Predicting drilling conditions in advance is of prime importance in oil and gas exploration. A close estimate of predrill pore pressure profiles helps reducing drilling operational risks (e.g. kicks) as well as optimizing drilling operations (e.g. mud weight window, casing design). In addition to undercompaction and other overpressure generation mechanisms, chemical compaction effects can produce overpressure in lithologies subject to high temperature conditions. This work presents a pore pressure model for an exploratory well that reproduces the depositional history of the well including chemical and mechanical compaction. The model has been calibrated and verified by comparison to drilling experience and available logs. Advantages and limitations of the model are presented, as well as its capabilities for pore pressure prognosis for new offset exploratory wells in the region. For such a purpose, the calibrated lithology models have been verified by reproducing the depositional and overpressure generation in two offset wells, obtaining results that compare well with their drilling experience. Finally, the model is used to predict the pore pressure conditions along a proposed new well trajectory. The models also provide estimates of overpressure timing in all the trajectories as valuable information for geologists studying the basin evolution.
Exploratory operations can have major uncertainty in predicting drilling conditions, especially in areas that lack drilling experience or in areas that due to some specific depositional history and conditions depart from the general trend observed in nearby wells. One of such special cases is related to overpressure mechanisms due to chemical diagenesis and compaction related processes, which can play a big role in specific rock fabrics exposed to high-temperature conditions [1, 2].
Predicting pore pressure in advance helps reducing wells operational risks and costs by optimizing drilling operations (e.g. mud weight windows, trajectory analysis and casing design amongst others). Traditional methodologies for pore pressure prediction are based on quasi-empirical estimations of the mechanical compaction disequilibrium that can appear in low permeability rocks (e.g. equivalent depth or Eaton's methods). These methods identify deviations from normal compaction trends using logs like neutron density, compressional velocities (sonic) or/and resistivity, to quantify the amount of overpressure that remains in the formation (e.g. ). Finite element modeling can be also used to reproduce the depositional sequence of sediments and study the evolution of overpressure and thermal maturation in sedimentary basins . Other aspects like buoyancy and centroid effect, increase in compressive stress or chemical compaction amongst others can produce overpressure conditions too . In the particular case of overpressure due to chemical processes, this can be triggered by temperature increases and the chemical composition of the rock, which can induce changes in the porosity and pore-pressure generation mechanisms [6, 7]. Overpressure related to chemical compaction might not correlate well with sonic velocities. In order to account for the combination of mechanical and chemical compaction processes, alternative methods to the classical approach based on normal compaction trends can be used, one of them being numerical modeling.
Bedayat, H. (Louisiana State University) | Hosseini, S. A. (Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas) | Moghadam, M. (Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas)
The mechanical response of the target formations during Carbon geological storage strongly depends to the fluid pore pressure alterations in the formation. The storage formations must have sufficient capacity and could avoid migration of CO2 to the surface. When the highly pressurized CO2 is injected into the geological repository the fluid pressure increases in the formation, which results in changes in stresses and deformation of the medium. From geomechanics point of view, the pressure caused by CO2o injection, should not exceed the formation strength and should not cause activation of existing faults. Therefore, finding a realistic estimation of alteration of stresses during the CO2 injection job, has been subject of several studies in the literature. Most of the analytical methods available in the literature to calculate the stress distribution are well established for single-phase flows, but it requires extension when a second fluid, in this case CO2, is also flowing. In the current work, an approximate analytical model is developed for calculating different stresses caused by injection of CO2 in a saline formation, assuming two phase flow pressure regime. Here, we investigated the stress regime under induced fluid pressure and temperature alterations during the injection time.
Carbon dioxide storage or carbon dioxide sequestration refers to the processes by which captured CO2 is securely stored in deep geologic formations. Carbon dioxide storage in geologic formations includes oil and gas reservoirs, unmineable coal seams, and deep saline reservoirs. It has been reported by The Intergovernmental Panel on Climate Change (IPCC) that the global capacity of deep saline storage sites is more than thousands of gigatons of CO2, which is hundreds of times greater than the annual CO2 emissions from industrial sources [1–3]. Therefore studying the behavior of CO2 when stored in these sites are crucial for the industry.
