- Home
- Conferences
- Conference Proceedings
- Conferences
EAGE GeoTech 2021 First EAGE Workshop on Induced Seismicity
- Conference date: March 1-5, 2021
- Location: Online
- Published: 01 March 2021
-
-
Monitoring and forecasting failure in laboratory using coda wave decorrelation
Authors A. Veltmeijer, M. Naderloo and A. BarnhoornSummaryGas extraction has caused pressure differences along the field, triggering earthquakes, which are causing a lot of damage and social unrest in the Groningen area. Predicting the degree of these stress changes, and as a result, the potential onset and exact location of failure and seismicity, is very challenging.
Therefore, developing good techniques that can monitor these changes is crucial for a better prediction and thus mitigation of failure and seismicity in the subsurface. Laboratory active acoustic-monitoring techniques are used to determine parameters that can forecast upcoming failure and seismicity.
We show the use of coda wave decorrelation as a monitoring tool using sandstones analogues for the Groningen reservoir. Failure of the rock sample is preceded by the formation of micro-fractures. These fractures change the scattering properties of acoustic waves. The decorrelation coefficient K, as the indicator of the amount of scattering and thus be used as precursor to failure. We show that by monitoring K we can forecast the upcoming failure of the rock samples in the laboratory.
-
-
-
Seismic monitoring network for induced seismicity in the Netherlands
Authors P. Kruiver, J. Spetzler, E. Ruigrok and B. DostSummaryThe seismic monitoring network in the Netherlands consists of a dense network of shallow borehole arrays (200 m depth) and surface accelerometers in the Groningen region and a sparser network in the rest of the country. The dense network in Groningen is targeted at induced seismicity due to gas extraction. Other sources of induced seismicity include mining activities like salt mining, subsurface gas storage and geothermal energy. The Dutch ambitions for geothermal energy are high. Hundreds of systems need to be installed in the coming decades. Monitoring of potential induced seismicity is one of the risk measures. The monitoring network is currently being assessed for detection of induced seismicity for a magnitude of completeness of ML = 1.5. This threshold is met for a large part of the country, except for the middle and southwestern part. The network will be upgraded to fill the gaps.
The densification of the network allows for lower uncertainty in hypocenter determination. In order to achieve this, a detailed 3D nationwide seismic velocity model and new location methods will be developed and applied. In addition, the borehole vertical arrays will be used more effectively in the seismic analysis.
-
-
-
EPOS-NL is the Dutch solid Earth science infrastructure for research on georesources and geohazards
SummaryThe European Plate Observing System – Netherlands (EPOS-NL) is the Dutch research infrastructure for solid Earth sciences. EPOS-NL is a cluster of large-scale geophysical facilities for research on georesources and geohazards. It is a partnership between Delft University of Technology (TU Delft), the Royal Netherlands Meteorological Institute (KNMI) and Utrecht University (UU) and is funded by NWO, as part of the national roadmap for large-scale research infrastructure. EPOS-NL facilities include 1) The Earth Simulation Lab at UU, 2) The Groningen gas field seismological network and the ORFEUS Data Centre at KNMI, 3) The deep geothermal (DAP-)well to be installed on the TU Delft campus, and 4) A distributed facility for multi-scale imaging and tomography (MINT) at UU and TU Delft. EPOS-NL aims to further develop the infrastructure for solid Earth scientific research. It also makes cutting-edge research facilities and data available to (inter)national researchers, aiming to address key geo-societal challenges, notably:
- Exploration for (renewable) geo-energy resources
- Storage of fuels, CO2 and wastewater in the sub-surface, and
- Hazards such as induced or natural earthquakes
Addressing these challenges requires a multi-physics, multi-scale approach, and open access to state-of-the-art research facilities and data. EPOS-NL contributes to addressing these needs.
-
-
-
Towards efficient probabilistic characterisation of induced seismic sources in the Groningen Gas field
Authors L.O.M. Masfara and K. WeemstraSummaryWe present a probabilistic scheme to invert for hypocenters and moment tensors (MTs) of induced seismic events in 2D heterogeneous media. Our scheme is based on a variant of the Hamiltonian Monte Carlo scheme that uses the linearization of a misfit function. This algorithm, however, requires a sufficiently accurate prior, which could be taken from earthquake catalogues. In our case, a detailed subsurface velocity model allows the hypocenter prior to be obtained via a first arrival based algorithm. Fixing the induced event’s location to this prior, the MT’s prior is subsequently obtained. Having all necessary priors, we then recover the posterior probability density of the induced event’s source characteristics, i.e., hypocenter and MT. By comparing the recovered posterior with the directly modelled events, we infer that our scheme is efficient and practical to invert for source characteristics of induced seismic events.
