Browsing by Author "Eaton, David"
Now showing 1 - 20 of 20
Results Per Page
Sort Options
- ItemOpen AccessCenozoic exhumation history of the northern Richardson Mountains: Results from apatite and zircon (U-Th-Sm)/He analysis(2020-08-17) McKay, Ryan D.; Enkelmann, Eva; Hadlari, Thomas; Enkelmann, Eva; Hadlari, Thomas; Hubbard, Stephen; Eaton, DavidNew low-temperature thermochronology data from clastic sedimentary rocks in the northern Richardson Mountains, Canada, indicate significant exhumational cooling during late Eocene–early Oligocene time. Apatite (U-Th-Sm)/He (AHe) data were collected from 19 Proterozoic–Paleocene rocks across a 115 km transect. Eighty-eight single-grain AHe dates range from 300 ± 5 Ma to 16 ± 4 Ma but most dates (65%) range 56–24 Ma. AHe dates are generally younger than the stratigraphic ages, indicative of thermal resetting by burial. Additionally, zircon (U-Th)/He (ZHe) data were collected from two Jurassic sandstones. The ten ZHe dates range from 831 ± 12 Ma to 291 ± 4 Ma and are older than the stratigraphic age, which limits maximum burial to <160°C. Our results from thermal history modeling indicate three phases of cooling, during the Paleocene–early Eocene (>65–50 Ma), late Eocene–early Oligocene (40–30 Ma), and late Oligocene–early Miocene (30–15 Ma). Most samples were sensitive to cooling during the first and second phases, whereas the third phase is not as well constrained. The rocks were close to surface temperatures and below the sensitivity of AHe analysis since the early–middle Miocene. The results suggest a previously unrecognized phase of cooling and inferred deformation in the northern Richardson Mountains during late Eocene–early Oligocene time. Our findings contribute to previous work that recognizes Late Cenozoic deformation along the eastern margin of the Northern Cordillera, in the northeastern Brooks Range (e.g. O’Sullivan et al., 1993; 1998b) and in the Mackenzie Mountains (Enkelmann et al., 2019). We investigated the potential mechanisms of this widespread deformation and found that the exhumation may relate to kinematic changes of the North American plate relative to structural trends along the margin of the Northern Cordillera.
- ItemOpen AccessCharacterization of damage processes in Montney siltstone under triaxial compression using acoustic emission and diagnostic imaging(Oxford University Press, 2021-11-04) Jia, Suzie; Wong, Ron; Eaton, DavidCrack nucleation and rock failure processes in a fine-grained siltstone (Montney Formation) under triaxial compression are investigated using combined diagnostic techniques, including ultrasonic-wave measurement, acoustic-emission (AE) monitoring, computed tomography (CT) scanning, and thin-section imaging. The sample displays a weak-to-moderate inherent seismic anisotropy and noticeable stress-induced anisotropy prior to failure. No AE event was detected until the applied axial stress reached 95% of the peak value. The signal-to-noise ratio is relatively low, however, and detectable AE events are more diffuse than those observed in highly brittle rocks. The AE locations correlate with a shear fracture zone imaged by CT scanning. AE moment-tensor analysis reveals that events with larger relative magnitudes are characterized by high volumetric (tensile or compressive) components, and the initiation of the failure zone is dominated by combined shear-tensile failure. Stress inversion of the AE events with high tensile components is in good agreement with the known applied stress. Microscopic imaging of thin sections from the failed sample shows that the failure zone is an en echelon structure consisting of a major fracture with branching micro and minor cracks. This failure mechanism is consistent with a shear-tensile source mechanism and is interpreted to be associated with the fine granular structure and mineral composition of Montney siltstone.
