- Home
- Conferences
- Conference Proceedings
- Conferences
81st EAGE Conference and Exhibition 2019 Workshop Programme
- Conference date: June 3-6, 2019
- Location: London, UK
- Published: 03 June 2019
81 - 93 of 93 results
-
-
Transdimensional Bayesian Thermochemical Joint Inversion of Seismic, Gravity and Surface Elevation Data
Authors D. Molodtsov and J. FulleaSummaryWe introduce a transdimensional probabilistic inversion algorithm in which seismic traveltime, gravity and surface elevation data are inverted for thermochemical parameters of the crystalline crust.
-
-
-
An Overview of the Geophysical Challenges in Monitoring and Locating Induced Earthquakes with Downhole Networks
Authors S. Minisini, C. Willacy, E. Van Dedem, J. Li and J.W. BloklandSummaryWe analyze the challenges that may be encountered when trying to locate microseismic events recorded by downhole networks in complex geological settings. The observed seismic events are characterized by significant complexity due to the high velocity contrasts and full waveform modeling is used to better understand and interpret the arrivals. Conventional location methods based on ray tracing for microseismic event location and moment tensor inversion may not be optimal for such an environment. To address this challenge, we developed a full waveform matching methodology for locating microseismic events and show its application on real dataset.
-
-
-
Avoiding Pitfalls and Extracting Value: Lessons for Induced Seismicity Monitoring.
Authors B. Dando, V. Oye, B. Goertz-Allmann and A. WuestefeldSummaryWe present case studies from microseismic monitoring including the Groningen gas field and the Decatur CCS site. We highlight often overlooked monitoring considerations including choice of location algorithm and waveform complexity, and demonstrate how modelling can avoid or mitigate the negative effects. Interpretation of microseismicity can lead to valuable insight into reservoir processes. However, to extract the most value, analyses must go beyond standard processing techniques. We show how combining waveform cross-correlation, full-waveform modelling, travel-time and ray-path analysis, and relative relocation methods can provide the necessary constraints to deliver improved interpretations from microseismic data sets.
-
-
-
The Need for Advanced Traffic Light Systems for Risk Mitigation of EGS Projects
Authors P. Meier, F. Bethmann and O. ZinggSummaryMany enhanced geothermal systems (EGS) projects rely solely on traditional magnitude based traffic light systems in order to mitigate seismic risks. However, in several geothermal projects (e.g. Basel and St. Gallen in Switzerland, Pohang in South Korea) traditional traffic light systems have fallen short to limit induced seismicity to a predefined level of event magnitude.
Our risks studies for the planned EGS project in Haute-Sorne in Canton Jura (Switzerland) and the evaluation of the Basel case highlight the need (1) for an advanced traffic light system taking into account the spatial distribution of seismicity and integrating continuously throughout all project stages new data and especially information about fault structures and (2) to update the risk studies throughout the duration of the project.
Such a procedure has been the basis for the permit by the authorities of the Canton Jura for the planned EGS project in Haute-Sorne and is also recommended by the “Good practice guide for managing Induced Seismicity in Deep Geothermal Energy Projects in Switzerland” published by the Swiss Seismological Survey.
Firstly we list some of the conditions of the permit and secondly we discuss selected points in light of the experiences of the Basel and Pohang EGS projects.
-
-
-
Using Vertical DAS Arrays for Continuous Monitoring of Induced Seismicity
Authors A. Lellouch, Z. Spica, B. Biondi and W. EllsworthSummaryOne of the most promising applications of DAS to the field of seismology is the possibility to permanently record data with downhole arrays with a much denser spatial sampling than would be possible with conventional geophones arrays. The data recorded by these arrays can be used for continuous (or periodical) monitoring of both natural and anthropogenic seismicity. Using two datasets we show how the data recorded from vertical arrays can be used: 1) to estimate 1D velocity models and 2) to detect events with high-sensitivity based on the events moveout computed with the estimated velocity models. The continuous nature of the DAS array allows us to accurately retrieve a high-resolution P-velocity model and extract an S-velocity model that couldn't be estimated from conventional VSP data.For one of the datasets we compare our detection results with an existing events catalogue. We find we are able to detect above 75% of cataloged events within a radius of 15 km and one weak new (uncatalogued) event.
-
-
-
The Role of Parallel Fracture Networks for Induced Seismicity in the Duvernay Formation
Authors Nadine Igonin, James P. Verdon, M. Kendall and David EatonSummaryFluid injection and hydraulic fracturing can cause induced seismicity. Two commonly proposed causative mechanisms are pore pressure and stress perturbations, or a combination of the two. Since most studies of induced seismicity due to hydraulic fracturing to date have been recorded on a regional scale, they lack the resolution to shed light on some of the key questions currently in the field. Using a high-quality dataset obtained during hydraulic fracturing in the Fox Creek, Alberta area, the detailed mechanisms of fault activation were examined. This experiment, dubbed the Tony Creek dual Microseismic Experiment (ToC2ME), contains several events over MW 2.0, and shows high-resolution fault activation. Seismic anisotropy derived from induced events was used to determine the direction of a pre-existing fracture network, which is attributed to be the main conduit by which fault activation was triggered. Pore pressure modelling was carried out to show that the delay times between injection and activation are consistent with the time necessary for a sufficiently significant pore pressure perturbation to reach the main fault features.
