Dissertations / Theses on the topic 'Engineering ; Geophysics'

The first part of my thesis is mainly focused on the effect of grain size distribution on compaction localization in porous sandstone. To identify the microstructural parameters that influence compaction band formation, I conducted a systematic study of mechanical deformation, failure mode and microstructural evolution in Bleurswiller and Boise sandstones, of similar porosity (∼25%) and mineralogy but different sorting. Discrete compaction bands were observed to develop over a wide range of pressure in the Bleurswiller sandstone that has a relatively uniform grain size distribution. In contrast, compaction localization was not observed in the poorly sorted Boise sandstone. My results demonstrate that grain size distribution exerts important influence on compaction band development, in agreement with recently published data from Valley of Fire and Buckskin Gulch, as well as numerical studies.

The second part aimed to improve current knowledge on inelastic behavior, failure mode and brittle-ductile transition in another sedimentary rock, porous carbonates. A micritic Tavel (porosity of ∼13%) and an allochemical Indiana (∼18%) limestones were deformed under compaction in wet and dry conditions. At lower confining pressures, shear localization occurred in brittle faulting regime. Through transitional regime, the deformation switched to cataclastic flow regime at higher confining pressure. Specifically in the cataclastic regime, the (dry and wet) Tavel and dry Indiana failed by distributed cataclastic flow, while in contrast, wet Indiana failed as compaction localization. My results demonstrate that different failure modes and mechanical behaviors under different deformation regimes and water saturation are fundamental prior to any geophysical application in porous carbonates.

The third part aimed to focus on investigating compaction on quartz aggregate starting at low (MPa) using X-ray diffraction. We report the diffraction peak evolution of quartz with increasing pressures. Through evaluating the unit cell lattice parameters and the volume of the quartz sample, macroscopic stress and strain were resolved. Moreover, we observed quartz peak broadened asymmetrically at low pressure, such extent is more prominent in axial than in radial direction. Our evaluation on peak [101] (highest intensity among peaks) demonstrated that full width at half maximum can be a good proxy for microscopic stress distribution. We observed deviations in the pressure-volume curves at P = ∼0.4 GPa and speculated that it was the point of which onset of grain crushing and pore collapse occur in quartz. This is on the same order of which onset of grain crushing (commonly known as P*) is observed in sandstones in the rock mechanics literature. This demonstrated that there is potential in estimating grain crushing and pore collapse pressure with our technique.

Brittleness is a key characteristic for effective reservoir stimulation and is mainly controlled by mineralogy in unconventional reservoirs. Unfortunately, there is no universally accepted means of predicting brittleness from measures made in wells or from surface seismic data. Brittleness indices (BI) are based on mineralogy, while brittleness average estimations are based on Young's modulus and Poisson's ratio. I evaluate two of the more popular brittleness estimation techniques and apply them to a Barnett Shale seismic survey in order to estimate its geomechanical properties. Using specialized logging tools such as elemental capture tool, density, and P- and S wave sonic logs calibrated to previous core descriptions and laboratory measurements, I create a survey-specific BI template in Young's modulus versus Poisson's ratio or alternatively λρ versus μρ space. I use this template to predict BI from elastic parameters computed from surface seismic data, providing a continuous estimate of BI estimate in the Barnett Shale survey. Extracting λρ-μρ values from microseismic event locations, I compute brittleness index from the template and find that most microsemic events occur in the more brittle part of the reservoir. My template is validated through a suite of microseismic experiments that shows most events occurring in brittle zones, fewer events in the ductile shale, and fewer events still in the limestone fracture barriers.

Estimated ultimate recovery (EUR) is an estimate of the expected total production of oil and/or gas for the economic life of a well and is widely used in the evaluation of resource play reserves. In the literature it is possible to find several approaches for forecasting purposes and economic analyses. However, the extension to newer infill wells is somewhat challenging because production forecasts in unconventional reservoirs are a function of both completion effectiveness and reservoir quality. For shale gas reservoirs, completion effectiveness is a function not only of the length of the horizontal wells, but also of the number and size of the hydraulic fracture treatments in a multistage completion. These considerations also include the volume of proppant placed, proppant concentration, total perforation length, and number of clusters, while reservoir quality is dependent on properties such as the spatial variations in permeability, porosity, stress, and mechanical properties. I evaluate parametric methods such as multi-linear regression, and compare it to a non-parameteric ACE to better correlate production to engineering attributes for two datasets in the Haynesville Shale play and the Barnett Shale. I find that the parametric methods are useful for an exploratory analysis of the relationship among several variables and are useful to guide the selection of a more sophisticated parametric functional form, when the underlying functional relationship is unknown. Non-parametric regression, on the other hand, is entirely data-driven and does not rely on a pre-specified functional forms. The transformations generated by the ACE algorithm facilitate the identification of appropriate, and possibly meaningful, functional forms.