- ItemOpen AccessCrustal structure beneath Hudson Bay from ambient-noise tomography(2012) Pawlak, Agnieszka; Eaton, DavidHudson Bay is a shallow inland sea located in north-central Canada. The underlying lithosphere preserves a complex deformational history that dates back to the Archean. The Hudson Bay Lithospheric Experiment, HuBLE, is a collaborative initiative aimed at understanding the lithospheric evolution beneath the Bay. The recent emergence of a methodology called ambient-noise tomography provides a tool to image the crust and upper mantle beneath the Bay with higher resolution than previously possible. Using ambient-noise generated by the Earth as a source, this technique requires continuous recordings of ground motion. The ambientnoise method is based on the cross-correlations of daily noise signals between station pairs to estimate empirical Green's Functions. This thesis is made up of there separate studies. The first is an isotropic application of the ambient-noise tomography method to image crustal structure beneath Hudson Bay. Results show crustal thinning beneath the Bay, allowing us to reject a hypothesis for eclogitization and crustal thickening, with support instead for an extensional hypothesis for the formation of the Hudson Bay basin. The next study focuses on anisotropic variations of velocity within the subsurface. Inversion results show a distinct outline of geologic boundaries in the upper to mid-crust that does not carry through into the lower crust. A significant change in anisotropic fabric is evident across the Trans-Hudson orogen (THO) suture zone, which allows us to establish that tectonic fabrics formed prior to collision. The third study employs joint inversion of ambient-noise data and teleseismic surface wave data for increasing resolution of the crust and upper mantle. Results show that the THO suture zone dips to the southeast within the crust and becomes vertical in the upper mantle. This feature is interpreted as a zone of weakness that extends through the lithosphere, providing a locus for initiation of localized lithospheric stretching.
- ItemOpen AccessDelineating the Shallow Free-phase Gas Distribution at an Abandoned Exploration Well, Crystal Geyser(2022-01-20) Lagasca, Patrick Arceo; Ryan, M. Cathryn; Bentley, Laurence; Eaton, DavidCrystal Geyser is an abandoned exploration well and active cold-water geyser driven by the expulsion of CO2 gas. It is a case study of an unsealed well transporting stray CO2 gas to the surface. Previous studies in the Crystal Geyser area inferred the existence of free-phase gas (FPG) accumulations trapped beneath low permeability layers near the geyser, but did not specify their exact locations or depths. In our primary study, we used electrical resistivity tomography to image a siltstone-capped FPG plume within the sandstone aquifer beneath Crystal Geyser. The plume was identified from an anomalously high resistivity zone that was not caused by lithofacies changes or fault displacement, representing a portion of the aquifer with elevated FPG content in the pore space. FPG is transported via Crystal Geyser’s conduit and the nearby Little Grand Wash Fault into a plunging anticline trap. The poor well casing integrity allows FPG to migrate from the conduit into the surrounding aquifer and gas trap. The subsurface FPG plume delineation was compared to the “equivalent CO2 bubbling depth” – the shallowest depth at which the total CO2 mass (free-phase and dissolved-phase CO2) expelled from the geyser would occur in the dissolved phase. The bubbling depth was extrapolated from mass balance calculations based on field estimates of Crystal Geyser’s CO2 and water emissions. The estimated bubbling depth was much shallower than the imaged FPG plume. This underestimation can be caused by the leakage of ascending FPG into the surrounding subsurface, allowed by the poor casing integrity of the well.
