 2012 and 2013 focused on applying the strategy for creating public acceptance of EGS-projects on two case studies: the Deep Heat Mining project in Basel (Switserland) and the EGS research project in Soultz-Sous-Forêts (France). we would like to validate the draft strategy for creating public acceptance for EGS-projects. The results of the literature study in 2011 and the analysis of two case studies have been descripted in the final report D6.4 “Laying the Groundwork for Public Acceptance of Enhanced Geothermal Systems”, which also includes recommendations for creating public acceptance of future EGS-projects.
The proposed strategy for creating public acceptance entails the following steps:
• Preparation and context analysis: To identify different interests and (perceived) risks regarding a specific EGS project. This entails a cost-benefit balance for the stakeholders throughout the entire exploration and production workflow. This balance requires to take into account both technical-spatial and social-economic aspects in order to define the project strategies for creating public acceptance of EGS at a specific location (Figure 9.4).
• Process design: The different interests and (perceived) risks allows to characterize the policy challenge(Figure 9.4). The policy challenge needs to be proper addressed through the project strategy, including a communication strategy and process definition of involvement of actors and associated actions.
• the execution of the preparation, context analysis and process design are recommended to overlap with the common planning phase of EGS projects
• Implementation and evaluation: The next phases of developing an EGS project (drilling, logging and testing; stimulation; operation; post-operation) should correlate to the implementation phase of the project strategy. During the implementation of the project strategy (cq. stages of development of the EGS-project) the progress of the process will be constantly monitored and evaluated. If needed, the project strategy will be changed and adopted to the process dynamics.
References
Aochi, H., B. Poisson, R. Toussaint, X. Rachez, and J. Schmittbuhl (2012), Self-induced seismicity due to fluid circulation along faults, Geophysical Journal International, in press.
Bachmann C.E Wiemer S., Goertz-Allmann B.P. & Woessner J., 2012, Influence of pore-pressure on the event-size distribution of induced earthquakes, Geophysical Research Letters 39, L09302, doi:10.1029/2012GL051480.
Bethmann F., Deichmann N., & Mai M., 2011, Scaling relations of local magnitude vs. moment magnitude for sequences of similar earthquakes in Switzerland, Bulletin of the Seismological Society of America 101(2), 515-534, doi:10.1785/0120100179.
Calò M., Dorbath C., Luzio D., Rotolo S.G. & D’Anna G., 2009, Local Earthquakes Tomography in the southern Tyrrhenian region: geophysical and petrological inferences on subducting lithosphere, In: Subduction Zone Geodynamics, Frontiers in Earth Sciences, S. Lallemand and F. Funiciello (eds.), Springer-Verlag Berlin Heidelberg, 85-99, doi: 10.1007/978-3-540-87974-9.
Calò M., Dorbath C., Cornet F.H. & Cuenot N., 2011, Large-scale aseismic motion identified through 4-D P-wave tomography, Geophysical Journal International 186, 1295-1314, doi: 10.1111/j.1365-246X.2011.05108.x.
Calò M., & Dorbath C., 2013a, Different behaviours of the seismic velocity field at Soultz-sous-Forêts revealed by 4D seismic tomography: case study of GPK3 and GPK2 injection tests, Geophysical Journal International, doi: 10.1093/gji/ggt153.
Calò M., Kinnaert X., & Dorbath C., 2013b, Procedure to construct 3-D models of geothermal areas using seismic noise cross-correlations: application to the Soultz-sous-Forêts enhanced geothermal site, Geophysical Journal International, doi: 10.1093/gji/ggt205.
Catalli, F., Meier, M.A. and S. Wiemer (2013), The role of Coulomb stress changes for injection-induced seismicity: The Basel enhanced geothermal system, Geophys Res Lett, 40 (1), 72–77.
Deichmann N., & Giardini D., 2009, Earthquakes induced by the stimulation of an Enhanced Geothermal System below Basel (Switzerland), Seismological Research Letters 80(5), 784-798, doi:10.1785/gssrl.80.5.784.
Deichmann N., Kraft T., & Evans K.F. 2013, Identification of faults activated during the stimulation of the Basel geothermal project from cluster analysis and focal mechanisms of the larger magnitude events. Submitted to Geothermics Special Issue on Induced Seismicity.
Douglas, J. (2013), Seismic network design to detect felt ground motions from induced seismicity, Soil Dynamics and Earthquake Engineering, 48, 193-197, doi:10.1016/j.soildyn.2013.01.030.
Douglas, J. and P. Jousset (2011), Modeling the difference in ground-motion magnitudescaling in small and large earthquakes, Seismological Research Letters, 82(4), 504-508, doi: 10.1785/gssrl.82.4.504.
Douglas, J., B. Edwards, V. Convertito, N. Sharma, A. Tramelli, D. Kraaijpoel, B. M. Cabrera, N. Maercklin and C. Troise (2013), Predicting ground motion from induced earthquakes in geothermal areas, Bulletin of the Seismological Society of America, 103(3), in press, doi: 10.1785/0120120197.
Douglas, J. and H. Aochi (2013), Using estimated risk to develop exploitation strategies for Enhanced Geothermal Systems, Geophysical Journal International (?), in preparation.
Edwards B., and J. Douglas (2013), Magnitude scaling of induced earthquakes, Geothermics, submitted November 2012.
Edwards B. and J. Douglas (2013), Selecting ground-motion models developed for induced seismicity in geothermal areas, Bulletin of the Seismological Society of America (?), in preparation.
Deichmann N., & Ernst J., 2009, Earthquake focal mechanisms of the induced seismicity in 2006 and 2007 below Basel (Switzerland), Swiss Journal of Geosciences 102(3), 457-466, doi: 10.1007/s00015-009-1336-y.
Gischik, V., and S. Wiemer (2013) A stochastic model for induced seismicity based on non-linear pressure diffusion, Geophys J Int, 194, 1229-1249, DOI: 10.1093/gji/ggt164.
Goertz-Allmann B.P. Goertz A., & Wiemer S., 2011, Stress drop variations of induced earthquakes at the Basel geothermal site, Geophysical Research Letters 38, L09308, doi: 10.1029/2011GL047498.
Goertz-Allmann B.P. & Wiemer S., 2013, Geomechanical modeling of induced seismicity source parameters and implications for seismic hazard assessment, Geophysics 78(1), 25-39, doi: 10.1190/GEO2012-0102.1
Hakimhashemi, A. Yoon JS, Heidbach, O., Zang, A., Grünthal, G. (2013), FISHA - FORWARD INDUCED SEISMIC HAZARD ASSESSMENT APPLICATION TO SYNTHETIC SEISMICITY CATALOG GENERATED BY HYDRAULIC STIMULATION MODELING, Geothermics, in press.
Häring M.O. Schanz U., Ladner F., & Dyer B.C. 2008, Characterization of the Basel 1 Enhanced Geothermal System, Geothermics 37, 469-495, doi: 10.1016/j.geothermics.2008.06.002.
Kraft T., & Deichmann N., 2013, High-precision relocation and focal mechanism of the injection induced seismicity at the Basel EGS, Submitted to Geothermics.
Kwiatek G., Bohnhoff M., Martínez-Garzón P., Bulut F., & Dresen G., 2013,. High-resolution reservoir characterization using induced seismicity and state of the art waveform processing techniques. First Break 31(7), 81-88.
Martínez-Garzón P., Bohnhoff M., Kwiatek G., & Dresen G., 2013, Stress tensor changes related to fluid injection at The Geysers Geothermal Field, California, Geophysical Research Letters, in press, doi: 10.1002/grl.50438.
Mena, B., S. Wiemer, and C. Bachmann (2013), Building Robust Models to Forecast the Induced Seismicity Related to Geothermal Reservoir Enhancement, B Seismol Soc Am, 103(1), 383-393.
Mignan, A., D. Landwing, B. Mena and S. Wiemer (2013), Induced seismicity risk assessment for the 2006 Basel, Switzerland, Enhanced Geothermal System project: Role of parameter uncertainty on risk mitigation, Geothermics, under revision.
Oye V., & Roth M., 2003, Automated seismic event location for hydrocarbon reservoirs, Computers & Geosciences 29, 851-863.
Rowe C.A. Aster R.C. Borchers B., & Young C.J. 2002, An Automatic, Adaptive Algorithm for Refining Phase Picks in Large Seismic Data Sets, Bulletin of the Seismological Society of America 92, 1660-1674.
Schoenball, M, Baujard, C and Dorbath, L. (2012) The role of triggering by static stress transfer during geothermal reservoir stimulation, J. Geophys. Res., 117, DOI: 10.1029/2012JB009304
Waldhauser F., & Ellsworth W.L. 2000, A Double-Difference Earthquake Location Algorithm: Method and Application to the Northern Hayward Fault, California. Bulletin of the Seismological Society of America 90(6), 1353-1368.
Potential Impact:
The main impact expected from the project is the establishment of a procedure to realise the goals of enhancing geothermal systems with a reliable concept for the mitigation of induced seismicity. This concept will ensure that geothermal energy can reach its full efficiency and profitability thresholds at the European scale.
The guidelines provided by GEISER will help to reduce uncertainties with respect to effects of induced seismicity and how to mitigate them. As induced seismicity has been one of the major road blocks in the development of EGS, the results of this project will help to increase EGS market introduction by providing:
Licensing guidelines
Planning security
These guidelines not only address the measures to be taken to monitor, control and mitigate induced seismicity during injection in geothermal operations but also include recommendations on how to build trust in the population and increase public acceptance.
Regulatory guidelines handling operational hazards.
With the provision of regulatory guidelines handling operational hazards due to underground exploitation, authorities will have better defined regulations and legal guidelines to cope with both political and physical damage of induced seismicity. This will provide planning security for developers and communities and is likely to unleash activity in the geothermal energy market.
Optimal monitoring infrastructures of underground exploitations.
The GEISER project successfully compiled and categorized data sets of induced seismicity in geothermal projects that can now be accessed by other users. Optimized monitoring systems were proposed that can be used by developers. It should be noted that, although the monitoring methods are specifically designed for the development of EGS, but they can also be of great interest in hydrocarbon exploration, in the search for deep aquifers, in the planning of waste disposals and in other applications.
The success of the GEISER-project should have a kick-off effect on the development of geothermal resources in Europe but also it shall enhance the capability of European industry to compete on the world market of geothermal business. The project will represent a technological milestone in the development of a renewable, cost-competitive geothermal energy (electric power and heat) which makes it possible for Europe to achieve:
the improvement of the standard of living and safeguard of the environment, contributing to the minimisation of gas emissions, thereby reducing the greenhouse effect. Limiting the considerations to the actual geothermal energy utilisation all over the world, it is equivalent to the burning of 150 million barrels of oil per year. In Europe alone, every year geothermal electricity avoids discharges of 5 million tons of carbon dioxide, 46000 tons of sulphur dioxide, 18000 tons of nitrogen oxide and 25000 tons of particulate matter into the atmosphere compared to the same production from a typical coal-fired plant. These numbers will increase with the successful development of more geothermal power plants
the creation of new employment, involved in exploration, well drilling, plants design and construction and all related activities of the operation and maintenance of power plants. Oil and gas service companies, drilling companies as well as machinery manufacturers and engineering suppliers will all benefit from the development. All these actions, applied on a vast scale, would not only guarantee new jobs, but also create new professional positions such as dedicated service men and builders
the development of new technologies and skills, as an outcome of the project will improve the competitiveness of the and increase the capability of geothermal as well as service companies to open up to new markets both in geothermal and hydrocarbon exploration sectors in Europe and in the developing areas of the world. The project target is to deal with methodological development of advanced techniques applicable to mitigate induced seismicity. This innovative approach applied to geothermal energy provision is also relevant in the prospect of geothermal resources in Europe (e.g. Italy, Portugal, Greece, Iceland, France, Germany, Eastern Europe and Turkey) and in the development of other areas of the world (Central and South America, South East Asia, Africa).