This dissertation pursues three main objectives: (1) to investigate the seismic response of tall reinforced concrete core wall buildings, designed following current building codes, subjected to pulse type near-fault ground motion, with special focus on the relation between the characteristics of the ground motion and the higher-modes of response; (2) to determine the characteristics of a base isolation system that results in nominally elastic response of the superstructure of a tall reinforced concrete core wall building at the maximum considered earthquake level of shaking; and (3) to demonstrate that the seismic performance, cost, and constructability of a base-isolated tall reinforced concrete core wall building can be significantly improved by incorporating a rocking core-wall in the design.

First, this dissertation investigates the seismic response of tall cantilever wall buildings subjected to pulse type ground motion, with special focus on the relation between the characteristics of ground motion and the higher-modes of response. Buildings 10, 20, and 40 stories high were designed such that inelastic deformation was concentrated at a single flexural plastic hinge at their base. Using nonlinear response history analysis, the buildings were subjected to near-fault seismic ground motions as well as simple close-form pulses, which represented distinct pulses within the ground motions. Euler-Bernoulli beam models with lumped mass and lumped plasticity were used to model the buildings.

Next, this dissertation investigates numerically the seismic response of six seismically base-isolated (BI) 20-story reinforced concrete buildings and compares their response to that of a fixed-base (FB) building with a similar structural system above ground. Located in Berkeley, California, 2 km from the Hayward fault, the buildings are designed with a core wall that provides most of the lateral force resistance above ground. For the BI buildings, the following are investigated: two isolation systems (both implemented below a three-story basement), isolation periods equal to 4, 5, and 6 s, and two levels of flexural strength of the wall. The first isolation system combines tension-resistant friction pendulum bearings and nonlinear fluid viscous dampers (NFVDs); the second combines low-friction tension-resistant cross-linear bearings, lead-rubber bearings, and NFVDs.

Finally, this dissertation investigates the seismic response of four 20-story buildings hypothetically located in the San Francisco Bay Area, 0.5 km from the San Andreas fault. One of the four studied buildings is fixed-base (FB), two are base-isolated (BI), and one uses a combination of base isolation and a rocking core wall (BIRW). Above the ground level, a reinforced concrete core wall provides the majority of the lateral force resistance in all four buildings. The FB and BI buildings satisfy requirements of ASCE 7-10. The BI and BIRW buildings use the same isolation system, which combines tension-resistant friction pendulum bearings and nonlinear fluid viscous dampers. The rocking core-wall includes post-tensioning steel, buckling-restrained devices, and at its base is encased in a steel shell to maximize confinement of the concrete core. The total amount of longitudinal steel in the wall of the BIRW building is 0.71 to 0.87 times that used in the BI buildings. Response history two-dimensional analysis is performed, including the vertical components of excitation, for a set of ground motions scaled to the design earthquake and to the maximum considered earthquake (MCE). While the FB building at MCE level of shaking develops inelastic deformations and shear stresses in the wall that may correspond to irreparable damage, the BI and the BIRW buildings experience nominally elastic response of the wall, with floor accelerations and shear forces which are 0.36 to 0.55 times those experienced by the FB building. The response of the four buildings to two historical and two simulated near-fault ground motions is also studied, demonstrating that the BIRW building has the largest deformation capacity at the onset of structural damage.

(Abstract shortened by UMI.)

Hydraulic fracturing was first carried out in the 1940s and has gained popularity in current development of unconventional resources. Flowing back the fracturing fluids is critical to a frac job, and determining well cleanup characteristics using the flowback data can help improve frac design. It has become increasingly important as a result of the unique flowback profiles observed in some shale gas plays due to the unconventional formation characteristics.

Computer simulation is an efficient and effective way to tackle the problem. History matching can help reveal some mechanisms existent in the cleanup process. The Fracturing, Acidizing, Stimulation Technology (FAST) Consortium at Colorado School of Mines previously developed a numerical model for investigating the hydraulic fracturing process, cleanup, and relevant physics. It is a three-dimensional, gas-water, coupled fracture propagation-fluid flow simulator, which has the capability to handle commonly present damage mechanisms.