- ItemOpen AccessDownhole microseismic monitoring: processing, algorithms and error analysis(2014-04-09) Akram, Jubran; Eaton, DavidBasic microseismic data processing for hypocentre location estimates is a pre-requisite to extracting information about the stimulated reservoir volume as well as understanding the geomechanics of the fracturing process. The primary objective of this thesis is to investigate the basic processing of microseismic data acquired with receivers in a single observation well, with emphasis on the evaluation and parameter selection of processing algorithms and the associated hypocentre location error analysis. This thesis is made up of five independent studies. The first study discusses the development of a MATLAB based microseismic data processing package (Calgary microseismic processing system; CaMPS). This software is used to process the microseismic data from a hydraulic fracture treatment in western Canada. For reference, the results are compared with those obtained independently by a microseismic services company. The second study examines several single-trace event-detection and arrival-time picking algorithms for microseismic data. A dynamic threshold criterion for event detection and a hybrid, arrival-time picking approach are proposed. The performance of these algorithms is evaluated using synthetic and real microseismic data. The third study describes an iterative cross-correlation based workflow to refine the initial arrival-time picks. This workflow is compared with other single-trace and multi-trace techniques. The proposed workflow provides an arrival-time accuracy of ±0.5 − 1ms for both synthetic and real microseismic data examples considered in this study. The fourth study examines hypocentre location uncertainty and errors due to inaccurate velocity model. The Monte Carlo uncertainty analysis suggests that the velocity errors have a greater impact on hypocentre locations than arrival-time pick errors. The hypocentre location errors resulting from the model calibration process are also discussed, in particular the use of single vs. multiple calibration shots, a priori information, and first vs. direct arrival times. The fifth study discusses the remaining hypocentre location errors after anisotropic model calibration. The behaviour of hypocentre location is discussed when a 1-D layered, isotropic and homogeneous model is used to locate hypocentres from anisotropic and heterogeneous subsurface. The results emphasize the use of a detailed model with anisotropy and lithological or structural variations for improving hypocentre location accuracy.
- ItemOpen AccessExperimental Simulation and Life Cycle Assessment Study of the Dense Hot Fluid Injection Process(2016-02-03) Hovsepian, Christian Nubar; Pereira-Almao, Pedro; Bergerson, Joule; Eaton, David; Hejazi, Hossein; Nassar, NashaatThe rapid decline in international oil prices forced oil and gas industries to reduce costs, improve productivity and layoff thousands of employees. Specifically, the majority of resources located in Alberta require thermal recovery methods due to their heavy characteristics. Innovation is required to displace current extraction methods that are becoming economically challenging. The present work studies the dense hot fluid injection (DHFI) process, which targets the in situ catalytic upgrading and extraction of Athabasca bitumen. A bench-scale set up was designed and built. The experimental results were integrated using process simulation and life cycle assessments (LCA) techniques to estimate GHG emissions of the new technology. Results confirmed quality improvement of the feedstock while maintaining a stable product. The experimental setup proved differences in the heat distribution profiles between SAGD and the DHFI. LCA demonstrated that the DHFI could produce less GHG emissions than SAGD under a set of assumed scenarios.
- ItemEmbargoExtended-FEM analysis of injection-induced slip on a fault with rate-and-state friction: Insights into parameters that control induced seismicity(Springer, 2023-03-09) Hosseini, Navid; Priest, Jeffrey; Eaton, DavidThe extended finite element method (X-FEM) is utilised to simulate the behavior of a heterogeneous fault characterized by rate-state frictional rheology, embedded within a poroelastic medium. The displacement and pore-pressure fields that are discontinuous across the fault are computed using X-FEM, by enriching the standard finite element approximation with additional degrees of freedom for elements intersected by the fault. We investigate a Mw 4.1 injection-induced earthquake in western Canada; this model incorporates depth-varying rate-slip behavior wherein a high-pressure zone due to hydraulic fracturing stimulation intersects the fault within a stable layer, producing aseismic slip that progressively loads an unstable fault region, thereby triggering dynamic rupture. Parametric studies using our numerical approach provide insights into the influence of rate-state parameters on fault activation, as well as hydraulic properties of a damage zone that surrounds the fault. Results confirm that aseismic slip near the injection zone propagates outwards to seismogenic unstable regions of the fault. The coseismic slip profile, seismic moment, and slip latency are determined by the difference a − b for rate-state parameters of the unstable fault regions. Hydraulic diffusivity in the damage zone controls the rate of pore-pressure diffusion along the fault, which affects timing of the initial seismic event and aftershock productivity.