The project successfully bridges the gap between the scientific development of advanced monitoring technologies and in the application of operational services. Moreover, geophysical methods and tools developed within the GEISER-project, will apply to waste water injection and other sources of mining induced seismicity. The GEISER-project will be an important contribution to focusing on the development of advanced geophysical technologies aimed at the improvement of the production of geothermal systems and at providing more reliable tools to reduce the mining risk of geothermal field development. The scientific know-how generated by GEISER represents a major contribution to the geological and petrophysical interpretation of seismic sounding carried out in many European countries and for the investigation of georesources.
List of Websites:
www.geiser-fp7.eu
GEISER worked specifically at developing a better understanding of the key parameters that control induced seismicity in response to an injection. This improved understanding serves as input for the necessary seismic hazard and risk assessment of future EGS projects. As a basis for all analyses the GEISER research team has analyzed a series of past stimulation experiments from projects such as Soultz-sous-Forêts (France), Basel (Switzerland), Berlìn (El Salvador) and Gross Schönebeck (Germany) as well as from numerous Icelandic fields, where injections caused seismicity. In addition, datasets from The Geysers in California and from Cooper Basin (Australia) were provided to GEISER.
Analyses of the temporal and spatial evolution of the average size distribution and stress drop of the thousands of seismic events observed during the injections at the various sites revealed a pattern with the largest seismic events often occurring in the hours and days after the stop of injection. The relation between injected hydraulic energy and released seismic energy seems to be characteristic for many projects. This observation provides clues to the development of mitigation strategies and soft stimulation techniques such as a cycling of the injection pressure.
The relative size distribution of micro-earthquakes at sedimentary seems to be different from sites accessing reservoirs in or near the granitic bedrock in the Rhine valley. In some cases, a lower seismic hazard is evident in sedimentary rocks.
From the joint research efforts, GEISER presents a number of results and recommendations of major relevance for geothermal operations:
For the first time, ground motion prediction equations have been derived specifically for geothermal sites. At geothermal sites small magnitudes, high frequencies and short distances are much more relevant than for normal tectonic hazard assessment. Ground motion prediction equations forecast the level of shaking expected at the surface for a given seismic event. They are a key ingredient in any seismic hazard assessment.
A comprehensive framework for the assessment of seismic hazard and risk was developed. The framework is probabilistic in nature, allowing the representation of the uncertainty in numbers and in our understanding of the processes to be captured. GEISER also provides guidelines for safe and reliable EGS operations. Key is a dynamic –forewarning- advanced traffic light system that can be used in all phases of a project. This approach allows adjustment the operational conditions to mitigate unsolicited effects and to improve system performance. It incorporates site specific parameters, derived from the geological studies, from drilling as well as logging.
A strategy is proposed to enhance public support for EGS projects, based on lessons learned from past projects. A cost-benefit balance for the stakeholders throughout the entire exploration and production workflow is important, capable of identifying and properly addressing different interests and risks regarding a specific EGS project. Special attention has to be paid to risk perception. Nuisance and trivial damage should be addressed with care and considered as a significant project risk. For non-structural damage, a pre-agreed procedure is needed to evaluate and compensate the costs.
The GEISER research efforts and guidelines are important steps to enable the efficient and safe use of deep geothermal energy resources throughout Europe.
The GEISER results also point to future challenges. Especially, the modeling approaches and software developed for the assessment of induced seismicity now needs to be validated against a wider range of experimental data, including future injections. Further improvements depend on our ability to image larger and potentially tectonically loaded faults and to incorporate empirical experience. However our means to forecast a distinct seismic response in the underground will remain limited.
Project Context and Objectives:
Geothermal operations require production and injection of fluids. In the case of Enhanced Geothermal System (EGS) operations are usually related to engineering of hot, low permeability rock. Similar operations are common, and will be applied in the future to a large extent in conventional fracture dominated geothermal systems. The reason for this is that geothermal energy can only be extracted from a small volume of hot rock surrounding the fracture systems. Permeability enhancement in boreholes and creation of artificial fracture for production, therefore, are considered crucial for future development of the reservoirs. The same applies to the re-injection wells in fracture-dominated reservoirs. The fluid injected must go either into open fractures that are a part of a tectonic fault system, or into new fractures that must be created. In general, the basic knowledge on design and safe operation of fluid injection is based on a geologic map with fault/fracture distribution, tectonic movements, stress field, and natural seismicity with estimates of the moment magnitude of the largest seismic event to be expected in the area of interest.
The need to inject fluids to enhance productivity of a geothermal well often induces seismicity. Earthquakes induced by human activities occur in most mining-related operations, depletion of oil and gas reservoirs, fluid injection in the subsurface and dam impoundment and they often reduce public acceptance of such ventures. Promising geothermal projects jeopardized by this problem are at Soultz-sous-Forêts and at Basel. In the latter case, repeated seismic events, although not destructive, were felt by the local population and prompted the authorities to halt operations. To avoid these problems, action has to be taken in order to better understand and mitigate induced seismicity in the development of geothermal reservoirs.
An improved understanding of the causes and processes leading to induced seismicity in geothermal engineering and the development of appropriate mitigation strategies were the general objectives of the GEISER project. In particular, specific goals were defined
to understand why seismicity is induced in some cases but not in others
to determine the potential hazards depending on geological setting and geographical location
to work out licensing and monitoring guidelines for local authorities, which should include a definition of what level of ground motion is acceptable
to develop strategies for successful stimulation and improvement of the hydraulic properties of the geothermal reservoir without producing large magnitude events that pose a threat to buildings and disturb the public.
These objectives were approached in several work packages, addressing specific aspects of the general topic.
WP 2: Compilation of induced seismicity data from geothermal sites. The compilation served as a basis for all subsequent analyses. In this WP, data on induced seismicity from representative sites were collected, to provide an overview of lessons learned from previous experience. Data were checked and homogenized for comparison and further use in the other workpackages.
WP 3: Analysis of Induced Seismicity. Different seismological approaches were applied and further developed to analyse data sets of induced microseismicity from geothermal areas. Based on the evolution of induced seismic activity in space and time, the interrelation between the specific local geological settings, injection parameters and the occurrence of fluid-injection induced seismic events was addressed.
WP 4: Understanding the Geomechanical Causes and Processes of Induced Seismicity. Different modelling approaches and laboratory experiments were performed to investigate some of the key factors influencing induced seismicity: The goal was to come to a better understanding of the basic physical mechanisms that induce microseismicity and to deliver management and production scenarios with relative estimates of the stress-state changes for different geothermal settings.
WP 5: Seismic Hazard Assessment. On the basis of the results from WPs 3 and 4 and considering natural seismicity, WP5 assembled all the components of hazard assessment to be conducted before the selection and start of operations at an EGS site, to result in Guidelines for best practice. This included the analysis of the natural background seismicity, an estimate of the ground shaking produced by the microseismicity induced during the initial stimulation phase and the probability of triggering a large earthquake (M>4) ahead of its natural time of occurrence, either during the stimulation phase or during the long-term EGS operation.
WP 6: Strategies for EGS operations with respect to Induced Seismicity (Mitigation). In this workpackage strategies for operators and regulatory bodies were proposed to develop guidelines for the selection, licensing and long-term operation of EGS sites in different geological settings. Recommendations for proper monitoring of induced seismicity are given, and for a minimization of the risk associated with induced seismicity.
GEISER worked specifically at developing a better understanding of the key parameters that control induced seismicity in response to an injection. This improved understanding was part of the seismic analysis in WP3 and the mechanical considerations in WP4 and serves as input for the necessary seismic hazard and risk assessment of future EGS projects. For this purpose, the GEISER team has analyzed a series of past stimulation experiments from projects such as Soultz-sous-Forêts (France), Basel (Switzerland), Berlìn (El Salvador), various Icelandic sites where re-injection of thermal fluids caused seismicity and Gross Schönebeck (Germany). In addition, datasets of seismic events recorded at The Geysers in California were provided to GEISER as well as stimulation data including the corresponding seismicity from Cooper Basin (Australia).
As a result of the seismic analyses, GEISER provided Guidelines for techniques/methodologies for seismological investigations to be applied in future EGS operations, developed on the basis of successful analyses of past sequences (Deliverable D3.3). In these guidelines it is recommended that all observables (1) in-situ stress, 2) hypocenter locations, 3) seismic moments, 4) stress drop and 5) b-value) should be investigated during or immediately after stimulation as their combination allows direct estimates of the spatial and temporal evolution of pore pressure in the stimulated rock volume. Thus, semi-automatic methodologies need to be developed to determine these parameters in near-real-time. The last three points, if available in real-time, constitute also essential input to advanced, dynamic traffic-light systems that are based on probabilistic seismic hazard assessments during ongoing stimulation.
To investigate the effects of changes in pore pressure, temperature and the effect of pre-existing fault segments, laboratory experiments and several modeling activities with complementary approaches were performed throughout WP4. Different numerical tools have been developed, or adapted, in order to study the coupled mechanisms that induce seismicity during EGS operations. The efforts put in WP4 also include the development of tools to simulate catalogs of synthetic events, based on rock mass physical features and on injection rates, that are useful in order to assess the seismic hazard associated to EGS operations.
Another ingredient to understand and mitigate induced seismicity was provided by WP5 which provided a General probabilistic hazard and risk assessment framework for hazard associated with natural seismicity and to the seismicity induced and triggered during and after EGS operations. Within this workpackage, ground motion prediction equations have been derived specifically for geothermal sites am. At geothermal sites small magnitudes, high frequencies and short distances are much more relevant than for normal tectonic hazard assessment. Ground motion prediction equations forecast the level of shaking expected at the surface for a given seismic event. They are a key ingredient in any seismic hazard assessment. In addition, hazard models for specific test areas were developed. These models are based on a) statistical forecast approaches such as the Gutenberg-Richter model, and on b) physical considerations including geomechanical aspects, so called deterministic approaches.