Coupling between fluid flow and mechanical deformation in porous media plays a critical role in subsurface hydrology, oil and gas operations and seismic activity in the Earth's crust. For carbon capture and storage (CCS) projects, these coupled phenomena determine the interactions between the underground injection of CO2, the geomechanical properties of the reservoir and the stability of existing faults. As a result of the inherent uncertainty that is present in complex geological structures, assessing fault stability and the potential for induced seismicity is a fundamental challenge in any modeling effort for CCS projects. Here we present a formal framework for uncertainty analysis and data assimilation, which relies on a two-way-coupled computational modeling strategy for fluid flow and poromechanics of faults. We first quantify the sensitivity of key earthquake attributes (time of triggering, hypocenter location, and earthquake magnitude) to geologic properties such as rock permeability and coefficient of friction of the fault. We then perform a Bayesian inversion that combines Gaussian Processes with Markov Chain Monte Carlo (MCMC), from which we determine the posterior distribution of the system parameters. We show that this posterior distribution correctly combines information from the synthetic earthquake observations with a priori knowledge about the unknown parameters.
Geologic CO2 sequestration is regarded as a promising technology to prevent rising CO2 concentration in the atmosphere from industrial emissions. While various types of underground geological formations have been considered for permanent storage of supercritical CO2 (IPCC, 2005), deep saline aquifers provide the most attractive option for gigatonne-scale storage, given their capacity and ubiquitous nature (Szulczewski et al., 2012).
A principal concern expressed about the practical implementation of CO2 sequestration in deep geological formations is the difficulty in evaluating potential geomechanical risk and the possibility of generating induced or triggered seismicity through stress perturbations in the underground system (NRC, 2013; Zoback and Gorelick, 2012). Complex computational models that combine multiphase flow and fault poromechanics have been proposed to analyze this type of problems (e.g., Cappa and Rutqvist, 2011; Jha and Juanes, 2014). A fundamental challenge associated to the coupled flow-geomechanics models is the inherent uncertainty associated with model parameters, which stems from the always-difficult description of complex geological structures in the subsurface. In addition, for problems where seismicity could be potentially associated to a CO2 storage project, the problem of data assimilation (i.e., model inversion using the observed seismic data) becomes a challenging task due to the complex characteristics of the seismic waveform signals.
No continuous stress field measures can be made inside the Earth's crust, so we developed a technique to estimate the possible stress field from surface stress data. Stress initialization is the aim of this work. The method presented here couples an inversion technique with a finite element model fed with local data coming from data such as in situ measurements of the stress tensor (leak-off tests, log data…). This method guarantees the stability of the initial stress field with the regional stresses thanks to modeling with a finite element (FE) software. The inversion technique consists in minimizing the difference between the FE model and the measured in situ stress by adjusting the elastic parameters of each geological unit of our model at each iteration of optimization. For this minimization problem we use a regular method which has been successfully tested on synthetic 2D models, and we developed the adjoint-state method by formulating the gradient of the error functional towards its parameters in a constraint optimization form, which is still being tested. The result of the minimization problem is a possible initial stress field which satisfies the in situ measurements.
For monitoring the exploitation of sites, and more specifically for monitoring stress changes due to production, oil companies need to finely estimate the geomechanical properties and the behavior of underground formations for all their activities. To reach this goal, it is necessary to model the exploited site. Geomechanical models, as sophisticated as they might be in terms of behavior law, are useless if the stress field is not well initialized. Stress initialization is the aim of this work.
Clearly no entire stress field measurements can be made inside the Earth's crust, so we developed a technique to estimate an admissible stress field from surface stress data deduced from leak-off tests, log data etc. This technique consists in an inverse problem allowing to match a model to discrete stress measurements.
The inverse problem is a mathematical tool which allows to get indirectly scientific measurements(Ganse, 2007). A classical technique of inverse problem is determining model parameters from observed data (Tarantola, 1987). The method presented here is different, it consists in determining the model full stress field from observed data such as stress measurements and elastic parameters.
The shape of the model is also supposed to be given by properly interpreted seismic data.