- ItemOpen AccessFinite-difference Simulation of Synthetic Microseismograms in the Presence of Anisotropy and Attenuation(2015-09-25) Rashedi, Hoda; Eaton, DavidThis study presents a suite of 2D synthetic microseismic events computed using three velocity models, incorporating varying degrees of anisotropy and attenuation. An explicit second-order finite-difference method is used, capable of handling general anisotropy, up to triclinic symmetry. The microseismic source is specified as an arbitrary moment tensor, subject to be located in an isotropic layer. The anisotropy in our models is confined to VTI (Vertically Transverse Isotropic) media. Attenuation can be simulated for both isotropic and VTI medium, so the method is suited to modeling wave propagation in shale. All of the examples shown in this thesis are for a downhole acquisition geometry using a vertical monitor well. Care is required to avoid contamination of the synthetic data by edge reflections due to the ineffectiveness of currently known FD absorbing boundary for general anisotropy. Synthetic data (SEGY) are available for benchmarking purposes including hypocentre location, moment tensor inversion, etc.
- ItemOpen AccessFour Phenomena Observed on Microseismic Data(2013-09-13) St-Onge, Andre Maurice; Eaton, DavidPassive seismic data recorded in observation wellbores for hydraulic fracture monitoring are usually used to detect P- and S-wave arrivals from induced microseismic events. The time arrivals for these events are used to compute source locations to build a fracture location map. Along with these arrivals are a number of other coherent signal and noise events. The primary objective of this thesis is to examine various other types of coherent signal and noise on four microseismic datasets. This thesis is made up of four separate studies. The first study examines numerous coherent low-frequency (<100 Hz) arrivals characterized by conspicuous dispersion and quasi-linear moveout. Their apparent velocities and dispersion characteristics are consistent with Lamb waves, an elastic guided wave that propagates within finite media such as plates or cylindrical casing. Lamb waves may be used for monitoring degradation of casing cement over time. The second study examines two datasets to characterize frequency shifts in discrete narrow passbands caused by wellbore acoustic transmissivity due to geophone clamping. The passbands changed throughout two fracture stimulations in proportion to the hydraulic fracture pressure. Analytical and finite-element models suggest that these temporal variations may be caused by a relative change in geophone clamping force as the pressure external to the wellbore varies. This sensitivity suggests that passive seismic monitoring can be used for downhole detection of relative stress changes. The third study focuses on frequency content variations of 20 high S/N P- and S-wave events. The 20 events were grouped into four clusters based upon a geolocation algorithm and event cross-correlations. Spectral analysis shows variation in the signal within event sets. A 2-D elastic finite-difference simulation using a layered model shows that this variation could be ascribed to path effects associated with propagation of the signal from the source to the receivers. The fourth study examines a potential long-period long-duration (LPLD) seismic event and analyzes the phenomena in detail. LPLD events observed on microseismic data have been likened to deep low-frequency tremor attributed to slow-slip processes along pre-existing fractures or strike-slip regimes at subduction zones. The event may be the result of hydraulic fracture fluid leak-off out of the intended zone into a pre-existing fracture.
- ItemOpen AccessHydraulic-fracturing induced seismicity in Alberta, Canada: Analysis and interpretation using dense local arrays(2021-04-28) Igonin, Nadine; Eaton, David; Lauer, Rachel; Lawton, DonaldUnconventional oil and gas development is associated with anthropogenic earthquake activity worldwide. In particular, hydraulic fracturing, which involves the injection of high-pressure fluids into the subsurface, has been observed to trigger induced seismicity in some areas. The Duvernay Formation in Alberta is a target of hydraulic fracturing and has experienced induced earthquakes up to magnitude 4.2. To study hydraulic fracturing-induced seismicity in greater detail, a local dense array of sensors was used to collect four weeks of passive seismic data over a 4-well pad. In this thesis, this dataset is processed and interpreted using a variety of techniques. Supplementary information, such as waveform data from broadband seismometers and an accelerometer that were co-located with the dense geophone array, is also used to aid the interpretation. In some cases, it is found that the magnitude-frequency distribution obtained from the monitored seismicity deviates from the expected power-law form associated with the Gutenberg-Richter relationship. In particular, the largest earthquakes, greater than Mw 3.0, are larger than statistically expected given the observed seismicity. Spectral analysis of a subset of events reveals complex source spectra. For example, some of the events show evidence of a double corner frequency and a source time function with multiple peaks. A detailed analysis of the timing of events relative to the hydraulic fracturing stages reveals that natural fracture networks within the Ireton Shale, overlying the target Duvernay Formation, provided permeable conduits for an expanding region of elevated pore pressure. Using simple diffusive modelling, it is shown that these fracture networks allowed pore pressure to perturb fault systems at distances of over 1 km, consistent with observations of fault activation during treatment. In two datasets that are used in this thesis, shear-wave splitting analysis of microseismic waveforms provides evidence for stress rotation near the largest faults. Integration of these investigations provides a basis for a new working model for induced seismicity in the Fox Creek, Alberta, area.