The two types of proposed forecast models, statistical and physical, are complementary. At the present time, statistical models appear as a reasonable choice to forecast induced seismicity in a prospective way for decision support (e.g. traffic light systems). It has been shown that they fit the data well and that observed variations between the best models (Shapiro-type and ETAS modified - used in the logic tree of Fig. 1) have a low impact on the overall uncertainty (Mignan et al., in revision). Physical models can also well reproduce observed induced seismicity sequences (e.g. Gischig and Wiemer, 2013). While statistical models may outperform physical models due to a lower number of parameters and a faster computation, physical models are crucial for a better understanding of the evolution of induced seismicity over longer time horizons, of b-value changes, or of Mmax (to only cite a few). However at the present time, there is no consensus on which physical model best describes induced seismicity.
Based on the innovative hazard and risk assessment framework described above, it was possible to provide some recommendations to stakeholders (Deliverable 5.6 Recommendations for induced seismicity hazard and risk management).
On the basis of recommendations derived in WPs 3,4 and 5, guidelines for safe and reliable EGS operations were proposed in WP6. These guidelines are presented in project deliverables
D6.1 Description of the effect of different stimulation techniques on seismicity and strategies to mitigate induced seismicity
D6.2 Description of optimized monitoring strategy, both for the permanent receiver network and the network specifically installed during stimulation treatments
D6.3 Report summarizing the results of the tasks in form of input and boundary conditions for regulatory guidelines
D6.4 Report on development of public awareness and acceptance
In previous projects, a so-called traffic light system had been proposed to better control induced seismicity. This approach was based on a reaction to increases in reservoir pressure during injection. GEISER also provides guidelines for safe and reliable EGS operations. Key is a dynamic – forewarning – advanced traffic light system that can be used in all phases of a project. It requires, however, the determination of a maximum acceptable seismic magnitude and its accepted probability of occurrence. The reliability of the dynamic model is based on the availability of rock physics and seismic data, with models updated from real-time monitoring. This approach allows adjusting operational conditions to mitigate unsolicited effects and to improve system performance. It incorporates site specific parameters, derived from the geological studies, from drilling as well as logging. The efforts in induced seismic hazard and risk assessment have to mimic the potential damage.
The GEISER research efforts and guidelines are important steps to enable the efficient and safe use of deep geothermal energy resources throughout Europe.
This section must be of suitable quality to enable direct publication by the Commission and should preferably not exceed 40 pages. This report should address a wide audience, including the general public.
Project Results:
WP3 Analysis of Induced Seismicity in Geothermal Reservoirs
The objectives of the work package were to apply different seismological approaches to analyse data sets of induced micro-seismicity from geothermal areas. Induced seismic activity in space and time was examined with a view to establishing any systematic correlation with local geologic setting and injection parameters. These points were addressed in tasks defined to investigate three key aspects in detail
Spatiotemporal characteristics of fluid-injection induced microseismicity
Occurrence of large magnitude events (LME) in geothermal reservoirs
Role of pore pressure and crustal stress in induced seismicity
For this purpose, the datasets collected in WP 2 were used for detailed analysis.
The research activities in this WP were used directly for recommendations developed in WPs 5 and 6. Three deliverables resulted from the work performed.
D3.1 Evaluation of systematic relations between the seismic response to fluid injection with particular regard to the generation of LME, the depth of injection, lithology, injection pressure, crustal stress state, and local structural geology
D3.2 Source characterisation of LME and their occurrence in time and space
D3.3 Guidelines for techniques/methodologies for seismological investigations to be applied in future EGS operations, developed on the basis of successful analyses of past sequences.
In particular, D3.3 was designed to provide practical guidelines for future geothermal energy developments. Deliverables D3.1 and D3.2 were used to provide input for the guidelines provided in D3.3.
Summary of the results and recommendations
Enhanced Geothermal System (EGS) operations are usually related to engineering of hot, low permeability rock. Similar operations are common, and will be applied in the future to a large extent in conventional fracture dominated geothermal systems. The reason for this is that geothermal energy can only be extracted from a small volume of hot rock surrounding the fracture systems. Permeability enhancement in boreholes and creation of artificial fracture for production, therefore, are considered crucial for future development of the reservoirs. The same applies to the re-injection wells in fracture-dominated reservoirs. The fluid injected must go either into open fractures that are a part of a tectonic fault system, or into new fractures that must be created. In general, the basic knowledge on design and safe operation of fluid injection is based on a geologic map with fault/fracture distribution, tectonic movements, stress field, and natural seismicity with estimates of the moment magnitude of the largest seismic event to be expected in the area of interest.
Interface issues
In case no seismic catalogue is available for the geothermal site of interest, we advice to use synthetic catalogues based on physical model approaches (WP4). Different stimulation scenarios for one and the same reservoir and fracture geometry should be used in order to find safer injection strategies. In this context, we recommend to connect these models to probabilistic seismic hazard assessment (WP5). In particular we advise to use the method of Forward Induced Seismic Hazard assessment (FISHA) based on zero a priori seismicity information as opposed to the conventional probabilistic seismic hazard assessment (PSHA) applied to induced seismic events obtained real-time.
Conclusion
As the combination of all observables ((1) in-situ stress, (2) hypocenter locations, (3) seismic moments, (4) stress drop and (5) b-value) allows direct estimates of the spatial and temporal evolution of pore pressure in the stimulated rock volume, every effort should be made to perform such investigations during or immediately after stimulation. Thus, semi-automatic methodologies need to be developed to determine these parameters in near-real-time. The last three points, if available in real-time, constitute also essential input to advanced, dynamic traffic-light systems that are based on probabilistic seismic hazard assessments during ongoing stimulation.
Partner contributions
AMRA devoted their attention to two geothermal fields: The Geysers in California (USA) and Campi Flegrei caldera (Italy). They analyzed a seismic waveform dataset of more than 15000 events (1.0 <ML <4.5) recorded at The Geysers geothermal field by a dense surface seismic array during August 2007 to October 2011. The obtained results provided interesting issues concerning the investigated area characteristics, but above all, they highlighted very important general indications about the importance in geothermic contexts of seismological analyses for characterizing the medium mechanical properties.
The experience of The Geysers study showed that the strong variation in P-wave velocity and Vp/Vs (Fig. 1) are closely related to a variation of the attenuation. In turn, all these parameters have been found consistently with the subdivision of The Geyser field in two sub-regions characterized by significantly different medium and fracture source properties.
A further important issue related to the availability of high resolution 3D velocity models is that they would be very important for improving the earthquakes location, and thus for obtaining a better reconstruction of the fractures orientation and dimension.
Fig. 1 Map view showing the Vp/Vs ratio inferred from the 3-D tomography, at 0, 1, 1.5 and 2 km depth, at The Geysers.
NORSAR has focused on the analysis of data from two sites, (1) Paralana, in South Australia, a new enhanced geothermal system (EGS) where the first main stimulation occurred in July 2011 and (2) Basel, in Switzerland, an EGS which was stopped during the stimulation phase in 2003 due to the occurrence of a large magnitude event (LME).
As EGS operations normally produce a large number of microearthquakes, automatic processing is usually required. However, there is an uncertainty of automatic picks, if for example the data are noisy or if several events occur at the same time. A careful check of the picks, based for example on location error and residual values, is always a first step for improving the data base. Another step to refine phase picks is to perform waveform cross-correlations for pairs of earthquakes recorded at the same station (Rowe et al., 2002). The approach at Paralana was first to process the data automatically with MIMO (Oye and Roth, 2003), to assess the quality of the automatic picks using a reference subset of events picked manually, and to cross-check the database manually (for events with magnitude larger than one and for the large residuals). Then, we improved the travel-time differences with waveform cross-correlation (Fig. 2) and relocated the events using the double-difference algorithm (Waldhauser and Ellsworth, 2000)
Figure 2: Seismograms of two events (1 and 2, blue and red, respectively) recorded at a surface station. Waveforms are aligned according to the automatic P-picks (blue star): wrong P-picks for event 2 is then highlighted by the gap between the waveforms (blue line and red dashed line). After waveform cross-correlation, the waveform of the event 2 is correctly aligned (red line) with event 1 and the travel-time difference can be corrected.
ÍSOR
Seismicity data from three Icelandic geothermal fields, namely Krafla, Hengill and the Reykjanes peninsula, have been analysed. Data from Krafla that were obtained during the drilling of the IDDP well in 2009 have been analysed in some detail. The well was cased with a steel casing down to 1950 m and with a slotted liner around an aquifer at 1950-2080 depth at a top of a molten or partially molten magmatic intrusion. Low pressure stimulation with cold water during completion of the drilling and subsequent tests induced seismicity of magnitudes up to 1 in local magnitude. The epicentres initially followed the top of the magmatic layer horizontally away from the wellbore until it met an inclined fracture, most likely a pre-existing one. The epicentres then followed this fracture upwards showing the connection between the heat-mining zone at the top of the magmatic layer and the active fault systems.
The data from Hengill contain seismic and pressure recordings obtained during drilling of an injection well in February 2011 for the 303 MWe Hellisheidi power plant in the Western part of the Hengill geothermal area. The injection well was drilled into a complex system of normal NE trending faults belonging to the axial rift zone of the Mid Atlantic Ridge in Iceland and N-S trending right lateral strike slip faults with character of the South Iceland transform zone. During the drilling, the well entered an open fracture at 1320m and total loss of circulation (40 L/s) was observed. A swarm of earthquakes was immediately initiated with magnitudes up to ML 2.2. The earthquakes were clearly felt in the neighbourhood. These data were analysed and used to investigate the interaction between injection and pre-existing fractures. In September 2011 a full scale injection of 550 L/s started into the fissure swarm resulting in high level of induced seismicity. About 3000 earthquakes have been located in the area in 2011. The earthquakes came in intensive swarms with quiet intervals in between. The seismicity culminated on October 15th when several events of magnitude more than 3.0 were measured, the largest one exceeding 3.8. This earthquake swarm occurred in conjunction with major disturbance of the injection rate. This is among the largest quakes that have been triggered by geothermal re-injection in the world. It caused serious inconvenience in a nearby village as people were not prepared.
Figure 3: Location of induced events around the IDDP well at Krafla during low pressure stimulation with cold water.
Figure 4: Induced seismicity during circulation loss of cold water while drilling an injection well.
The data from these latest events at Hengill were not supposed to be included and analysed within the GEISER project as they were collected late in the year 2011, the second year of GEISER. However, as this dataset is of high importance in understanding the earthquake triggering effect of large scale re-injection, some of the obtained were used for the GEISER work. The power company that owns the data agreed to release them to the scientific community.
Data from Reykjanes peninsula include data from two production fields, Svartsengi and Reykjanes. In Svartsengi, production has been ongoing since 1977 and re-injection started in 1984. Reykjanes entered production of 100 MWe in 2006 but without re-injection until 2009. In addition to the national seismic network in Iceland the University of Iceland and the University of Wisconsin operated a local seismic network in Reykjanes from December 2008 until May 2009. Around 320 earthquakes have been located during that period. The seismic dataset consists of continuous waveforms from 11 seismic stations at the tip of the peninsula. The seismic activity was analysed, with relocation of earthquakes using the double-difference algorithm, evaluation of focal mechanisms of the earthquakes, and investigation of the relationship with injection and production data and tectonic structures.