Shear slip on the natural fractures is proposed as a viable stimulation mechanism in unconventional geothermal and petroleum reservoirs. Fracture mechanics studies show that the propagation of mechanically closed fractures often involves both mode I and II propagation. Wing cracks are triggered by shear deformation at the tip of the natural fractures. These wing cracks tend to reorient and extend in the direction of maximum compressive stress. In addition to mode I propagation (i.e., wing cracks), natural fractures may also propagate in mode II in a plane approximately parallel to the pre-existing natural fracture. A displacement discontinuity method with Mohr-Coulomb elements is used in this paper to study the response of natural fractures to water injection. Modeling results indicate that the onset of fracture slip occurs when the initial shear stress exceeds the shear strength of the Mohr-Coulomb contact elements. The results show that injection into a single natural fracture may lead to the coalescence of multiple natural fracture which can be regarded as an important advantage of this stimulation technique. However, it was found that network connectivity and fracture coalescence is less likely achievable through simultaneous injection into multiple natural fractures mainly due to the compressive stress shadow in the vicinity of the neighboring fractures caused by the shear slip.
EGS design concept in a number of field projects (Soultz, Desert Peak, Newberry) relied on the conceptual model of permeability increase by slip on natural fractures due to water injection. The conceptual model envisions injection pressures below the minimum in-situ stress to cause slip on the critically-stressed fractures and or induce shear failure of the rock mass. However, Jung (2013) presented a review of the results and observations from a number of EGS experiments and suggested that the hitherto adherence to stimulation by shear or “hydro-shearing” is the main reason for the poor progress in the EGS success. Based on interesting interpretations of a number of phenomena, Jung argued that tensile fracturing and not shear slip or propagation is the main mechanism of stimulation, recommending a return to the conventional stimulation concept. In this paper, we review the concept of wing-crack propagation and show this mechanism is in fact, an integral part of the shear slip stimulation mechanism and that a shear propagation mode is also plausible and can contribute to permeability and MEQ. The process is similar to the shear failure in laboratory triaxial compression tests on rock whereby tensile and shear cracks coalesce to form a macroscopic shear crack or fault across the sample. And although individual tensile cracks do form in the process, the failure is referred to as shear failure.
Swyer, M. W. (AltaRock Energy Inc.) | Cladouhos, T. T. (AltaRock Energy Inc.) | Forson, C. (Washington Department of Geology and Earth Resources) | Czajkowski, J. L. (Washington Department of Geology and Earth Resources) | Davatzes, N. C. (Temple University Department of Earth and Environmental Science) | Schmalzle, G. M. (BOS Technologies LLC)
This study seeks to better understand geothermal energy development risk in Washington State. In this region, crustal stress is dominated by the complex tectonics of the Cascade volcanic arc, and active faulting promotes and sustains geothermal reservoir permeability and provides connection to the postulated heat source. Three prospect-scale sites were selected for Phase 1 of a geothermal play-fairway analysis (PFA); Mount St. Helens seismic zone (MSHSZ), Wind River valley (WRV), and Mount Baker (MB). In Phase 1 of the PFA, heat and permeability potential was modeled from existing and publicly available data which are integrated into a map for each site of geothermal development potential using weighting derived from a multiple experts-opinion approach using an analytical hierarchy process. A heat potential model was created based on the locations of Quaternary volcanic vents, hot springs, Quaternary intrusive rocks, geothermometry, and temperature gradient data. Permeability potential was estimated using three dimensional modeled fault geometries in an elastic half space that slips in response to tectonic crustal stresses estimated from regional strain rates modeled from publicly available Global Positioning System (GPS) velocities. Volcanic deformation at MSHSZ and MB are modeled as Mogi sources of deformation. The resulting permeability potential analysis reveals 1) if faults are acting as fluid conduits or barriers, 2) the portions of faults likely to host fluid flow where slip is promoted and large slip gradients imply damage, and 3) the geometry of adjacent rock volumes that have dense fracture networks due to locally concentrated stresses that provide the porosity and permeability to host a commercially viable reservoir. Geometric fault location uncertainty is explored to determine where improved constraints would significantly alter predicted geothermal potential and thus target new data acquisition planned in Phase 2 of the project.
Geothermal resources require heat, fluid, and permeability. Active faulting and accompanying fractures can supply this permeability, allowing sustained deep circulation of hot water to shallow reservoirs accessible to wells and the high flowrates necessary for commercial electricity production.
This study presents a numerical approach to account for the non-linear clay plasticity on wellbore stability analysis for deep water drilling. The numerical model considers the stress path dependent (anisotropic) strain-hardening/post peak softening behavior, which is typically observed during undrained laboratory testing of clay materials, using the NGI-ADP soil model. A linear-elastic assumption estimated almost no drilling window that did not fit with the field drilling experience. The analysis including the hardening behavior results in a smaller failure zone around the wellbore wall than the linear-elastic perfectly plastic model. When the collapse criterion is defined as an extension of the failure zone that does not trigger a drastic reduction in the shear strength around the well, the proposed approach estimates a possible drilling window for a horizontal well, which was originally estimated as having no drilling window. This study indicates that proper modelling of clay plasticity can provide an efficient solution for deep water drilling of shallow reservoirs.