- ItemOpen AccessIntegrated approach for fracture characterization of hydraulically stimulated volume in tight gas reservoir(2019-04-17) Maulianda, Belladonna; Prakasan, Aruvin; Wong, Ron C; Eaton, David; Gates, Ian DAbstract Hydraulic fracturing is conducted on unconventional reservoir which has very low permeability. It increases the production from unconventional oil and gas reservoirs through the creation of a connected stimulated rock volume (SRV) with higher conductivity. The permeability and the SRVs dimension are important parameters which increase the performance of hydraulically fractured wells. Microseismic monitoring is used to estimate the seismically stimulated volume within the reservoir, which can provide a proxy for the SRV. Finite element analysis was used in this study in the determination of SRV characteristics by utilizing field data from a horizontal well hydraulic-fracturing program in the Hoadley Field, Alberta, Canada. Coupled fluid-flow geomechanics finite element (FE) model was used. The permeability of the SRV is altered to match the field bottom-hole pressure. The pressure drop and in situ stress changes within the SRV are determined through the matching of the FE model. Fracture aperture, number and spacing in the SRV are then inferred from the estimated reservoir parameters by using a semi-analytical approach.
- ItemOpen AccessInvestigating Fault-Sealing Effects on Pore Pressure Distribution, Induced Seismicity and Hydraulic Fracture Propagation: Numerical Modelling and Case Study in Northeastern British Columbia(2023-02-24) Esmaeilzadeh, Zahra; Eaton, David; Clarkson, Christopher; Gilbert, Hersh; Chen, Nancy; Lauer, Rachel; Fox, AmyThis thesis focuses on the effects of sealing faults and associated lateral pore-pressure gradients on induced seismicity and hydraulic fracturing (HF) operations. The studied formation is the Montney, a Triassic geological unit located in the Western Canada Sedimentary Basin that hosts one of the most productive unconventional plays in the world. The study area is located in northeastern British Columbia (BC), situated within the Kiskatinaw seismic monitoring and mitigation area (KSMMA), in the Septimus field where a significant induced earthquake sequence occurred in November 2018. Previously published reports show that pore pressure in the Montney Formation is compartmentalized, indicating that permeable and impermeable features such as fractures and sealing faults may control the reservoir fluid flow. Notably, the locations of pressure discontinuities are generally consistent with known faults, supporting a previous hypothesis that pressure compartments are fault-bounded. This thesis tests this hypothesis and models the effects of a horizontal pressure gradient (∆P) on fault activation and hydraulic fracture propagation. Novel methodologies have been developed for investigating complex interactions between pore-pressure gradient, seismicity and hydraulic fracturing mechanism, consisting of three workflows based, respectively, on statistical analysis of pore-pressure data, numerical simulation of hydraulic fracturing, and numerical modelling of fault activation. First, a comprehensive reservoir model is considered to characterize regional hydrodynamic, petrophysical and mechanical properties. The compiled data are used to construct a new pressure map of the Montney Formation, yielding insights into relationships between pressure boundaries and seismicity. The results confirm that pressure discontinuities are generally fault-bounded, but seal preservation depends on depth and fault properties. Earthquakes (both induced and natural) appear to cluster preferentially in areas of a high lateral pore-pressure difference. Next, using grid oriented hydraulic fracture extension replicator (GOHFER) software, HF growth is modelled in the presence of an impermeable fault and surrounding permeable damage zone. Different scenarios are considered based on ∆P and damage-zone properties. Results show that fault sealing behaviour depends on the pressure difference, stage shadowing effects, fault throw, injection rate and fault damage-zone permeability. The adverse effects of a damage zone on fluid loss of injected fluid are more pronounced when there is a lateral pressure gradient across the fault. The order of importance of these parameters implied by the modelling results is: stress shadowing > lateral pressure gradient > fault seal properties. Finally, the effect of lateral pressure gradient on fault activation is tested using 3D distinct element code (3DEC) software- 3D numerical modelling code that simulates the mechanical response of rock mass with discontinuities such as fractures and faults. Using this tool, sensitivity analysis of fault activation is performed based on the presence or absence of a significant pressure difference and a highly fractured and permeable damage zone on both sides of the fault. Strong ∆P may lead to asymmetric fracture geometry and considerable leakage of pore pressure into the lower pressure domain. A damage zone around the fault appears to channel pore pressure along the fault, leading to a more uniform distribution of fault aperture and pore pressure than in the absence of a damage zone, where the fault opening is concentrated near the HF intersection.