ETH Zürich focused their analyses on the vast data set acquired during the stimulation of the Basel EGS in 2006, which induced about 3500 seismic events. These occurred during the injection phase and the following months and were locatable by the six-station borehole network (Häring et al., 2008). Close to 200 of the strongest events were also recorded by various surface networks (Deichmann and Ernst, 2009; Deichmann and Giardini, 2009). Preliminary hypocenter locations and magnitudes were obtained from manually picked arrival-times and amplitudes. This information was sufficient to map the overall orientation and dimension of the stimulated rock volume in near real-time. More detailed insights into the physical processes were obtained only through systematic re-evaluation of the data set by various groups of scientists over the following years: hypocenter locations were refined in several stages by the application of various methods; magnitudes were re-evaluated and seismic moments were calculated by different groups; focal mechanisms and moment tensors were determined; b-values and stress-drops were mapped in space and time; pore pressure evolution was estimated.
Hypocenter locations are of fundamental importance for all subsequent analyses, providing the basis for seismic velocity models. Poorly calibrated velocity models and arrival-time errors lead to a more or less amorphous appearance of the so-called seismic cloud, which in addition can be systematically mislocated and misshaped (Kraft and Deichmann, 2013). In fact, a large component of the observed scatter of the initial hypocenter locations was due to faulty or inconsistent arrival time picks made under time pressure by inexperienced analysts.
A well-calibrated velocity model, while ensuring the proper location of the seismic cloud as a whole relative to the injection well, is not sufficient to resolve individual faults activated by the stimulation. The detailed analysis of the sequences associated with the larger magnitude events (Mw > 2) induced during the stimulation of the Basel EGS showed that the activated faults have dimensions on the order of several 100 m and are often oriented obliquely to the overall orientation of the microseismic cloud (Deichmann et al., 2013). These results reveal a complex internal structure of the flow paths in the rock volume stimulated by the water injection and imply that geo-mechanical models consisting of a single throughgoing structure are too simplistic.
Although the dimensions of the faults activated by the stimulation of the Basel EGS were estimated on the assumption of a constant stress-drop (Deichmann et al., 2013), the analysis by Goertz-Allmann et al. (2011) shows that stress-drop varies over a wide range and tends to increase systematically with distance from the injection point, which suggests that stress drop correlates with pore pressure perturbations due to the injection. This hypothesis was tested by calculating the injection-related pore pressure perturbation based on a simple linear pore pressure diffusion model and find a good correlation of the expected pore pressure perturbation with the estimated stress drops. These results were reported in more detail in periodic report 2 (2012).
An alternative approach to obtain direct observational estimates of pore pressure within the stimulated reservoir was taken by Terakawa et al. (2012). Using all available observations (orientation of stress field, stress field ratio, friction values determined in the laboratory) and some modelling assumptions (homogeneity of the stress field over the stimulated volume) it was possible to fully characterize the average stress field and to calculate the shear stress on each fault and the fluid pressure necessary to induce failure. Although overall the fluid pressures decrease towards the periphery of the stimulated volume, the pressure field within this volume is characterized by several patches of higher and lower pressures. More important is, however, that according to the results of this analysis the fluid pressures needed to trigger many of the observed earthquakes are in the range of 10 to 20 MPa even out to several 100 m from the injection well. This result is obtained totally independently of the applied injection pressures but is entirely consistent with these pressure values.
To map possible flow paths that have been opened during stimulation, and thus to constrain geo-mechanical models of permeability enhancement, estimates of the dimensions of the activated faults are also necessary. This requires reliable measures of seismic moment or moment magnitude (not just local magnitude) and, if possible, stress drops (Bethmann et al., 2011; Goertz-Allmann et al., 2011). The detailed analysis of the sequences associated with the larger magnitude events (Mw > 2) induced during the stimulation of the Basel EGS showed that the activated faults have dimensions on the order of several 100 m and are often oriented obliquely to the overall orientation of the microseismic cloud (Deichmann et al., 2013). These results reveal a complex internal structure of the flow paths in the rock volume stimulated by the water injection and imply that geo-mechanical models consisting of a single throughgoing structure are too simplistic.
Bachmann et al. (2012) apply high-resolution b-value mapping to induced seismicity, in order to obtain information on the stress regime and possibly the pore-pressure evolution in space and time inside the stimulated rock volume.
Figure 5 (a) Overview of the Basel induced seismicity showing the depth of the 3560 located events (circles) and the location of the seismic stations (triangles). The red plane marks the top of the crystalline basement, within which all events occurred. The color scale indicates the recording completeness ranging from Mw 0.7 to 1 (Mc). (b) Close-up of the events with the overall b-value distribution based on all events. While values range from Mw 0.8 to 3.5 the color bar is limited from 1 to 2 for a clearer visibility. (From Bachmann et al., 2012).
The key findings of Bachmann et al. (2012) are summarized in Figure 5, which shows the strong and highly systematic spatial heterogeneity of the b-values for the seismicity induced by the Basel EGS. Unusually high b-values shown in red are found near the injection point and earlier in the sequence; further out, b-values tend to be closer to the normal tectonic average of around 1.0. Bachman et al. (2012) and in even more detail Goertz-Allmann and Wiemer (2012) have developed a geo-mechanical model that explains these observations and that offers a framework for improved forecasting of induced seismicity. The fact that the largest events in induced seismicity often – but not always – occur after shut-in is well-explained by this model.
EOST
The EOST crew took advantage of the large amount of data gathered by the Soultz EGS scientific project to develop a seismological methodology and obtained a reliable knowledge of the deep structures developing reliable methods that can be applied at various scales and different moments of the development of geothermal projects. The focus was on studying the induced seismicity, seismic noise correlation for exploration and monitoring of geothermal reservoirs, and on seismic tomography using VSP data.
Specific focus of the investigation was on the temporal variations of the elastic parameters through 4D seismic tomography, which is very useful for the understanding of the mechanical behaviour of the geothermal reservoirs, when used with rigorous assessment tests and post-processing procedures (e.g. the Weighted Average Model method; Calò et al., 2009, 2011, 2013a) to avoid misunderstanding due to the presence of bias in the models.
Results of these studies show that injection tests performed in regions initially poorly connected to large faults are characterized by a low anomaly of the P-wave velocity mainly located around the zone where microseismic activity develops. In some specific periods (i.e. when the injected flow rate was suddenly increased) the velocity anomaly disappears suggesting that the velocity variations within the reservoir (and consequently the related variations of effective stress) are not associated with simple water diffusion from the injection well, but rather reflect the occurrence of large-scale aseismic events in the reservoir. In regions where pre-existing faults are well documented, the accumulation of effective stresses close to the well is avoided probably because the structures represent the main paths of the injected water. This results in a lack of large low Vp anomalies during the stimulation and in the occurrence of the induced seismicity located along the major structures.
The noise cross correlation tomography method was tested to image 3D structures of the Soultz-sous-Forêts EGS at small scale (about 1x1 km2). This method proved favourable for application in densely populated regions characterized by high geothermal potential, as is the case for the upper Rhine valley (Calò et al., 2013b). To image the geometrical features of the Soultz-sous-Forêts EGS Rayleigh waves were reconstructed from cross-correlations of 15 months of ambient seismic noise recorded by 23 seismological stations installed around the geothermal power plant. The reconstructed waveforms were used to measure group velocity dispersion curves at periods between 1.0 and 5.0 s. The obtained measurements were inverted for two-dimensional group velocity maps and finally for a 3-D S-wave velocity model of the Soultz region from 0 to 5.2 km depth.
The procedure applied at Soultz can be used as a guideline for the reliable 3D imaging of
geothermal reservoirs (Calò et al., 2013b). Furthermore cross-correlation of ambient noise could be applied to develop geothermal monitoring systems.
Analysis of the VSP data collected in 2008 to better understand permeable fractures serving as main circulation paths of the reservoir gave promising results. The main new insight is the evidence of a previously not identified structure clearly underlined by reflections and a well-marked high velocity anomaly. This structure is located between the wells GPK3 and GPK4 and could explain the low connection between these two wells already observed by hydrologists.
GFZ
The research group at GFZ German Research Center for Geosciences was focused on analysis of seismic data from Berlín Geothermal Field, El Salvador and The Geysers geothermal field in California, USA. The studies performed at these sites were related to the high resolution reservoir characterization using induced seismicity data and state of the art waveform processing techniques.
Three state-of-the-art algorithms, namely the Double-Difference (hypoDD) re-location technique, the Spectral Ratio (SR) technique and the Stress Inversion (SI) method were used to analyse IS generated by fluid injection and steam production. hypoDD significantly improves the precision of hypocenters allowing imaging of the fluid path and propagation in response to multiple injections with unprecedented detail (Kwiatek et al., 2013). In addition, the application of the SR technique provides refined source parameters that can later be used to interpret the subtle interactions between pressure perturbations, fluid flow and fracture (re-) activation within the reservoir. Finally, we show how the SI technique can contribute towards monitoring geomechanical processes occurring in the reservoir in response to short-term injection and long-term production activities (Martínez-Garzón et al., 2013).
Figure 6. Berlìn Geothermal Field (BGF): Spatial distribution of 581 seismic events recorded between Oct-2002 and Jan-2004 (locations provided by BGF operator). Thin grey lines display shifts in earthquake locations towards the centre of the BGF due to the application of DD relocation technique. (Kwiatek et al., 2013)
The techniques described above to characterize IS and geomechanical processes at the two geothermal reservoirs represent selected case studies. These methods, however, can be applied to any data set of IS with a reasonably good quality of the recorded waveforms obtained from a sufficient monitoring network.
WP4
The aim of the development of a geothermal site by hydraulic and/or chemical stimulation is to irreversibly increase the injectivity/productivity of a well by a local increase of the transmissivity of the fractures and/or faults intersecting the well and an improvement of the connectivity of the well to fracture and/or fault network. The generation of microseismicity is inherent in these methods of development of wells. However, if a large number of microearthquakes is inevitable, major events (magnitude between 2 and 3) must be studied carefully. Understanding these events, may lead to propose scenarios for the stimulation reducing or eliminating them while maintaining the effectiveness of the stimulation relative to the injectivity. The WP4 work package is dedicated to understanding the geomechanical causes and processes of induced seismicity in various contexts and at various scales. This work was carried out in three sub-tasks : sub-task 4.1 related to the role of pore pressure changes; sub-task 4.2 related to the role of temperature changes; and sub-task 4.3 related to the role of faults and fractures. The following provides an overview of the S&T results obtained within WP4.