Shallow clay sections in deep water settings are a complex challenge regarding wellbore stability analysis. Often, clay strength is low compared to pore pressures and the drilled formation behaves under an undrained condition due to relatively low permeability compared to typical drilling times. Consequently, collapse gradients, estimated using an idealized linear elastic or linear elastic perfectly plastic behavior, are commonly estimated as too high .
As illustrated in Figure 1, clay materials show significant non-linear plastic ductility, more than mudstones or stiff shales. When the idealized elastic model is used for the clay material, importance of nonlinear plasticity is ignored. Neglecting plasticity may result in inaccurate wellbore analyses. It is common in the field to observe that drilling through clays or soft shale formations are possible even beyond their elastic limit or the peak failure strain due to non-linear plasticity [2, 3]. Therefore, for an optimum well design in deep-water drilling scenarios, the non-linear plastic behavior of clay around the drilled wellbore should be properly taken into account.
This study investigates the effect of non-linear clay plasticity on wellbore stability by a numerical approach.
The numerical model considers anisotropic strain-hardening and softening behavior of clay, which is typically observed during undrained laboratory testing of clay material, using NGI-ADP soil model . First, a typical plastic behavior of clay while drilling is discussed and compared to laboratory tests. The model is then applied to a deep water drilling design, which initially estimated almost no drilling window using a linear-elastic assumption; a mismatch with a field drilling experience. In the end, applicability of calculated plasticity on the collapse gradient is discussed.
Creating a fracture network by tensile and shear failure in rock, and keeping it conductive with proppants has been the commonly used conceptual model of designing a fracture job. The key to optimizing reservoir stimulation by shear slip is knowledge of how fracture permeability increases with slip and how the resulting unpropped conductivity evolves as the stresses (shear and normal) on the fractures change during production. Such data can then be used in numerical models that consider shear slip and time-dependent unpropped conductivity loss to help design stimulation jobs that are more effective in self-propping, and retain their conductivity, thus, maximizing its benefits of ultimately needing less proppants, additives, and less water (reduced stages and re-frac). Therefore, fracture's deformation properties such as shear strength and friction are needed. In this work, we performed laboratory experiments on several shale samples from the Barnett, Mancos and Pierre, to characterize their geomechanical properties. A multistage triaxial testing strategy was applied on saw-cut jointed shale samples to measure deformation properties, and then Mohr-Coulomb and Barton's shear strength criteria were used to determine shear strength envelope and friction properties.
The mechanics behavior of rock is usually dominated by its discontinuities such as natural fractures, bedding planes, joints and faults, etc. As a result, fractures' deformation and failure properties are significant to engineering practices like wellbore stability and hydraulic fracturing. The natural fracture properties can be measured in two major experiments: direct shear test and triaxial shear test. Direct shear test is widely used to measure the strength and stiffness properties of rock fractures (Goodman, 1976; ASTM D5607-08, 2008), since it is easy to utilize large scale specimens at both laboratory and field site testing. However, the direct shear test has some disadvantages. For example, fractures are sheared without confinements, and the normal stress is limited by relatively low capacity of the shear apparatus, and displacement measurements are usually affected by grout deformation. The triaxial shear test can overcome some of these shortcomings as it can be conducted under confining pressure (Jaeger, 1959; Lane and Heck, 1964; Rosso, 1976; Li et al., 2012). In this test, a cylindrical specimen with an inclined fracture is subjected to a given confining pressure, and then the differential stress (σdiff = σ1- σ3) is increased until slip on the weakness plane is initiated. Thus, the fractures' deformation properties under confining pressure can be estimated from stress and displacement measurements.