- ItemOpen AccessMicroseismic Data Processing, Modeling and Interpretation in the Presence of Coals: A Falher Member Case Study(2014-09-29) Pike, Kimberly Anne; Eaton, DavidLow-velocity, low-density coal units are prevalent throughout the Cretaceous section of the Western Canadian Sedimentary Basin. Their elastic properties differ considerably from the surrounding clastic units, resulting in complex seismic propagation paths and interference patterns. This thesis presents a case study evaluating the implications these coal units have on the processing, finite-difference modeling and interpretation of a borehole microseismic dataset recorded in the Falher Member of western Alberta's Deep Basin. Seismic energy generated during hydraulic fracturing of the Falher Member manifests as low-velocity, high-amplitude channel waves. A Matlab-based processing workflow is developed to analyze the data set, with emphasis on velocity model calibration, receiver orientation and hypocentre location in the presence of coals. Finite-difference modeling is used to evaluate the propagation of seismic energy through the monitoring interval, leading to the identification of complex P- and S-wave arrival patterns which significantly complicate microseimsic processing. Alternative acquisition geometries are evaluated via finite-difference modeling and offer improvements to the original acquisition design. The resulting hypocentre distribution suggests that the completion design effectively stimulated the reservoir along the wellbore. Systematic error introduced by the coal layers may contribute to distance-dependant uncertainty in the final hypocentre locations.
- ItemOpen AccessMicroseismic Interpretation of Hydraulic Fracture Processes in Unconventional Reservoirs: Geomechanics, Numerical Simulations and Data Integration(2016) Boroumand, Neda; Eaton, David; Lawton, Don; Krebes, Edward; Aguilera, Roberto; Angus, DougNew techniques are presented to advance microseismic interpretation of the hydraulic fracture process. Numerical simulations were developed to present the majority of the claims of novelty and a few previously developed ones were modi ed to present new insights. First, two processing techniques are used for analysis of a microseismic double couple source mechanism. A time shifted hyperbolic Radon transform scheme was applied to a synthetic and a real microseismic event to obtain its lag time and spatial location. Second, a new method is developed to deal statistically with missing microseismic events that are too far from the monitoring well to be detected. This new method, b value energy correction method (MbEC), begins by unbiasing the catalogue as a function of distance and magnitude, then characterizing the frequency magnitude distribution and nally correcting for missing radiated seismic energy. The corrected energy is compared to the fracture energy to calculate seismic e ciency and the fracture energy is compared to injection energy to compute fracture e ciency. The energy correction and comparisons are useful for evaluating changes in microseismic activity. Third, a numerical algorithm is developed to analyse the energy breakdown of the hydraulic fracture process. This energy breakdown based model technique is developed by incorporating microseismic events in space. This model is able to constrain and validate the model geometry through parameter adjustments. The novelty is in the integration of microseismic observations in the numerical simulation that was developed speci cally to characterize the reservoir. Fourth, a numerical algorithm is developed that couples the multi-physics interaction between formation elasticity, fracture uid ow and mechanics equations to initiate and propagate a fracture. The novelty in this algorithm lies in using the microseismic events as the fracture propagation criteria, in time and space, for reservoir property characterization. Case studies are presented in all chapters to illustrate the applicability of each new idea.