Laboratory experiments (GFZ) coupled with acoustic emission (AE) monitoring during fluid injection have been conducted in dry and partially saturated sandstone specimens, with a pre-existing laboratory induced fracture plane. The experiments show that the re-activation of a fracture through fluid pressure perturbations into saturated specimen is characterized by fewer high amplitude events than into dry specimens. Furthermore, the reactivation occurred via collapse and shear cracks in places of the pre-existing fracture plane and shear and tensile cracks in regions far from it. Other laboratory experiments have also been conducted with specimens which had a predefined saw-cut plane oriented at different angles to the direction of the major imposed principal stress (Figure 1). Stick-slip test have been conducted, and the results show that slip along the saw-cut plane was accompanied by an increased number of AE events, all located in different patches along the saw-cut plane. New fractures, far from the saw-cut planes were not induced. Mohr-Coulomb envelopes reached sliding lines with different coefficients of static friction, which indicates a change in the roughness of the saw-cut planes (higher coefficients of static frictions, thus, more rough surfaces due to wearing). Shear cracks were dominant during the sliding, while the number of tensile cracks was not negligible. The latter were related to local pore pressure perturbations and local dilatancy taking place during the sliding
Figure 1: View of rock specimen with saw cut plane – Location of AE during pore pressure increase
Partner INGV was mainly involved in sub-task 4.1. Their contribution dealt with modeling of time-dependent Coulomb stress changes due to fluid injection and fluid withdrawal in wells by using a thermo-fluid dynamical approach in porous media with a coupling of the two codes THOUGH2 and COMSOL. The approach consists in a two-step procedure: in the first step, injection of withdrawal of water is simulated by TOUGH2, by computing pressure and temperature changes in the volume; in a second step, the incremental stress tensor is computed by COMSOL. According to the models, stress changes can attain considerable values, able to significantly re-orient the mechanisms of most stressed fractures, with respect to the orientation due to the regional stress field alone. Also, the results highlighted that the main causes of induced seismicity during stimulation are due to the Coulomb stress changes generated by water injection, rather than by the effect of the temperature of the injected fluid.
Changes of Coulomb stress due to induced seismicity during stimulation have also been analyzed by partner GEOWATT. For this purpose an efficient method has been developed to calculate coseismic stress changes from an elliptical slip distribution on a circular fracture using superposition of rectangular sources. The method has been applied on a dataset of 715 focal mechanisms derived from seismic recordings of the Soutz-sous-Forêts GPK2 2000 stimulation to calculate temporal evolution of change of Coulomb stress (Figure 2). It comes out that the distribution of stress changes at the hypocenters reveals no evidence for events to be triggered from previous induced seismicity. However, about 5% of the events occur in areas where the stress is increased by more than 1 MPa and are thus candidates for triggering by previous events.
Figure 2: 3D views of coseismic stress changes in the Soultz-sous-Forêts reservoir. The wells GPK2 (red), GPK3 (green) and GPK4 (blue) are displayed with bold open hole section. The microseismic events used for the computation are represented by sphere symbols which scale with the magnitude, not with the geometry
Partner TNO was involved in the three sub-tasks of WP4. TNO has shown within sub-task 4.1 that for creating an EGS in compartmentalized sedimentary reservoirs, both the effects of pressure changes in the rock matrix and the impact of large scale heterogeneities should be taken into account. It has been shown within sub-task 4.2 that for simple systems dominated be few large fractures, poro-elasticity processes were most likely to control the overall thermo-mechanical process, by increasing the shear strength resistance of the fracture and balancing local traction and fracture opening effects, thus preventing large shear failure to develop. Within sub-task 4.3 partner TNO developed a specific coupled continuous approach with FLAC3D in order to study the role of pore pressure changes causing fault zone reactivation and inducing seismicity during the EGS operations. A sensitivity analysis has been performed on key parameters such as the in-situ stress regime, the fracture or fault strength and fault frictional behavior after the onset of failure.Results of the FLAC3D model were compared to results of the Block-Spring model as proposed by Baisch. The models were also applied on the GPK3 stimulation case that was performed in Soultz-sous-Forêts in 2003 (Figure 3). The basic features of the observed seismicity were reproduced. The 3D physical approach and the Block-Spring model showed their specific advantages: the Block-Spring model, if calibrated, establishes a fast modeling tool for sensitivity analyses; the FLAC3D implementation allows better understanding as it is based on actual physics.
Figure 3: Stimulation of Soultz-sous-Forêts GPK3 well with 3D continuous approach – Distribution of pore pressure (a), fracture aperture (b) and permeability (c)
Partner ARMINES was also involved in the three sub-tasks of WP4. ARMINES studied with a Discrete Fractured Network approach the impact of the thermal effect during a long term flow circulation test based on a forward thermo-hydro-mechanical modeling of the in-situ observations made at Rosemanoves site (UK). It has been shown that when injection pressure is maintained at constant level over time, shear can be re-activated at places where a previous rupture occurred, as soon as thermal tractions accumulated with time reach a magnitude similar to the stress drops generated during reservoir stimulation. The delay of a new event to occur is controlled by heat diffusion phenomenon in the rock blocks adjacent to the fracture. Note that calculated seismic moments due to the long term cooling of the rock are one order of magnitude lower, but that a small increase in injection pressure can precipitate new seismic activity as soon as some thermal stress has been stored. It comes out that seismic activity during the life cycle of a reservoir is not independent of pre-stimulation phases during initial reservoir development. This is of prior interest for mitigation strategies. The role of fractures heterogeneities has also been undertaken by partner ARMINES by studying different reservoir configurations: a 3D multi fracture system and a single fault segment with heterogeneous properties under a given stress regime (Figure 4). According to the models, the development of shear is controlled by pressure diffusion, and therefore in both cases hydraulic properties, contrasts in properties of the various structures and boundary conditions are governing the mechanical process. The critical conditions with regard to the occurrence of large induced seismic events are obtained in the situation of some large fracture with unknown hydraulic properties exists at an intermediate distance of the well, and when the ‘bulk’ hydraulic diffusivity of the massif is limited. In the case this structure is also of poor natural hydraulic diffusivity, or locked by other fault segments, the risk of inducing there a significant event is higher because it can be pressurized over a large region. Therefore tracking by geophysical methods any of these structural surfaces of potential failure at distances up to some hundreds of meters from a well is crucial. Testing them separately, if cross cut by a well, even with short term hydraulic tests would produce most valuable data, that would be helpful to design the most reasonable stimulation scenario, with the constraint of a pre-identified admissible pressure in the fault. This recommendation should form part of the mitigation strategy. Drilling capabilities allows now the possibility to develop deviated wells, or wells with multiple branches in smaller diameters. Targeting such geological structures should be evaluated in case VSP had identified them for instance. Although not affordable at first glance, these investments would anyhow participate to the long-term survey of any EGS production field at industrial scale and have the potential of allowing re-entry in the future for further injection or production developments. It is therefore suggested that the deepest section of a well in a new site should be first used at survey and not be automatically designed to be included in the final system.
Figure 4: Left: Prescribed step-rate injection test and calculated hydraulic head at injection well. Right, corresponding shearing events occurring in time with their calculated moments in blue, with hydraulic head (green area) in the fault when shear starts to develop into it, after about 3 days of injection. No events are simulated during the late injection phase
Partner BRGM was involved in sub-task 4.3 related to the role of faults and fractures. BRGM focused its study on the coupled hydro-mechanical response of a fault network intersecting a deep geothermal well to a hydraulic stimulation. The stimulation zone around Soultz-sous-Forêts deep wells was modelled using a 3D Distinct Element Method. According to the stimulation phase and post-stimulation phase carried out here, the models reproduce the evolution of the Overpressure-Flowrate curves observed in-situ, even if the total increase of injectivity has not been reached yet. Contrary to what is traditionally done, looking at the orientation of the most favourably oriented fault zone in respect to the principal stress does not allow determining the most efficient fault network. On the basis on the study performed, the three kinds of behaviour can be met (Figure 5):
• Direct stimulation due to shearing in fault segment favourably oriented with respect to σH. But, all the favourably oriented with respect to σH are not stimulated;
• Little/no stimulation, even if favourably oriented with respect to σH;
• “Indirect stimulation” of fault segment, not favourably oriented with respect to σH, due to the connectivity in 3D of the fault network.
The adaption of a joint model, that takes into account progressive damage during shear, has allowed discriminating the coseismic part of the shear from the aseismic one and thus better estimating the event magnitude. It results that:
• In the directly stimulated fault segments, the maximum seismic moments reach a plateau (M < 2) due to the repartition of the shear in several distinct seismic areas, clearly lower than the maximum seismic moments calculated with the standard Mohr-Coulomb model used so far (M around 3);
• In the indirectly stimulated fault segments, there is barely any shear displacement at the origin of major seismic events.
Finally, with the developments undertaken by BRGM, tools are available to study the occurrence of potential major induced seismic events on the same mechanical basis as the evaluation of the efficiency of the hydraulic stimulation of a deep well. Moreover, from the point of view of the knowledge of EGS well, the results obtained show the importance of taking into account the 3D reality of the fault network around the well, for the efficiency of the stimulation as well as for the estimation of the magnitude of the major seismic events. In order to lead stimulation scenarios, limiting the maximum magnitude of seismic events, an improvement is still needed in choosing the parameters of the hydro-mechanical behaviour to reach the final value of the well injectivity.
Figure 5: Stimulation of Soultz-sous-Forêts GPK4 well 3D Distinct Element approach - Location of the nodes reaching the rupture, superimposed to the shear displacement for a fault segment directly stimulated by shearing (left) and another fault segment indirectly stimulated (right)
The implementation of the effect of fluid injection into a pre-existing Boundary Integral Element Method code modeling dynamic rupture has also been done by partner BRGM in order to build, based on physical parameters, catalogues of synthetic events (location, time, magnitude) that are necessary to evaluate the seismic hazard (WP5). With this approach, it is assumed that the evolution of fluid migration and induced seismicity takes place along a single planar fault, described as a finite permeable zone of variable width. A sensitivity analysis has been performed in order to study the parameter interdependencies. It turns out that seismicity and fluid migration are strongly influenced by the injection rate and the fault rheology heterogeneity. The simulated seismicity generally tends to rapidly evolve after triggering, independently of the injection history. This self-induced seismicity takes place in the case where shear rupturing on the planar fault becomes dominant over the fluid migration process. On the contrary, if healing processes take place in the fault, so that the fluid mass becomes trapped within the fault, rupturing occurs continuously during the injection period.
To conclude, laboratory experiments and several modeling activities with complementary approaches have been addressed throughout WP4. Different numerical tools have been developed, or adapted, in order to study the coupled mechanisms that induce seismicity during EGS operations. The efforts put in WP4 also include the development of tools (partners ARMINES, BRGM, TNO) to simulate catalogs of synthetic events, based on rock mass physical features and on injection rates, that are useful in order to assess the seismic hazard associated to EGS operations. A considerable progress has been therefore made in GEISER.
WP5: Seismic Hazard Assessment
The objectives of WP5 of the GEISER project were to (i) develop a comprehensive framework to assess the earthquake hazard associated with natural seismicity and to the seismicity induced and triggered during and after EGS operations, (ii) develop statistical methodologies to assess the stability or increase in seismic hazard associated with EGS operations and (iii) test modelling tools of incremental accuracy to assess shaking and where possible damage which could be produced by EGS induced and triggered seismicity.