The overall goal of this research effort is to validate the model on real data and to investigate the dominant physics in well cleanup for the Cana Field, which produces from the Woodford Shale in Oklahoma.

To achieve this goal, first the early time delayed gas production was explained and modeled, and a simulation framework was established that included all three relevant damage mechanisms for a slickwater fractured well. Next, a series of sensitivity analysis of well cleanup to major reservoir, fracture, and operational variables was conducted; five of the Cana wells' initial flowback data were history matched, specifically the first thirty days' gas and water producing rates.

Reservoir matrix permeability, net pressure, Young's modulus, and formation pressure gradient were found to have an impact on the gas producing curve's shape, in different ways. Some moderately good matches were achieved, with the outcome of some unknown reservoir information being proposed using the corresponding inputs from the history matching study. It was also concluded that extended shut-in durations after fracturing all the stages do not delay production in the overall situation.

The success of history matching will further knowledge of well cleanup characteristics in the Cana Field, enable the future usage of this tool in other hydraulically fractured gas wells, and help operators optimize the flowback operations. Future improvements can be achieved by further developing the current simulator so that it has the capability of optimizing its grids setting every time the user changes the inputs, which will result in better stability when the relative permeability setting is modified.

As the Arctic sea ice extent shrinks, it becomes feasible to navigate through the Arctic Ocean. The Arctic routes shorten the marine transport between the American and Asian-European continents. To enable navigation planning, reliable wave forecasts in the ice covered area is highly demanded. However, as one component of the ocean wave models, the wave-ice interaction modelling is still under development. To obtain good wave forecasts, the effect of all ice types on wave propagation must be modeled correctly.

This dissertation contributes to the wave-ice interaction modelling for general sea ice-covered waters. For this purpose, the research questions addressed include investigating a theoretical model that assumes ice covers as a continuous layer of viscoelastic material. The derived dispersion relation contains two parameters associated with the equivalent viscoelastic properties of different ice types. Implementation of this model in an operational ocean wave model is a numerical problem to solve. Parameters in this viscoelastic model require data calibration. Inverse methods are developed using measurements from a recent field campaign to establish a relation among ice types and these theoretical parameters.

Three main questions of this study are answered as the following. 1) To understand the physical nature of ice-water layered system in the viscoelastic model. The wave characteristics are compared with those from developed theories of wave propagation in other layered systems. It concludes that the roots of the dispersion relation are identified as the flexural gravity, pressure, shear, evanescent and Rayleigh-Lamb waves. A wave mode swap phenomenon is also discussed. 2) To solve the numerical issues in applying the model in a global ocean wave model WAVEWATCH III^®. Strategies of determining the dominant wave mode and expediency of the numerical procedure are proposed. The updated ice source module for WAVEWATCH III ^® performs better in accuracy, efficiency and robustness than its predecessor. 3) Inverse methods are applied to calibrate the model using data collected in the western Arctic Ocean, populated predominantly with pancake ice. The calibrated parameters can be used for wave forecasts in fields of the same ice type in the future. Furthermore, a combined laboratory and numerical study is conducted for wave propagating through an array of uniformed floes. The effective rigidity of the cover is explained by the change of elastic strain energy due to the free edges of each floe. An empirical relation is obtained for the effect rigidity in terms of the floe size and other length scales. This relation may be used to estimate the effective rigidity of an ice cover by in situ or remote sensing images. By answering the above questions, this dissertation contributes to the application of a viscoelastic model for wave hindcasts/forecasts in the whole ice-covered waters.

Since its completion in 1910, the Chicago Sanitary and Ship Canal (CSSC) has become a pathway for invasive species (and potentially Asian carp) to reach the Great Lakes. Currently, an electric barrier is used to prevent Asian carp migration through the canal, but the need for a secondary method is necessary, especially when the electric barrier undergoes maintenance. The underwater Asian carp “cannon” (water gun) provides such a method. Analysis of the ground movement produced by an 80 in³ water gun in the CSSC was performed in order to establish any potential for damage to the either the canal or structures built along the canal. Ground movement was collected using 3-component geophones on both the land surface and in boreholes. The peak particle velocities (PPVs) were analyzed to determine if damage would be caused to structures located along the canal. Vector sum velocity ground movement along the canal wall was as high as 0.28 in/s (7.11 mm/s), which is much lower than the United States Bureau of Mines (USBM) ground vibration damage threshold of 0.75 in/s (19.1 mm/s), causing no potential for damage to structures along the canal wall. The dominant frequency of ground motion produced by the water gun is primarily above 40 Hz, so the wave energy should attenuate fairly quickly away from the canal wall, with little disturbance to structures further from the wall.

