<p> Given the terrifying potentialities linked to global warming, some have suggested that the only means of abating a worst-case scenario is to double down, so to speak. Geoengineering is the intentional manipulation and augmentation of the global climate system. Critics and enthusiasts have commenced a lively debate around this complex issue, and scenarios have recently emerged as a constitutive practice to confront the uncertainties permeating research, implementation, and prospective governance. Using a synthesis of critical political frames to engage with a range of geoengineered imaginaries, this dissertation employs both textual and practice-based modes of research to argue that there are more dynamic and efficacious means to engage people in thinking through the radical possibilities and postnormal potentialities inherent to geoengineering. Turning to games and deploying play as a modality for experimentation, this dissertation assembles a design for exploring the core themes of the debate and enacting an embodied politics for geoengineering. <i>GeoFutr</i> is an alternative futures-driven gaming platform designed to critique, create, and ultimately contest geoengineered imaginaries.</p><p>

<p> Unstable failure in underground mining occurs when a volume of material is loaded beyond its strength and displaces suddenly. It is recognized on various scales, from small rock bursts to the collapse of pillars or entire sections of a mine. The energy that is released during smaller scale events is manifested through the ejection of material, which can pose a hazard to the safety of miners. Larger scale events generate seismic waves as mine workings are damaged and may entrap miners or terminate production. </p><p> This dissertation focuses on the analysis of unstable failure in an underground room and pillar mining environment. The potential for violent pillar failure is assessed using numerical modeling techniques and a parametric approach to loading conditions and material strength properties. The magnitude of instability is quantified by calculating the release of kinetic energy that occurs as failure progresses in each simulation. </p><p> Fundamental mechanisms associated with the release of kinetic energy are analyzed in a series of finite difference models, and the results are compared with analytical solutions to illustrate the applicability of the energy calculations to increasingly complex modes of failure. Back analyses are performed on two room and pillar mine collapse events from the western United States by constructing large-scale models and reproducing widespread failure. The values of energy released in two-dimensional models are extrapolated by assuming a depth of failure in the third direction, and the total energy values are compared to the documented seismic magnitudes from each collapse through empirical equations. With further development of this numerical modeling approach, energy consideration may be used to study the potential for instability in a wide variety of mining excavations and identify the associated range of hazards.</p>

The Bagdad Mine, an open pit mine located in northwestern Arizona, has experienced a slope failure related to a weak weathered tuff bed. The unit has been altered to a clay-like material in areas of the pit. Overburden pressures have caused the tuff to be squeezed out, and blocks of the overlying basalt unit have toppled forward. The failure has been remedied by reducing the pit slope to about 20°, but this solution is not desirable. Modelling of the slope is done using a general purpose finite element method program. The Drucker-Prager theory is used to model the tuff bed, while the other units are treated as being linearly elastic. Material properties for the three rock units are obtained from a variety of sources. Results suggest that loss of shear strength in the tuff, due to blasting-induced liquefaction, may be the cause of slope failure. Suggestions are made for remedial action.

Although a number of models have been proposed to characterize behavior of geological materials including elastic, plastic, and cyclic loading responses, few constitutive models have been developed for the behavior of fully saturated sands and interfaces including liquefaction under dynamic loading. Such realistic constitutive models play an important role in analyzing and predicting the response and design of soil-structure interaction systems. Also, in the engineering view of the complexity of the material behavior, it becomes necessary to develop and use computer procedure (finite element method) for the analysis and prediction of behavior of geotechnical problems. A general concept, called the disturbed state concept (DSC), that can characterize behavior of geological material is developed in this dissertation for the behavior of saturated sands and sand-steel interfaces. The DSC model is an unified approach and allows hierarchical use of the model for factors such as elastic and plastic strains, damage, and softening and stiffening. The model parameters for a saturated sand and a sand-steel interface are evaluated using data from comprehensive laboratory tests; truly triaxial test device for the saturated sand and cyclic multi-degree-of-freedom device (CYMDOF-P) for the interface. The laboratory test results are also used for the verification of DSC model. In general, the model predictions were found to provide satisfactory correlation with the test results. The DSC model with the foregoing parameters is implemented in a nonlinear dynamic finite element program(DSC-DYN2D). It is used to solve two boundary value problems-an axially loaded pile and a shaking table test-involving the interface behavior for pile and the liquefaction for shake table. A new and highly efficient method is used to determine the liquefaction in saturated materials. This method is based on the changing microstructure of the material and allows identification of liquefaction based on the critical disturbance, Dc which is the point where the curvature of the disturbance function is the minimum. Based on the results of this research, it can be stated that the DSC model is capable of characterizing the cyclic behavior of saturated sands and interfaces and the liquefaction instability under dynamic and earthquake loading.