This work led to the publication of 15 peer reviewed articles by the different groups involved in WP5 (Douglas and Jousset; 2011; Goertz-Allmann et al., 2011; Bachmann et al., 2012; Convertito et al., 2012; Douglas et al., 2013; Gischig and Wiemer, 2013; Aochi et al., 2012; Goertz-Allmann and Wiemer, 2013; Mena et al., 2013; Hakimhashemi et al., 2013; Douglas, 2013; Catalli et al., 2013; Edwards and Douglas, 2013a, b; Mignan et al., in revision; Schoenball et al., 2012), plus six GEISER deliverables. Several additional publications are work in progress.
The main results of WP5 are threefold:
1) General probabilistic hazard and risk assessment framework (D5.1 2, 4, 5)
Figure 1 summarizes the main characteristics of the proposed approach, in the form of a logic tree. It was first shown that the three types of seismicity (natural, induced and triggered) could be treated in a similar manner at the hazard and risk level, hence the proposal of a common framework. This framework (Mignan et al., in revision) combines state-of-the-art time-dependent earthquake forecasting methods (probabilistic part) with standard seismic hazard assessment (so-called hazard curves) and standard risk assessment (here the RISK-UE macroseismic method). This is the first time that such an approach has been applied to EGSs, based on lessons learned from a long tradition of tectonic event analysis.
The three types of seismicity behaviour have some unique specificities, which do not impact the structure of the framework but yield different input parameters to be taken into account. In particular, induced seismicity differs from natural seismicity in the fact that the rate of occurrence is correlated to the volume of injected fluids (e.g. Shapiro-type or modified Epidemic-Type Aftershock Sequence ETAS models - light blue in Fig. 1). Distinction between induced and triggered seismicity can be made based on the maximum magnitude Mmax parameter (dark blue in Fig. 1), the size of an induced event being controlled by the size of the reservoir while the size of a triggered event is controlled by the maturity of the existing fault network. Another characteristic of induced seismicity, compared to tectonic events, is the range of magnitudes considered. Induced seismicity can cause nuisance for magnitudes as low as m = 2 while standard hazard and risk modelling is defined for m = 5+ (damaging events). It means that extrapolation of hazard and risk results from m = 5+ to 2 ≤ m ≤ 4 inevitably leads to biases. This has been corrected at the hazard level with the definition of ground motion prediction equations (GMPEs) for induced seismicity data sets (Douglas et al., 2013; Edwards and Douglas; 2013; see inset in Fig. 1) and with the calibration of vulnerability curves to avoid overestimating losses (red in Fig. 1) due to induced events (based on the SERIANEX study).
The logic tree approach captures all of these specificities, but more generally all epistemic uncertainties to be considered for hazard and risk assessment in the EGS context. It also provides a simple way to incorporate results of statistical methodologies to assess changes in hazard before, during and after injections of fluids. In particular the likelihood method was used to compare different time-dependent forecasts, which gives the weights to be included in the logic tree (Mena et al., 2013). Finally, the logic tree can be updated following the testing of modelling tools of incremental accuracy to assess shaking (and damage). In particular, the 36 GMPEs proposed for induced seismicity represent a large range of possibilities, which can be refined if more data becomes available (Douglas et al., 2013). In Figure 1 (in yellow/orange), weights are based on improved knowledge on GMPE parameters for the case of Switzerland. If GMPEs are not available, one could instead use Intensity Prediction Equations (IPEs), based on global data or on regional data (e.g. Swiss IPEs in Fig. 1).
Figure 1: Logic tree for induced seismicity probabilistic hazard and risk assessment. Ground Motion Prediction Equations (GMPEs) are the ones proposed by Douglas et al. (2013). Weights are the same for most branches except for induced-seismicity based GMPEs for which weights are fixed for Switzerland. IPE refers to Intensity Prediction Equation and GMICE to Ground Motion-Intensity Conversion Equations. Source: Mignan et al. (in revision).
2) Hazard models for specific test areas (D5.3)
Within the scope of WP5, a series of hazard models have been described, based on different forecast models and hazard metrics and tested on various EGS induced seismicity sequences. Sequences include: Basel (Switzerland), Rosemanowes (United Kingdom), Soultz-sous-Forêts (France) and The Geisers (USA). Statistical forecast include: Gutenberg-Richter law, ETAS model modified with a flow-induced background rate and the Shapiro-type model. Physical models include: Geomechanical models, non-linear flow model and static stress transfer model. Hazard metrics considered are: event magnitude, European Macroseismic Scale (EMS) intensity and peak ground acceleration (PGA). Here is a synthesis of these results, which is also used as input for the development of guidelines for best practice (see next section).
The two types of proposed forecast models, statistical and physical, are complementary. At the present time, statistical models appear as a reasonable choice to forecast induced seismicity in a prospective way for decision support (e.g. traffic light systems). It has been shown that they fit the data well and that observed variations between the best models (Shapiro-type and ETAS modified - used in the logic tree of Fig. 1) have a low impact on the overall uncertainty (Mignan et al., in revision). Physical models can also well reproduce observed induced seismicity sequences (e.g. Gischig and Wiemer, 2013). While statistical models may outperform physical models due to a lower number of parameters and a faster computation, physical models are crucial for a better understanding of the evolution of induced seismicity over longer time horizons, of b-value changes, or of Mmax (to only cite a few). However at the present time, there is no consensus on which physical model best describes induced seismicity.
Groups affected by an EGS project are public authorities, local residents, the operator company and insurances. Depending on the group, a different metric may be required. The simplest metric is to use the probability of exceeding a given magnitude m (Bachmann et al., 2012; Goertz-Allmann and Wiemer, 2013; Mignan et al., in revision). Adding more information, such as Mmax and ground motion properties, while adding more parameters and thus more uncertainties, provides a better communication tool. The probability of exceeding a given intensity measure (e.g. EMS based on an IPE) will inform local residents and public authorities as to what events are likely to be felt or damaging (Gischig and Wiemer, 2013; Mena et al., 2013; Mignan et al., in revision). The probability of exceeding a given PGA value (Convertito et al., 2012), being based on an instrumental measure, can be useful to the EGS operator to follow specific thresholds.
3) Recommendations for induced seismicity hazard and risk management (D5.6)
Based on the innovative hazard and risk assessment framework described above, it was possible to provide some recommendations to stakeholders. This is described in the section on Task 5.5 below. Figure 2 shows a schematic example of a advanced traffic-light system.
Figure 2: Advanced Traffic Light System: Schematic overview of the foreseen software framework. Legend: W = weighting; GMPE = ground motion prediction equation; EGF = empirical green’s function; PSHA = probabilistic seismic hazard assessment.
For such a comprehensive hazard and risk assessment framework we provide the following guidelines:
(1) Use a deterministic risk approach prior to injection to estimate roughly the expected losses (some hazard metrics could also be used) in function of induced event expected magnitude;
(2) Use a probabilistic risk approach during injection by combining deterministic results to time-dependent induced seismicity forecasts;
(3) Distinct output metrics to communicate the risk to different interest groups (public authorities/regulators, operators, public, insurances);
(4) Include epistemic uncertainty using a logic-tree approach as well as realistic aleatory uncertainty (e.g. the seismic hazard "sigma") for an objective estimation of hazard and risk associated to EGS activities;
(5) Define a clear traffic-light system following well-established methodologies and regulatory standards.
4) Achievements by Tasks
Below more detailed outcomes of WP5 are given, broken down by each task:
Task 5.1: Assessment of seismic hazard associated to natural seismicity
A good definition of the background hazard and the probability of natural earthquakes is needed to assess the probability that an earthquake that occurred in the vicinity of an EGS site is causally related to the activity. This judgment may have substantial legal and financial consequences (in terms of insurance and liability); it also requires local and regional monitoring capable of resolving hypocenters with sufficient accuracy. Hazard assessment is a core responsibility of national authorities, and all nations in Europe maintain and update a probabilistic seismic hazard model. While the degree of complexity and sophistication varies, the basic methodology of probabilistic seismic hazard assessment (PSHA) is well established and accepted. We evaluated the available models for their suitability for EGS relevant estimation of the background hazard at a European level. We concluded that the SHARE seismic hazard model, to be published in 2013 as part of the EC FP7 project on Seismic Hazard Harmonization (www.share-eu.org/) is the ideal reference framework.
Task 5.2: Assessment of seismic hazard associated to EGS induced seismicity
We developed a probabilistic hazard assessment method for induced seismicity observed in EGSs. We first generated forecasts based on induced seismicity models available in the literature. Forecasts were then included in time-dependent seismic hazard assessment and combined to ground motion intensity prediction equations (IPEs) to compute induced seismicity hazard curves. Epistemic uncertainties due to the choice of the forecast model and IPE were captured by using a logic tree approach. The weight of the different forecast models was determined using the likelihood method on observed induced microseismicity (m < 4).
Task 5.3: Assessment of seismic hazard associated to EGS triggered seismicity
Triggered seismicity, in contrast to induced seismicity, corresponds to events that occur on existing faults, i.e. not produced during stimulation, but which are advanced in their seismic cycle due to increased stress during stimulation. The main criterion for the assessment of EGS triggered seismicity hazard is the maximum magnitude Mmax. We first presented a statistical approach, which investigates the role of Mmax in seismic hazard assessment. In that view, Mmax in the EGS regime can be the same Mmax considered in the tectonic regime (e.g. Mmax = 7.0) thus making no distinction between induced and triggered seismicity. The probability of occurrence of such event is very low but is a non-zero value (conservative probabilistic view). We then presented a physical approach, in which Mmax is estimated in a deterministic way. Distinction between induced and triggered seismicity as well as the debate on Mmax values remain open questions, as no consensus seems to exist. We concluded that the statistical approach should be preferred at the present state of knowledge, using the tectonic Mmax to remain conservative. However mapping of existing faults around EGS sites should help refining Mmax values, which has a significant role in hazard assessment compared to other input parameters for high hazard intensities.
Task 5.4: Shaking and damage scenarios from EGS induced and triggered events
Various ground-motion datasets of induced and natural seismicity (from Basel, Geysers, Hengill, Roswinkel, Soultz, and Voerendaal) were compiled and processed, and moment magnitudes for all events were recomputed homogeneously. These data were used to show that ground motions from induced and natural earthquakes cannot be statistically distinguished. Empirical GMPEs were derived from these data; and, although they have similar characteristics to recent GMPEs for natural and mining-related seismicity, the standard deviations are higher. To account for epistemic uncertainties, stochastic models subsequently were developed based on a single corner frequency and with parameters constrained by the available data. Predicted ground motions from these models were fitted with functional forms to obtain easy-to-use GMPEs. These are associated with standard deviations derived from the empirical data to characterize aleatory variability. As an example, we demonstrated the potential use of these models using data from Campi Flegrei.