- ItemOpen AccessMicroseismic moment-tensor inversion(2011) Forouhideh, Farshid; Eaton, DavidIt is common practice in the oil and gas industry to use a single vertical well with limited number of geophones to monitor the microseismicity induced by hydraulic fracture treatment. In this thesis, I have employed a synthetic and stochastic modeling approach to investigate the effects of monitoring geometry and noise on inverting recorded P- and S-wave amplitudes for the moment tensor and source parameters responsible for generating the observed seismograms. The results of the synthetic modeling have shown that microseismic moment-tensor inversion using data from a single vertical monitoring wellis ill-conditioned and incapable of resolving all six components of the moment tensor. Mitigating this problem requires introducing new constraints and a-priori assumptions on mversion.
- ItemOpen AccessMulti-scale analysis of the influence of sedimentary fabric and composition on the geomechanical properties of organic-rich mudstones: a case study from the Duvernay Formation, Alberta, Canada(2021-06-24) Venieri, Marco; Pedersen, Per; Eaton, David; Ghanizadeh, Amin; Priest, Jeffrey; Wust, Raphael; Hart, BruceIn this research, the influence of sedimentary fabric and composition on the geomechanical properties of organic-rich mudstones has been investigated at the seismic, wireline log and core scale with resolution ranging from 50m- to the cm-scale. Our analysis reveals that a relationship is observable between composition, fabric, and elastic properties of the rock units at each scale of observation. Generally, carbonate-rich facies show the highest values of Young’s modulus (YM) and Poisson’s ratio (PR); clay-rich facies show the lowest YM and intermediate PR; biogenic silica-rich (“organic-rich”) facies show intermediate YM and the lowest PR. At the seismic scale, for which the resolution does not allow for assessment of intra-Duvernay mechanical heterogeneity characterization, significant difference in computed elastic moduli is observed between the organic-rich Duvernay Formation and its bounding strata which show different mineralogy. Analysis at the m- and cm-scale of compositional and geomechanical properties of the Duvernay Formation from outcrop samples reveals significant heterogeneity within the Duvernay Formation. Our research suggests that geological and geomechanical heterogeneity within the Duvernay Formation is facies-dependent. In fact, organic-rich mudstones show nearly twice as much variability in composition and mechanical hardness than carbonate-rich facies. This, in conjunction with our analysis of vertical and lateral geological heterogeneity within the Duvernay Formation from subsurface data, suggests that a cautionary approach should be adopted when using averaged or upscaled data for subsurface modeling which risk oversimplifying reservoir complexity. In this case, heterogeneity of rock properties should be included in reservoir models as a measure of uncertainty on the input data.