A cement borehole plug and a crushed tuff/bentonite clay mixture borehole plug were tested insitu in highly welded tuff. The hydraulic performance of the cement plug was evaluated through steady-state and transient hydraulic tests with a hydraulic conductivity in the range of 10⁻¹⁰ cm/s. A crushed tuff/bentonite mixture plug was tested through a steady-state flow test with a measured hydraulic conductivity of 10⁻⁹ cm/s. The plug was installed in a fractured borehole which was grouted to reduce the overall rockmass permeability. Installation procedures were evaluated in the laboratory prior to field installation. Installation of the cement seal with a bailer indicated seal degradation with water present in the borehole. Degradation appeared as piping, both internal and along the interface, and mixing of the cement with the water. Tests on the mixture seal indicated the need for homogeneous placement and adequate compaction to resist internal water piping and channelling.

A simplified version of a model that accounts for distributed damage in brittle materials is adopted. It is then extended for the study of the effects of random initial inhomogeneity on material response. A model case regarding random initial inhomogeneity in a structure (specimen) subjected to uniform static compression is studied by including pre-processing and appropriate modifications to a general two dimensional finite element program. From the numerical calculations obtained, the character of damage growth in brittle materials is then studied and discussed.

A study on the possible use of melted waste thermoplastics to stabilize soils, and produce plastic-soil structural composites is presented. Specifically, the effects of melted thermoplastics on the shear strength, and creep characteristics of sands are examined. An extensive experimental program was conducted to examine the strength and creep characteristics of different plastic-sand compositions. It was found that increasing plastic content in the sand results in stronger, and stiffer soils with negligible creep deformations for loads which are at least 50% of the strength of the composite. It is expected that injection of melted waste thermoplastics in sands can improve significantly the engineering characteristics of foundation, and slope stability of sands. Other potential applications include construction of structural elements, bricks, etc.

The effects of an axisymmetric fluid jet impinging on a rock half-space are examined. A recently developed constitutive model for porous elastic materials, which explicitly provides for the compressibilities of the solid grains and fluid that comprise the material, is reconciled to the model developed by M. A. Biot for soils. An increase in pore fluid pressure is shown to reduce the compression in the rock matrix, displacing the Mohr's circle in the direction of the Griffith failure surface. Shear stress due to the subsequent radial flow of fluid, though small and neglected in all previous work, is shown to have significant effect. Shear stress increases the difference between the maximum and minimum effective principal stresses, enlarging the Mohr's circle toward the failure surface. Accurate predictions of threshold pressures to cause failure were achieved when the component compressibilities and shear stress were taken into account.

Fracture growth under compressive loading is studied using the maximum strain energy release rate criterion by means of both the finite element and the boundary element methods. Although this approach is computationally intensive, it is indispensable for this type of problem because other criteria cannot account for the friction effect on the fracture faces. We use a repulsion scheme to handle the frictional contact constraints on the fracture faces: the interpenetration is eliminated by adjusting the normal compressive force (repulsion), and the friction law is satisfied by modifying the friction resistance at each iteration. Our results explain the fact that a natural fracture under uniaxial compression often grows in its own plane, while an artificial cut grows by means of a kink: the reason lies in the lower friction coefficient on an artificial cut than on a natural fracture. Fracture growth under simple shear and under transtension occurs by a kink and along a smooth, slightly convex trajectory; the computed path is almost identical to the one obtained in the laboratory. Under transpression, fracture also grows by a kink and along a smooth trajectory which is of the opposite convexity than in the previous case, when compression is large. Right-stepping fractures under a left-lateral shearing run away from each other when their centers are more than one fracture length distant; when this is not the case, they turn toward each other. Interaction is thus significant only in this last case. Geologically, our results imply that essentially planar faults may be due to continuing remote compressional stress at about 30$\sp\circ$ to the fault, while abrupt changes in orientation may indicate that the previous stress has been replaced by a remote shear stress. Finally, a convex fault path may indicate simple shear or transtension, whereas a concave one may indicate transpression.

A study of the effects of dam-foundation interaction on the response of earth, rockfill and concrete-faced rockfill dams to obliquely incident P, SV and Rayleigh waves is presented. Emphasis is placed on the effects of the foundation flexibility, the spatial variability of the ground motion and the material nonlinearity. The study is based on a rigorous hybrid numerical formulation that combines the efficiency and versatility of the Finite Element Method (FEM) and the ability of Boundary Element Method (BEM) to account for the radiation conditions at the far field. The developed hybrid method is very powerful, and can be used efficiently to obtain accurate solutions of problems of complex geometry, material heterogeneity and, for time-domain analysis, material nonlinearity. The 2-D frequency-domain formulation is used at first to investigate the linear response of earth and rockfill dams to incident P, SV and Rayleigh waves and the response of concrete-faced rockfill dams to incident Rayleigh waves. Furthermore, the nonlinear time-domain coupled BE-FE formulation is used to investigate the response of earth and rockfill dams to vertically incident SV waves. By accounting rigorously for the energy radiated back into the halfspace, the study demonstrates the dramatic effect of the flexibility of the foundation rock in reducing the overall response of the dam. The effects of the spatial variability of the ground motion across the width of the dam are also shown to be important, but less dramatic than those of the foundation flexibility. Finally, the results from the nonlinear analysis of two different dams, each experiencing various degrees of nonlinearity, have demonstrated the great importance of the material nonlinear behavior on the response of dams subjected to strong ground motion.