Task 5.5: Guidelines for best practice in seismic hazard assessment for site selection and licensing
If geothermal energy from EGSs is to become a significant component in future energy policy, there is an urgent need to provide guidelines to help regulators in devising seismic hazard assessment specifications for the selection, licensing and long-term operation of EGS sites in different geological settings. Here we built upon previous results of WP5 on the assessment of natural, induced and triggered seismic hazard to develop a comprehensive framework for best practice. We provided the following guidelines:
(1) Use a deterministic risk approach prior to injection to estimate roughly the expected losses (some hazard metrics could also be used) in function of induced event expected magnitude;
(2) Use a probabilistic risk approach during injection by combining deterministic results to time-dependent induced seismicity forecasts;
(3) Include epistemic uncertainty using a logic-tree approach as well as realistic aleatory uncertainty (e.g. the seismic hazard "sigma") for an objective estimation of hazard and risk associated to EGS activities;
(4) Define sensible thresholds for the various stakeholders and implement these thresholds as part of an advanced traffic light systems, which is forward looking, validated abd dynamically updated on the fly as new data arrives.
The results of WP5 build upon work completed in WP3 and WP4, and are an integral input to WP6 and discussed there also. With respect to managing induced seismicity, we believe it is fair to say that GEISER WP5 has made substantial progress: We now have a framework for hazard and risk assessment that is accepted by a larger community and serves the need of a wide range of stakeholder. This framework will have to be expanded and validated in future pilot and demonstration projects.
WP6: Strategies for EGS operations with induced seismicity
WP6, -in close cooperation with WP5- provides guidelines for safe and reliable EGS Operations.
Task 6.1 soft stimulation investigated the influence of injection parameters to reservoir seismicity due to the stimulation treatment, reviewing past permeability enhancement experiences and drawing idealized scenarios.
At the moment there is no 3-dimensional fracture simulator simultaneously covering the reservoir development and the behaviour of the surrounding rock mass before, during, and after the stimulation treatment. Some of the capabilities that a complete tool to model the fracture creation/stimulation should possess have been addressed; a complete 3-dimensional simulator would prove useful in real-time operation to understand field data collected real-time, since new data allows refining and better definition of the stimulation treatment, not excluding the definition of a traffic light system capable of forecast based on injection parameters and not only reacting on recorded seismicity.
A combination of different numerical tools has been employed to define the response of the reservoir to different injection parameters (temperature, pressure, fluid volume variation in time and space), integrating results obtained in WP4.
ARMINES extended the outcome of the work done in WP 4.4 investigating the effect of injected fluid temperature and thermal stress build up due to stimulation treatment. The numerical model developed was validated against the Rosemanowes (UK) test site data. A significant result comes out from the modelling exercise: seismic activity during the life cycle of a reservoir is not independent of pre-stimulation phases during initial reservoir development. This is of prior interest for mitigation strategies, since the thermal stress magnitude depends on temperature and volume of injected fluid.
GEOWATT AG contributed to subtask 6.1 by evaluating the role of existing (or expected) large scale faults in proximity of the engineered reservoir. Faults with different orientations, but with the same mechanical properties have been investigated. An important outcome of the study is that the highest danger of large magnitude events might not come from fractures that are optimally oriented to fail, because of the anisotropy of the pressure front and stress field heterogeneities in the model. This result is in agreement with field observations in Soultz (FR) and in Basel (CH).
GFZ investigated the reservoir response in term of pore pressure and stress change due to a tensile fracture opening, relating the fracture volume with the influenced reservoir volume. Some of the characteristics that a 3-dimensional fracture simulator should possess have been addressed. Stress shadow effects from the fracture and delayed thermal stress effect have been evaluated and compared with observations from Groß Schönebeck (DE) geothermal test site.
Stimulation treatment design is a critical process in engineering a geothermal system, however the knowledge of the target reservoir before completing the actual field operations is quite limited. Therefore, predicting the seismic behaviour of the reservoir can be difficult. The assumption behind the designed stimulation treatment may have to be updated on the fly, during the field operations, especially with respect to fault orientation and mechanic properties.
It is possible, however, to perform sensitivity analysis on the relevant mechanical and hydraulical parameters of the structure that may be reactivated. A range of stimulation scenario can then be defined and eventually integrated into a traffic light system based on injection parameters, to provide a more robust forecast capability, with respect to a system based only on recorded microseismicity.
The results of Task 6.1 have been included in deliverable D6.1
Task 6.2 seismic monitoring developed guidelines for seismic monitoring at EGS projects. The goal is mitigation of induced seismicity, in particular the prevention of large magnitude events (LME), i.e. seismic events that may cause nuisance and/or damage to local population, industry and/or infrastructure. The guidelines as delivered in deliverable D6.2 are aimed at both regulators and operators and address two import aspects of seismic monitoring: (1) monitoring requirements, defined in terms of data quality, temporal/spatial coverage, resolution, etc., and (2) monitoring network design and optimization, which is presented in a number of case studies.
For the monitoring requirements (1) we choose to follow the recent recommendations by the FKPE (Forschungskollegiums Physik des Erdkörpers) working group on “Induced seismicity” for seismic monitoring in the general context of geotechnical facilities. The FKPE recommendation is intended for reliable observation of earthquakes one magnitude unit below the human perception threshold. We adopt the FKPE recommendations for what we refer to as (a) basic seismic monitoring. However, the FKPE explicitly notes that their recommendation is not sufficient when the monitoring goal is to gain insight into the seismogenic processes. In the context of EGS and mitigation of induced seismicity the understanding of the seismogenic processes is of critical importance. We therefore extend the FKPE recommendations down towards lower magnitudes for what we refer to as (b) reservoir seismic monitoring. In our requirements we identify three main themes that are developed for both basic (a) and reservoir (b) seismic monitoring: (i) data quality: defined in terms of signal-to-noise ratios, frequency content and time-stamping accuracy; (ii) spatio-temporal coverage: defining both spatial and temporal intervals for the monitoring efforts; and (iii) data policy: where a general recommendation is to be as open as possible.
For the monitoring network design and optimization (2), we note that monitoring at least requires a network of continuously operating, time-synchronized seismic sensors. The network should be designed in such a way that it satisfies the monitoring requirements discussed above. In a surface network the sensitivity of a usually goes down to magnitudes around M=1 (at typical EGS reservoir depths), which is roughly one to two magnitude units below the perception threshold (i.e. the level that can be felt by humans at the surface). To extend the sensitivity of the network further down the magnitude scale it is usually necessary to install sensors in (deep) boreholes. To optimize the network design various approaches are possible. With (partial) contribution of GEISER a number of case studies on network design and optimization have been performed and included in deliverable D6.2. These studies are provided as illustrations of proposed techniques for the design and optimization of seismic networks.
Task 6.3 Real-time tools to monitor the evolution of induced microseismicity and Task 6.5 Provide boundary conditions for regulatory guidelines have built input and boundary conditions for regulatory guidelines included in deliverable D6.3
Activities in these tasks focussed on building a shared conceptual framework for this deliverable in close feedback with other Work package activities providing input. To this end a number of workshops have been organised between WP 5 and WP 6, including a workshop in Utrecht in March 2011, one in Zürich in November 2011 and various work meetings. From a regulatory perspective the boundary conditions need to be closely connected to conditions and evaluation criteria for licensing of activities. As an outcome of joint work package meetings during the general assembly in 2011 it became apparent that we needed to further synergize the insights of WP 3 and WP4 into WP 5 and WP6. To this end in Q1 2012 a joint workshop between WP3,4,5,6 has been devoted to the outline of key parameters which control induced seismicity and to discuss strategies how a priori assessment, monitoring and validations of these can be incorporated in a regulatory framework and technical guidelines. As an outcome of the Q1 2012 workshop it has been decided to organize a follow up workshop Q3 2012 for the technical guidelines serving as input for regulatory quidelines.
These technical guidelines have been described in detail in the following deliverables of Geiser:.
1) Input Guidelines for best practice for seismic hazard assessment (D5.6)
2) effect of different stimulation techniques on the seismicity and strategies to mitigate induced seismicity (D6.1)
3) Technical best practices for monitoring (D6.2)
In addition a separate deliverable of Geiser deals with Socio-economic best practices(D6.4) , which is also of great relevance to providing regulatory guidelines
Activities in these tasks dealt largely with fitting the fore mentioned technical and socio-economic guidelines in existing regulatory frameworks (France, Germany, Iceland, Italy, Netherlands, Switzerland). To this end we introduced a conceptual development framework for EGS, identifying main toll-gates for regulatory guidelines. Next we outlined the main features of various national mining laws and its practical regulatory implementation. For each of these an outline is given how the law and regulatory framework are or could be applied to EGS development.
The following recommendations for European regulatory guidelines have been included in Deliverable 6.3 to prevent unsolicited effects of induced seismicity:
1. to set a maximum level of acceptable magnitude (Mtol), and associated treshold probability (Ptol) for earthquakes caused by stimulation and production. The project should be halted if the expected probability of Mtol is higher than Ptol and no operational adjustment can lower this probability.
2. to request the project developer for an assessment of expected probability of Mtol prior to stimulation and to set-up an advanced traffic light system to safeguard that the expected probability remains lower than Ptol, based on a physics based approach (see deliverable 5.6 of GEISER).
3. to request the project developer to set-up a seismic monitoring network which follows GEISER recommendations (deliverable 6.2)
4. to provide incentives for project strategies promoting public acceptance (cf deliverable 6.4 of GEISER). One such incentive can be a differentiation in exploration licensing, allowing for a desk study phase, prior to the drilling and stimulation license. The desk study phase allows to build an outreach program.
Task 6.4 Public awareness and sensitivity to ground shaking due to injection and/or during EGS operations
In 2011 the activities within task 6.4 have been focused on the development of a strategy for creating public acceptance of EGS-projects. First, we did a literature review, based on empirical literature about current/past EGS-projects and theoretical literature about public acceptance. Next, we developed a draft strategy for creating public acceptance for EGS-projects. The results of the research activities of task 6.4 in 2011 have been collected in a progress report task 6.4. The research activities in In 2012 and 2013 focused on applying the strategy for creating public acceptance of EGS-projects on two case studies: the Deep Heat Mining project in Basel (Switserland) and the EGS research project in Soultz-Sous-Forêts (France). we would like to validate the draft strategy for creating public acceptance for EGS-projects. The results of the literature study in 2011 and the analysis of two case studies have been descripted in the final report D6.4 “Laying the Groundwork for Public Acceptance of Enhanced Geothermal Systems”, which also includes recommendations for creating public acceptance of future EGS-projects.
The proposed strategy for creating public acceptance entails the following steps:
• Preparation and context analysis: To identify different interests and (perceived) risks regarding a specific EGS project. This entails a cost-benefit balance for the stakeholders throughout the entire exploration and production workflow. This balance requires to take into account both technical-spatial and social-economic aspects in order to define the project strategies for creating public acceptance of EGS at a specific location (Figure 9.4).
• Process design: The different interests and (perceived) risks allows to characterize the policy challenge(Figure 9.4). The policy challenge needs to be proper addressed through the project strategy, including a communication strategy and process definition of involvement of actors and associated actions.
• the execution of the preparation, context analysis and process design are recommended to overlap with the common planning phase of EGS projects
• Implementation and evaluation: The next phases of developing an EGS project (drilling, logging and testing; stimulation; operation; post-operation) should correlate to the implementation phase of the project strategy. During the implementation of the project strategy (cq. stages of development of the EGS-project) the progress of the process will be constantly monitored and evaluated. If needed, the project strategy will be changed and adopted to the process dynamics.