- ItemOpen AccessNonlinear Bayesian estimation of centroid moment tensors using multiple seismic data sets in the Kiskatinaw seismic monitoring and mitigation area(2022-12-22) Hamidbeygi, Mahdi; Dettmer, Jan; Eaton, David; Gilbert, HershCentroid moment tensor (CMT) parameters of earthquakes are routinely estimated to gain information on structures and regional tectonics. However, for small earthquakes (M<4) it is still challenging to determine CMTs due to the lack of high-quality waveform data. In this study, we propose to improve solutions for small earthquakes by incorporating multiple seismic data types in Bayesian joint inversion: polarities picked on broadband signals, amplitude spectra for intermediate frequency bands (0.2--2.0 Hz), and waveforms at low frequencies (0.05--0.2 Hz). Both measurement and theory errors are accounted for by iterative estimation of non-Toeplitz covariance matrices, allowing to objectively determine weights for the different data types in the joint parameter estimation. Validity and applicability of the method are demonstrated on simulation and field data. Results demonstrate that the combination of data, such as a single high quality waveform, a few amplitude spectra and many waveform polarities are able to resolve CMT parameters to comparable quality as if many high quality waveforms were available. Results of 10 induced events that occurred in northeastern British Columbia between January 2020 and February 2022 indicate predominant strike slip focal mechanisms with low non double-couple components. These events appear to be located at shallow depth with a short duration as expected for induced seismicity. These results are consistent with previous studies. Therefore, we learn that this method reduces the dependency of source inversion on high-quality waveforms and permits to resolve CMTs for earthquakes as small as ML 1.6.
- ItemOpen AccessTeleseismic Array Studies of Earth’s Core-Mantle Boundary(2011) Alexandrakis, Catherine; Eaton, David
- ItemOpen AccessTime-Lapse Numerical Modeling for a Carbon Capture and Storage (CCS) Project in Alberta, Using a Poroelastic Velocity-Stress Staggered-Grid Finite-Difference Method(2016) Moradi, Shahin; Lawton, Donald; Krebes, Edward; Eaton, David; Lines, Laurence; Landro, Martin; Fear, EliseA finite-difference algorithm was developed based on the Biot's equations of motion to model seismic wave propagation in poroelastic media. As opposed to the elastic case, in the poroelastic approach the properties of the pore fluid are taken into account in the modeling process. Poroelastic modelling could be useful in cases where the fluid content of the rock is of interest, such as Carbon Capture and Storage (CCS) projects. The developed program was then used to investigate the detectability of CO2 in a CCS project in Alberta. Two models were defined for the baseline and monitor scenarios that respectively, represented the subsurface before and after injection of CO2 and the corresponding synthetic seismic sections were generated. The difference between the calculated seismic sections for the two scenarios shows that the residual amplitude is comparable with the baseline amplitude. With this result, the injected CO2 in the Quest project over a year could be detected providing the data have good bandwidth and a high signal-to-noise ratio. Furthermore, a comparison between the poroelastic algorithm and the elastic algorithm shows that the time-lapse effect in the poroelastic case is smaller than the one in the elastic case. In the fluid saturated media, some of the wave energy is dissipated due to fluid viscosity, and the poroelastic approach helps us to take this loss into account in the modeling process.
- ItemOpen AccessVisualizing highly multidimensional time varying Microseismic Events(2012-10-16T16:48:30Z) Mostafa, Ahmed; Carpendale, Sheelagh; Brazil, Emilio; Eaton, David; Sharlin, Ehud; Costa Sousa, MarioMaking decisions about improving an oil and gas reservoir model based upon microseismic data is a difficult challenge for reservoir engineers and analysts. These difficulties arise because the available data contains inaccuracies, has high-dimensionality and has a high degree of uncertainty. Currently these difficulties are intensified by the lack of computational tools to support interactive visual interpretation and integration of geophysical data leading to robust structural models of the reservoir and its parameters. To address these difficulties domain experts are demanding better and more detailed visualization tools to help them as they explore their data. In this paper, we present a tool that contains a set of interactive visualizations that combines, merges and extends existing visualization techniques. We describe the iterative design process we undertook to develop the tool, relying on insight from domain specialists. Our tool supports 3D spatial analysis and exploration of the data with a set of manipulations designed to provide domain experts with insights into their highly complex microseismic data. Our microseismic visual-analysis tool also provides an extended parallel coordinates implementation to: (1) support interactive filtering and selection through combined filter and shadow boxes that can remove the uninteresting events from further analysis, (2) correlate between the data attributes by axes reordering and outlier discovery, and (3) visually correlate the data events rendering through additional visual elements such as color maps. Our multiple coordinated views link the insights gained from one view with other views instantaneously. We conclude with a discussion of the feedback provided to us by the domain experts.