As an alternative to spherical harmonics in modeling the gravity field of the Earth, we built a multiresolution gravity model by employing spherical regularization wavelets in solving the inverse problem, i.e. downward propagation of the gravity signal to the Earth's surface. Scale discrete Tikhonov spherical regularization scaling function and wavelet packets were used to decompose and reconstruct the signal. We recovered the local gravity anomaly using only localized gravity measurements at the observing satellite's altitude of 300 km. When the upward continued gravity anomaly to the satellite altitude with a resolution 0.5° was used as simulated measurement inputs, our model could recover the local surface gravity anomaly at a spatial resolution of 1° with an RMS error between 1 and 10 mGal, depending on the topography of the gravity field. Our study of the effect of varying the data volume and altering the maximum degree of Legendre polynomials on the accuracy of the recovered gravity solution suggests that the short wavelength signals and the regions with high magnitude gravity gradients respond more strongly to such changes. When tested with simulated SGG measurements, i.e. the second order radial derivative of the gravity anomaly, at an altitude of 300 km with a 0.7° spatial resolution as input data, our model could obtain the gravity anomaly with an RMS error of 1 ~ 7 mGal at a surface resolution of 0.7° (< 80 km). The study of the impact of measurement noise on the recovered gravity anomaly implies that the solutions from SGG measurements are less susceptible to measurement errors than those recovered from the upward continued gravity anomaly, indicating that the SGG type mission such as GOCE would be an ideal choice for implementing our model. Our simulation results demonstrate the model's potential in determining the local gravity field at a finer scale than could be achieved through spherical harmonics, i.e. less than 100 km, with excellent performance in edge detection.

The cause for low angle submarine landslide (SML) failures, at slope angles less than 4°, currently cannot be readily predicted using conventional terrestrial sources (i.e. excess pore pressure, weak horizons). Numerous models that have been developed pertaining to mass wasting on continental margins generally fall into two categories: post landslide occurrence (Tsunami wave run-up modeling) on coast lines and core sample description on costal margins. To date, there has been limited research on determining bed failure depth of submarine landslides through modeling. We propose a new method of 2D numerical modeling of rupture surface within unconsolidated sediments using the “Particle in Cell” method in combination with a conservative finite volume scheme. The software is written in Python, using the Numerical Python (NumPy) library to reach compiled-code-like performance. The Particle in Cell method was tested for accuracy, advection, and numerical diffusion. A set of six numerical model simulations are presented in which we investigate the role of material and external properties (i.e. hydraulic diffusivity and sedimentation rate), and geometry in the quest to determine bed failure depth. Through initial modeling simulations, it is confirmed that yield strength, diffusivity and sediment loading all play a role in predicting failure.

The operational ocean wave model needs a sea ice component to simulate the global ocean waves. Current ocean wave models treat ice covered regions crudely. The purpose of this thesis is to provide a unified continuum model for the wave ice interaction. Sea ice is constantly subject to environmental forcing. As a result, its physical appearance and mechanical properties vary dynamically. There are three existing classic wave ice interaction models: viscous layer, mass loading, and thin elastic plate models. Viscous layer models may be used to simulate grease ice, mass loading model is probably suitable for pancake ice, and thin elastic plate model may be used to describe a continuous ice sheet floating in water. This situation means that for different kind of sea ice we need different wave ice interaction models. Recently, a proposed viscoelastic wave ice interaction model synthesized the three classic models into one model. Under suitable limiting conditions this model converges to the three previous models. Based on this new development, the present study expands the viscoelastic model for wave propagation through two connected ice covered ocean regions. By doing so, the complete theoretical framework for evaluating wave propagation through various ice covers may be implemented in the operational ocean wave models. In this thesis, we introduce a three-layer viscoelastic model to include the eddy viscosity in turbulent boundary layer under the ice cover into previous viscoelastic model and the methods to calculate wave reflection and transmission. We also use recent results of a laboratory study to determine the viscoelstic model parameters with an inverse method. The thesis concludes with a numerical procedure for implementing the viscoelastic dispersion relation into the ocean wave model and some ideas of model parameterization.