References
Aochi, H., B. Poisson, R. Toussaint, X. Rachez, and J. Schmittbuhl (2012), Self-induced seismicity due to fluid circulation along faults, Geophysical Journal International, in press.
Bachmann C.E Wiemer S., Goertz-Allmann B.P. & Woessner J., 2012, Influence of pore-pressure on the event-size distribution of induced earthquakes, Geophysical Research Letters 39, L09302, doi:10.1029/2012GL051480.
Bethmann F., Deichmann N., & Mai M., 2011, Scaling relations of local magnitude vs. moment magnitude for sequences of similar earthquakes in Switzerland, Bulletin of the Seismological Society of America 101(2), 515-534, doi:10.1785/0120100179.
Calò M., Dorbath C., Luzio D., Rotolo S.G. & D’Anna G., 2009, Local Earthquakes Tomography in the southern Tyrrhenian region: geophysical and petrological inferences on subducting lithosphere, In: Subduction Zone Geodynamics, Frontiers in Earth Sciences, S. Lallemand and F. Funiciello (eds.), Springer-Verlag Berlin Heidelberg, 85-99, doi: 10.1007/978-3-540-87974-9.
Calò M., Dorbath C., Cornet F.H. & Cuenot N., 2011, Large-scale aseismic motion identified through 4-D P-wave tomography, Geophysical Journal International 186, 1295-1314, doi: 10.1111/j.1365-246X.2011.05108.x.
Calò M., & Dorbath C., 2013a, Different behaviours of the seismic velocity field at Soultz-sous-Forêts revealed by 4D seismic tomography: case study of GPK3 and GPK2 injection tests, Geophysical Journal International, doi: 10.1093/gji/ggt153.
Calò M., Kinnaert X., & Dorbath C., 2013b, Procedure to construct 3-D models of geothermal areas using seismic noise cross-correlations: application to the Soultz-sous-Forêts enhanced geothermal site, Geophysical Journal International, doi: 10.1093/gji/ggt205.
Catalli, F., Meier, M.A. and S. Wiemer (2013), The role of Coulomb stress changes for injection-induced seismicity: The Basel enhanced geothermal system, Geophys Res Lett, 40 (1), 72–77.
Deichmann N., & Giardini D., 2009, Earthquakes induced by the stimulation of an Enhanced Geothermal System below Basel (Switzerland), Seismological Research Letters 80(5), 784-798, doi:10.1785/gssrl.80.5.784.
Deichmann N., Kraft T., & Evans K.F. 2013, Identification of faults activated during the stimulation of the Basel geothermal project from cluster analysis and focal mechanisms of the larger magnitude events. Submitted to Geothermics Special Issue on Induced Seismicity.
Douglas, J. (2013), Seismic network design to detect felt ground motions from induced seismicity, Soil Dynamics and Earthquake Engineering, 48, 193-197, doi:10.1016/j.soildyn.2013.01.030.
Douglas, J. and P. Jousset (2011), Modeling the difference in ground-motion magnitudescaling in small and large earthquakes, Seismological Research Letters, 82(4), 504-508, doi: 10.1785/gssrl.82.4.504.
Douglas, J., B. Edwards, V. Convertito, N. Sharma, A. Tramelli, D. Kraaijpoel, B. M. Cabrera, N. Maercklin and C. Troise (2013), Predicting ground motion from induced earthquakes in geothermal areas, Bulletin of the Seismological Society of America, 103(3), in press, doi: 10.1785/0120120197.
Douglas, J. and H. Aochi (2013), Using estimated risk to develop exploitation strategies for Enhanced Geothermal Systems, Geophysical Journal International (?), in preparation.
Edwards B., and J. Douglas (2013), Magnitude scaling of induced earthquakes, Geothermics, submitted November 2012.
Edwards B. and J. Douglas (2013), Selecting ground-motion models developed for induced seismicity in geothermal areas, Bulletin of the Seismological Society of America (?), in preparation.
Deichmann N., & Ernst J., 2009, Earthquake focal mechanisms of the induced seismicity in 2006 and 2007 below Basel (Switzerland), Swiss Journal of Geosciences 102(3), 457-466, doi: 10.1007/s00015-009-1336-y.
Gischik, V., and S. Wiemer (2013) A stochastic model for induced seismicity based on non-linear pressure diffusion, Geophys J Int, 194, 1229-1249, DOI: 10.1093/gji/ggt164.
Goertz-Allmann B.P. Goertz A., & Wiemer S., 2011, Stress drop variations of induced earthquakes at the Basel geothermal site, Geophysical Research Letters 38, L09308, doi: 10.1029/2011GL047498.
Goertz-Allmann B.P. & Wiemer S., 2013, Geomechanical modeling of induced seismicity source parameters and implications for seismic hazard assessment, Geophysics 78(1), 25-39, doi: 10.1190/GEO2012-0102.1
Hakimhashemi, A. Yoon JS, Heidbach, O., Zang, A., Grünthal, G. (2013), FISHA - FORWARD INDUCED SEISMIC HAZARD ASSESSMENT APPLICATION TO SYNTHETIC SEISMICITY CATALOG GENERATED BY HYDRAULIC STIMULATION MODELING, Geothermics, in press.
Häring M.O. Schanz U., Ladner F., & Dyer B.C. 2008, Characterization of the Basel 1 Enhanced Geothermal System, Geothermics 37, 469-495, doi: 10.1016/j.geothermics.2008.06.002.
Kraft T., & Deichmann N., 2013, High-precision relocation and focal mechanism of the injection induced seismicity at the Basel EGS, Submitted to Geothermics.
Kwiatek G., Bohnhoff M., Martínez-Garzón P., Bulut F., & Dresen G., 2013,. High-resolution reservoir characterization using induced seismicity and state of the art waveform processing techniques. First Break 31(7), 81-88.
Martínez-Garzón P., Bohnhoff M., Kwiatek G., & Dresen G., 2013, Stress tensor changes related to fluid injection at The Geysers Geothermal Field, California, Geophysical Research Letters, in press, doi: 10.1002/grl.50438.
Mena, B., S. Wiemer, and C. Bachmann (2013), Building Robust Models to Forecast the Induced Seismicity Related to Geothermal Reservoir Enhancement, B Seismol Soc Am, 103(1), 383-393.
Mignan, A., D. Landwing, B. Mena and S. Wiemer (2013), Induced seismicity risk assessment for the 2006 Basel, Switzerland, Enhanced Geothermal System project: Role of parameter uncertainty on risk mitigation, Geothermics, under revision.
Oye V., & Roth M., 2003, Automated seismic event location for hydrocarbon reservoirs, Computers & Geosciences 29, 851-863.
Rowe C.A. Aster R.C. Borchers B., & Young C.J. 2002, An Automatic, Adaptive Algorithm for Refining Phase Picks in Large Seismic Data Sets, Bulletin of the Seismological Society of America 92, 1660-1674.
Schoenball, M, Baujard, C and Dorbath, L. (2012) The role of triggering by static stress transfer during geothermal reservoir stimulation, J. Geophys. Res., 117, DOI: 10.1029/2012JB009304
Waldhauser F., & Ellsworth W.L. 2000, A Double-Difference Earthquake Location Algorithm: Method and Application to the Northern Hayward Fault, California. Bulletin of the Seismological Society of America 90(6), 1353-1368.
Potential Impact:
The main impact expected from the project is the establishment of a procedure to realise the goals of enhancing geothermal systems with a reliable concept for the mitigation of induced seismicity. This concept will ensure that geothermal energy can reach its full efficiency and profitability thresholds at the European scale.
The guidelines provided by GEISER will help to reduce uncertainties with respect to effects of induced seismicity and how to mitigate them. As induced seismicity has been one of the major road blocks in the development of EGS, the results of this project will help to increase EGS market introduction by providing:
Licensing guidelines
Planning security
These guidelines not only address the measures to be taken to monitor, control and mitigate induced seismicity during injection in geothermal operations but also include recommendations on how to build trust in the population and increase public acceptance.
Regulatory guidelines handling operational hazards.
With the provision of regulatory guidelines handling operational hazards due to underground exploitation, authorities will have better defined regulations and legal guidelines to cope with both political and physical damage of induced seismicity. This will provide planning security for developers and communities and is likely to unleash activity in the geothermal energy market.
Optimal monitoring infrastructures of underground exploitations.
The GEISER project successfully compiled and categorized data sets of induced seismicity in geothermal projects that can now be accessed by other users. Optimized monitoring systems were proposed that can be used by developers. It should be noted that, although the monitoring methods are specifically designed for the development of EGS, but they can also be of great interest in hydrocarbon exploration, in the search for deep aquifers, in the planning of waste disposals and in other applications.
The success of the GEISER-project should have a kick-off effect on the development of geothermal resources in Europe but also it shall enhance the capability of European industry to compete on the world market of geothermal business. The project will represent a technological milestone in the development of a renewable, cost-competitive geothermal energy (electric power and heat) which makes it possible for Europe to achieve:
the improvement of the standard of living and safeguard of the environment, contributing to the minimisation of gas emissions, thereby reducing the greenhouse effect. Limiting the considerations to the actual geothermal energy utilisation all over the world, it is equivalent to the burning of 150 million barrels of oil per year. In Europe alone, every year geothermal electricity avoids discharges of 5 million tons of carbon dioxide, 46000 tons of sulphur dioxide, 18000 tons of nitrogen oxide and 25000 tons of particulate matter into the atmosphere compared to the same production from a typical coal-fired plant. These numbers will increase with the successful development of more geothermal power plants
the creation of new employment, involved in exploration, well drilling, plants design and construction and all related activities of the operation and maintenance of power plants. Oil and gas service companies, drilling companies as well as machinery manufacturers and engineering suppliers will all benefit from the development. All these actions, applied on a vast scale, would not only guarantee new jobs, but also create new professional positions such as dedicated service men and builders
the development of new technologies and skills, as an outcome of the project will improve the competitiveness of the and increase the capability of geothermal as well as service companies to open up to new markets both in geothermal and hydrocarbon exploration sectors in Europe and in the developing areas of the world. The project target is to deal with methodological development of advanced techniques applicable to mitigate induced seismicity. This innovative approach applied to geothermal energy provision is also relevant in the prospect of geothermal resources in Europe (e.g. Italy, Portugal, Greece, Iceland, France, Germany, Eastern Europe and Turkey) and in the development of other areas of the world (Central and South America, South East Asia, Africa).
The project successfully bridges the gap between the scientific development of advanced monitoring technologies and in the application of operational services. Moreover, geophysical methods and tools developed within the GEISER-project, will apply to waste water injection and other sources of mining induced seismicity. The GEISER-project will be an important contribution to focusing on the development of advanced geophysical technologies aimed at the improvement of the production of geothermal systems and at providing more reliable tools to reduce the mining risk of geothermal field development. The scientific know-how generated by GEISER represents a major contribution to the geological and petrophysical interpretation of seismic sounding carried out in many European countries and for the investigation of georesources.
List of Websites:
www.geiser-fp7.eu