-
-
-
Modelling the Seismic Response of Flooded Faults in an Ultra-Deep South African Gold Mine
Authors J. Gerber and G. Van AswegenSummaryIn this study, the displacement-discontinuity boundary-element method is used to simulate large-magnitude, slip-type failures associated major geological structures in an ultra-deep, South African gold mine. The above-mentioned slip-type failures are simulated using a so-called RIDE algorithm, and the modelled seismic response is described in terms of the expected number of seismic events with some potentially-damaging size. In this case, the modelling results are well correlated with the observed seismic response.
-
-
-
Induced Seismic Background Disturbance Due to Geomagnetic Pulses
By S. RiabovaSummaryThe purpose of this work is to determine the response of the microseismic background to pulsed geomagnetic variations for the conditions of the mid-latitude Mikhnevo Geophysical Observatory of Institute of Geosphere Dynamics of Russian Academy of Sciences. During pulsed variations of the magnetic field of SSC and SI types, in most cases, increased variations of the seismic background are found. Spectral analysis shows that the induced variations in the seismic background are mainly observed in the frequency range of 0.01 – 0.1 Hz (in some cases, 0.001 – 0.1 Hz). The statistically significant correlation between amplitudes of induced microseismic background and geomagnetic pulses is indicated by high correlation coefficients calculated by different methods and by surrogate data analysis. The quantitative relationship between the studied variables was obtained. The application of methods of statistical data processing made it possible to establish that it is the geomagnetic field variations that entail a change in the microseismic background and with some delay.
-
-
-
The Future of 4D Subsurface Modelling: Reflections from a Multi-Sector Conference
Authors G. Burridge, T. Finkbeiner, J. Herwanger, W. Hohl, R. Plumb, K. Royse, J. Booth and G. McKinleySummaryIn February this year, the Geological Society of London held a conference entitled 4D Subsurface Modelling: Predicting The Future. Its objective was a novel one: To explore the lessons to be gained by comparing the four industrial geology sectors: Oil & Gas, Mining, Civil Engineering and Geothermal.
-
-
-
On Strategies for Consistent Local Geomodel Updating
By G. CaumonSummaryModel updating is an important way forward to efficiently incorporate new information about the subsurface. Four local components are needed to address common needs: finding the subregion to be updated, updating the discrete model features (topology) locally or semi-globally, locally updating the feature geometry and locally updating the associated properties. We present a few existing strategies and highlight that topological and geometric editing are essential for geologically consistent and automatic or semi-automatic geomodel updates.
-
-
-
Big Loop Reservoir Forecasting in the Machine Learning era
Authors F.O. Alpak and M. Araya-PoloSummaryWith the arrival of Machine Learning (ML) techniques as effective alternatives to many legacy modeling steps, classical static and dynamic reservoir modeling workflows need re-adjustment. In particular, we will focus on Big Loop (BL) approaches to reservoir modeling, where subsurface disciplines create an integrated representation of the subsurface, calibrated to static and dynamic information, for reliable field development and reservoir management decision making. The commonality is
Finally, we show some of the specific ingredients of an evergreen ML-driven Big Loop workflow.
-
-
-
Technology of Modeling and Simulation Integration and its Application
Authors Zhang Yuxiao, Xia Jian and Han ZhiyingSummaryDue to oil development areas are gradually entering high or extra-high water cut stage and the distribution of remaining oil is extremely sophisticated, a set of integrated technology for fine reservoir modeling and reservoir numerical simulation is formed to perform a deep research, according to different characteristics of reservoir and difficulties of exploration and development, The technology is based on the multi-scale data matching of the fine reservoir modeling, combined with the method of numerical simulation. The technology follow the rules of “analyzing contradictions, resolving contradictions” to continuously optimize the reservoir geological model and fluid distribution model by using the method of well-to-seismic integration. The series of integrated technology involve in many technical fields and a variety of reservoir types. It has been successfully applied in several development areas.
-
-
-
A Multi Domain Machine Learning Approach for Reliable Production Forecast
Authors S. Walia and V. AgarwalSummaryBig Loop™ solution meets these evolving needs and provides oil and gas operators with a platform to manage the challenges and complexities of modern reservoirs. The Big Loop workflow results in a calibrated flow model that is consistent with the underlying geology. The workflow is easy to update and allows experts to spend more time analyzing the results and building a common understanding of the reservoir instead of manual adjustments. Big Loop's automated, ensemble based, stochastic workflow tightly integrates the static and dynamic domains, ensuring that all the relevant reservoir uncertainties are captured and used as input parameters integrated to the reservoir simulator. By adjusting uncertainty ranges these input parameters, multiple realizations of the static and dynamic model, constrained by factors such as production history, are created via an iterative loop. (Big Loop - Trade mark of Emerson)
-