-
-
-
Using next-generation seismic monitoring and processing to help manage injection and extraction operations
By A. BaigSummaryThe management of induced seismicity during hydraulic stimulations and other injection or extraction operations benefits greatly from any information on geohazards or the state of stress in the subsurface. The ability for pore-pressure and poro-elastic stress perturbations to propagate beyond a modest buffer around the reservoir has underscored the need for such knowledge over a footprint wider than the pad. Often, the seismicity itself allows for such knowledge to be brought better into focus. Locating events with precision will illuminate fault geometries that are often unobserved though other means. By recording with a sufficiently dense network, enough lower magnitude events will be observed to maximize the opportunities to image these faults. For larger events recorded across a sufficiently dense network, the first motions of the P and S waves can constrain the moment tensors, allowing for possible fault geometries to be explored, the faulting regime to be understood with greater nuance, and the stress and strains in and around the reservoir to be understood.
-
-
-
Next maximal event magnitude from statistical method
Authors N. Cao, L. Eisner and Z. JechumtalovaSummaryThis article shows the promising results of maximum magnitude event estimation, in the case of hydraulic injections, based on statistical methods used in mining industry.
-
-
-
Picking Automatization by Pattern Matching: A New Methodology Using EAT Events
Authors L. Duboeuf, V. Oye and I. BerreSummaryThe drastic increase of the observed seismicity at fluid injection sites has led to the need for further development of automatic picking methods for seismic events. Here, we focused on a pattern matching technique, which has traditionally relied on a representative Master event. However, this method strongly depends on the selection of the Master Event, which can be challenging with low Signal to Noise Ratios (SNR) or data gaps. In this study, we developed the Empirically Aggregated Template (EAT) concept. An EAT is a type of Master Event built using the seismic traces with the best SNR at each sensor, based on a set of events that exhibit high waveform similarity, and which are spatially close to one another. Thus, an EAT event will present the highest possible SNR of the database, and is not affected by data gaps. The efficiency and quality of this methodology has been tested by comparing with a set of manually picked events. The picking accuracy, location error, and travel-time residuals demonstrate similarities between manually and EAT picked events and allow us to validate this method. Finally, we have highlighted the advantages and improved results obtained using an EAT compared to a standard Master Event.
-
-
-
Keynote 5: Using Microseismicity to Manage Induced Seismicity During Hydraulic Fracturing at the Preston New Road PNR1 Well
Authors J. Verdon, T. Kettlety, A. Baird and M. KendallSummaryIn this case study, from the Preston New Road PNR-1 well in Lancashire, England, we show how high-quality microseismic data can be used to manage induced seismicity in real time. A combination of microseismic observations were used to characterise the interplay between hydraulic fractures and a pre-existing fault. We then used the observed event population to make forecasts of the largest expected event magnitude. We show how observations of this type can be used to guide operational decision-making during operations.
-
-
-
Systematic Recovery of Instrumental Timing Errors Using Interferometric Surface Waves Retrieved from Large N Seismic Arrays
Authors C. Weemstra, J. De Laat and A. VerdelSummaryTiming errors are a notorious problem in seismic data acquisition and processing. A technique is presented that allows such timing errors to be recovered in a systematic fashion. The methodology relies on virtual- source responses retrieved through the application of seismic interferometry (SI). In application to recordings of ambient seismic noise, SI involves temporal averaging of time-windowed crosscorrelation measurements. The retrieved interferometric responses are typically dominated by surface waves. Under favorable conditions, these interferometric responses therefore approach the surface-wave part of the medium’s Green’s function. Additionally, however, its time-reverse is often also retrieved. This implies time-symmetry of the time-averaged receiver-receiver crosscorrelations. In this study, this time-symmetry is exploited: by comparing the arrival time of the direct surface wave at positive time to the arrival time of the direct surface wave at negative time for a large a number of receiver- receiver pairs, relative timing errors can be determined in a weighted least-squared sense. The proposed methodology is validated using synthetic data. The results hold particular promise for large N seismic arrays.
-
-
-
Identification of Tectonic Structures Based On Tremors in Bobrek Coal Mine Using A Collapsing Method
Authors A. Lesniak, E. Weglinska and K. MirekSummaryIn Bobrek hard coal mine located in the area of Upper Silesia Coal Basin, two types of tremors can be distinguished. The first type is seismic shocks occurring in the vicinity of longwall mining fronts, which are the result of mining works. The second type of tremors is related to the location of renewed seismogenic structures. The article presents a collection of tremors recorded during the exploitation of the longwall 3 in a seam No. 503 from April 2009 to July 2010. An attempt was made to relocate tremors to separate clusters that could correspond to microtectonic structures. In order to group events into more spatially ordered structures, the collapsing method was used. The results of relocation after the collapsing procedure were shown on the whole data set. The linear structures responsible for the generation of shocks related to the progress of the mining front together with those indicating tectonic microstructures activated as a result of mining operations were presented.