MOIL Limited (formerly Manganese Ore India Limited), a premiere manganese ore mining company of the Government of India, is producing annually around 1.14 million tonnes of manganese ore. It operates 7 underground and 3 opencast mines. The underground mines are operated at shallow to moderate depth. Asia's largest and oldest underground manganese mine is being operated by MOIL at Balaghat. Continuous rock mechanics investigations and instrumentation has enabled to change the safety and productivity in cut and fill mining. The passive timber square set support and manual back filling of waste rock has been replaced with advanced techniques of pre-mining rock reinforcement by Cable bolting and hydraulic sand stowing operation in all the underground mines of the MOIL. The panel operations have started with phased mechanization. The side discharge loaders have been introduced for mechanical handling of run of mine rock in the stope and load, haul and dump machines for drift development. Further rock mechanics investigations changed the level interval from 30 m to 45 m at Balaghat Mine. Moreover, a single-boomer electro hydrostatic crawler mounted drill jumbo has been introduced for drift development. This paper presents the rock mechanics investigation in CAF stope that have helped to improve the safety standards and underground face productivity from 2.5 to 9.0 metric tons output per man shift by pre-mining support and underground mechanization in the mines of MOIL.
Manganese ore deposits in India have been exploited well over one hundred years. These deposits are mainly of secondary origin and are associated with the older Achaean meta-sediments (Murthy et al, 2009). The deposits in India were originally classified as three fold (Fermor, 1909), which were subsequently modified as fourfold (GSI News 1973; Khrishnaswamy, 1979) as mentioned below:
The manganese deposits in central India are unique in respect of their formation (Mahadik, 2000). The ore body is basically a sedimentary deposit subject to various types of geological disturbances. This area has the oldest meta-sedimentary deposit named as Sausar Group of pre Cambrian age covering all 10 mines of MOIL Limited from East - Ukwa mine to West – Gumgaon mine. It is producing annually around 1.14 million tonnes of manganese ore. The exploitation of ore bodies with weak host rock masses, as found in the manganese deposit in central India is a difficult and challenging task. All the 7 underground mines of MOIL operating at shallow to moderate depth are practicing cut and fill (CAF) stoping with various support systems like pack pillar and timber square set with manual back filling of waste rocks. This system of supporting does not reinforce the strata prior to mining.
Rock behavior frequently does not fit the classical theory of continuum mechanics because of rock heterogeneous structure. Particularly, rock failure may be accompanied by zonal disintegration formation.
The key to building the non-classic model of rock failure is the structure with grain boundaries. Deformations of solid bodies with microscopic flaws can be described within the scope of non-Euclidean geometry, and non-trivial deformation incompatibility can be referred to as a fracture parameter.
The new continuum model presented in this paper enables prediction of the fractured zones initializing and developing as a periodic structure. The non–Euclidean description of phenomenon initiates an appearance of two unusual material constants that can be called ‘non-classic’ moduli. The coupled model must comprise the fourth–order parabolic equation on disintegration thermodynamic parameter to be solved with the classical hyperbolic system of equations for the dynamics of continuous media.
In the classic approach to the rock cutting problem, the fracture is caused by critical shear stresses when failure criterion based on Tresca (or more general Mohr) theory is applied. In this case it is hard to predict the chip size and optimal depth of cut because the criterion is beyond the model.
In this paper, the developed theory of zonal disintegration is applied to analysis of rock cutting by PDC cutter process. It is shown that initial shear stress at the cutter tip causes the disintegration (destruction) developing deep into the medium with contouring of the rock failure scale; and along the cutting direction with the forming of rock failure scale as well.
The difficulties of mathematical description of the rock cutting process, as it is emphasized in Cherepanov , arise from the elongated plastic zones around the cutting edge of a cutter and the interaction of multiple fractures generated from cutting. Cutting the rock formation with an apparent shear plane typically occurs under the brittle mode with cuttings formed as rock particles. The size of the rock particles is characteristic for the cutting process. The main experimental results of the cutting process are well studied (e.g. see Borisov ,) and are typically reduced to describing the key patterns of the transitional process leading to the steady mode, as well as to characteristic features of the steady mode. On the other hand, the steady cutting mode is characterized with the measureable force of resistance to cutting, including cases when it is a function of the linear velocity of the cutter (Cherepanov ). Any theoretical model of the rock cutting process under the brittle fracturing mode proposed must confirm the experimental results. Adding new characteristic features of the cutting process does not bring more clarity to the physics of this phenomenon. The main problem of the rock cutting process may be reduced to finding the characteristic scale of formation fracturing under the conditions of a load applied to the cutter (weight on bit and torque on bit), taking into account the features of the stressed state of the formation in the vicinity of the cutter tip. Another issue that must be clarified in order to optimize the drilling process is an optimal cutting depth at a given cutter velocity. The presented study is an attempt to answer these two questions.