-
-
-
Keynote: Corkscrew Well Paths for Improved Microseismic Event Locations with DAS Recordings
More LessSummaryDistributed Acoustic Sensing (DAS) measurements become ever more common to monitor oil fields. Event locaitons iwht single component DAS data is however difficult. We present a method to model the response of the fiber fro different wrapping angles to improve detection capabilities. Furthermore, we discuss how corkscrew well paths can improve event locations
-
-
-
Effective Monitoring of Hydraulic Fracture Stimulations Using Microseismcity: Part 1, Source Scaling Behavior and Stress Release
Authors T. Urbancic and E. ArdakaniSummaryTo utilize the full potential of microseismic data collected during a hydraulic fracture stimulation, there needs to be an effective approach to determining additional parameters of the rupture process associated with each microsiemic event. The use of ‘penny shaped crack’ models of rupture and variants allows for the determination of estimates of fracture dimensionality and stress release, such as seismic stress drop or apparent stress. Accordingly, global investigations of seismicity over seven orders of magnitude, suggest that seismicity behaves in a self-similar manner, where the ‘penny shaped crack’ dimensions (source radius) scales with seismic moment resulting in a constant stress drop over the size scale range considered. In this paper we examine the scaling relationship of microseismicity for events with M<-0.5, as recorded with downhole array configurations for a hydraulic fracture stimulation. We identify departures from self-similarity and suggest that those are related to issues in data collection and processing, and when, appropriately addressed, the resultant behavior can be considered as self-similar. Utilizing scaling relations can form the basis for quantifying the robustness of data for post-acquisition processing and interpretation and deriving meaningful interpretive results.
-
-
-
Effective Monitoring of Hydraulic Fracture Stimulations Using Microseismcity: Part 2, Tapered Gutenburg – Richter Relationship
By T. UrbancicSummaryConsideration is made of how a tapered Gutenburg-Richter relationship could be adapted and applied to hydraulic fracture stimulation related seismicity and used to improve data completeness. Combined with the observed self-similar source scaling behavior the recorded events are related to fracture length and surface area of individual events and cumulatively for the reservoir volume under consideration. As shown, the cumulative surface area increases by 40% by eliminating recording biases, and over 100% when considering the minimum fracture length for flow (set to 1 m). Based on these analyses, it can be suggested that a tapered Gutenburg-Richter fit provides a more robust approach to characterizing reservoir behavior that the traditional Gutenburg-Richter relationship.
-
-
-
Developing Approaches for Background Seismicity Monitoring of Potential CO2 Storage Sites in Horda Area Offshore Norway
Authors R. Dehghan-Niri, V. Oye, A. Wuestefeld, R. Bakke, M. Wilks, A. Furre and P. RingroseSummaryEquinor in collaboration with Shell and Total is working on maturing the carbon storage project to store industrially produced CO2 into geological subsurface offshore Norway. However, the selected candidates for CO2 storage are part of a region with moderate natural seismicity. In order to assure a safe storage, the background seismicity should be monitored to understand both the nature of natural seismicity and to detect possible events induced by the imposed pressure changes due to CO2 injection.
Current existing seismic network onshore is rather sparse and due to the limitation of recording mostly from the Norwegian side, event location uncertainty is rather high which complicates associating individual earthquakes to specific faults. Integration of onshore network with selected offshore PRM stations improved detectability. However, a more local monitoring system is required to improve the detectability and reduce the location uncertainty. We here present a proposal for a seismic monitoring array design for the offshore setting around the Smeaheia and Aurora sites that includes an upgrade of the onshore monitoring network as well as some ocean bottom nodes. We argue that such a network will provide an adequate baseline dataset, which is crucial for understanding the site prior to injection.
-
-
-
Automatic 3D-Seismic Fault Interpretation and Slip Potential Analysis from Hydraulic Fracturing in the Bowland Shale, UK
More LessSummaryThe Bowland Shale in central United Kingdom, has shown to be very susceptible to induce anomalously large seismicity during hydraulic fracturing stimulations of shale-gas wells. In part, due to the presence of pre-existent, critically-stressed, strike-slip faults that were not previously detected in 2D or 3D seismic surveys due to their low (and in some cases even null) vertical displacement, and only visible through the location and fault planes of microseismic events once these faults were reactivated. To better identify fault planes from 3D seismic images, and their reactivation potential due to hydraulic fracturing, a high-resolution fault-detection attribute was first tested in the same 3D seismic survey and at the same depth where the two most recently shale-gas wells were hydraulic-fractured in the Bowland Shale in 2018 and 2019. Their fault slip potential was then estimated by integrating the obtained faults with the formation’s stress and pore pressure conditions, where several critically stressed faults were identified near the previously hydraulic fractured wells and with similar location and orientation than the induced seismicity reported for the same wells.
-