Transport model of buoyant metamorphic fluid by hydrofracturing in leaky rock

Abstract Aqueous fluid released in metamorphism is transported upwards from depth to the Earth's surface. I propose a hydrofracturing model for the fluid transport. In the model, fluid is transported by the upward propagation of a two-dimensional vertical fluid-filled crack from a fluid reservoir (e.g. overpressured compartment under a seal) at depth to the Earth's surface; fluid is injected consecutively from the reservoir into the crack at a given (but not necessarily constant) injection rate; some of the injected fluid is lost by infiltration from the crack walls into the surrounding permeable rock. An approximate solution of the crack propagation is obtained using fluid dynamics for turbulent film flow and linear elastic fracture mechanics. The solution shows the transition from a regime in which the excess pressure of the fluid in the reservoir drives the propagation to a regime in which the buoyancy of the fluid in the crack drives the propagation. For example, if the net injection rate of H₂O is 1 m²/s, the regime transition occurs when the vertical crack length becomes 280 m; after the transition, the propagation velocity and average aperture are constant: 21 m/s and 4.8 cm. If the injection rate is lower than a critical value, hydrofracturing cannot be an effective mode for the fluid transport because of the significant fluid loss by infiltration from the crack walls into the surrounding permeable rock. Assuming a fluid-saturated crust with hydrostatic pore fluid pressure, a lower limit can be estimated for the injection rate required to transport H₂O by hydrofracturing without significant fluid loss. For example, the lower limit for transport from a depth of 15 km to the Earth's surface is estimated at 0.2 m²/s if the crustal permeability is 10^-17 m². The lower limit decreases with decreasing crustal permeability.     

Propagation of a hydraulic fracture parallel to a free surface

This paper analyses the plane strain problem of a fracture, driven by injection of an incompressible viscous Newtonian fluid, which propagates parallel to the free surface of an elastic half‐plane. The problem is governed by a hyper‐singular integral equation, which relates crack opening to net pressure according to elasticity, and by the lubrication equations which describe the laminar fluid flow inside the fracture. The challenge in solving this problem results from the changing nature of the elasticity operator with growth of the fracture, and from the existence of a lag zone of a priori unknown length between the crack tip and the fluid front. Scaling of the governing equations indicates that the evolution problem depends in general on two numbers, one which can be interpreted as a dimensionless toughness and the other as a dimensionless confining stress. The numerical method adopted to solve this non‐linear evolution problem combines the displacement discontinuity method and a finite difference scheme on a fixed grid, together with a technique to track both crack and fluid fronts. It is shown that the solution evolves in time between two asymptotic similarity solutions. The small time asymptotic solution corresponding to the solution of a hydraulic fracture in an infinite medium under zero confining stress, and the large time to a solution where the aperture of the fracture is similar to the transverse deflection of a beam clamped at both ends and subjected to a uniformly distributed load. It is shown that the size of the lag decreases (to eventually vanish) with increasing toughness and compressive confining stress. Copyright © 2005 John Wiley & Sons, Ltd.     

Comparing Reservoir and Outcrop Specimens for Mixed Mode I–II Fracture Toughness of a Limestone Rock Formation at Various Conditions

Summary A fracture toughness study was conducted on a limestone rock formation from a petroleum reservoir in Saudi Arabia, and results were compared with those for outcrop specimens from the same geological formation. The objective was to investigate the possibility of using outcrop specimens to estimate the fracture toughness behavior of reservoir rock at in-situ conditions of temperature and confining pressure. The study was made on reservoir specimens from a depth of about 3.5 km, at both ambient and reservoir conditions. Mixed mode I–II fracture toughness at reservoir conditions of high temperature and confining pressure was studied using straight notched Brazilian disk (SNBD) specimens under diametrical compression. Tests were conducted at ambient conditions, at an effective confining pressure (σ₃) of 28 MPa (4000 psi), and at a temperature of 116^°C. The results showed a substantial increase in fracture toughness under confining pressure. Under σ₃=28 MPa, the pure mode-I fracture toughness (K _IC), increased by a factor of about 3.2, and the pure mode-II fracture toughness (K _IIC) increased by a factor of 4.4, compared to those under ambient conditions. On the other hand, K _IC at 116^°C was only 25% more than that at ambient conditions. These results were compared with recent results for outcrop specimens from the same geological formation. The results reveal that outcrop specimens can be successfully used to predict the fracture behavior of reservoir specimens at in-situ conditions, in spite of some differences at ambient conditions. Additionally, fracture toughness envelopes were obtained for reservoir specimens at ambient and high pressure conditions, in both positive and negative regions. Received September 14, 2000; accepted February 22, 2002 Published online September 2, 2002     

Influence of carbon dioxide on coal permeability determined by pressure transient methods

The permeability of coal samples from Pittsburgh Seam was determined using carbon dioxide as the flowing fluid. The confining pressure was varied to cover a wide range of depths. The permeability was determined as a function of exposure time of carbon dioxide while the confining stress was kept constant. The porosities of the coal samples were found to be very low and most of the samples had porosities less than 1%. The permeability of these coal samples was very low—less than 1 μD. Since the objective of this study was to investigate the influence of CO₂ exposure on coal permeability, it was necessary to increase the initial permeability of the coal samples by introducing a fracture. A longitudinal fracture was induced mechanically, and CT scans were taken to ensure that the fracture was present throughout the sample and that the sample was not damaged otherwise during the process. In this study, the permeability of coal was determined by using pressure transient methods. Two types of pressure pulses were used: A-spike and Sine-6 pressure transients. It was first established that the permeability of fractured coal samples did not change with exposure time when an inert gas (Argon) was used as the fluid medium in the experiments. However, the permeability of coal samples decreased significantly when carbon dioxide was used as the fluid medium. This reduction can be attributed to the coal swelling phenomenon. The results show that the permeability reduction in fractured coal samples can be over 90% of the original value, and the exposure time for such reductions can range from 1.5 days up to a week, typically about 2 days under laboratory conditions. The permeability decreased significantly with the increase in confining pressure. The higher confining pressure appears to close internal fractures causing a reduction in permeability.     

The influence of pumping parameters in fluid‐driven fractures in weak porous formations

In this work, we investigate the main pumping parameters that influence a fluid‐driven fracture in cohesive poroelastic and poroelastoplastic weak formations. These parameters include the fluid viscosity and the injection rate. The first parameter dominates in the mapping of the propagation regimes from toughness to viscosity, whereas the second parameter controls the storage to leak‐off dominated regime through diffusion. The fracture is driven in weak permeable porous formation by injecting an incompressible viscous fluid at the fracture inlet assuming that the fracture propagates under plane strain conditions. Fluid flow in the fracture is modeled by lubrication theory. Pore fluid movement in the porous formation is based on the Darcy law. The coupling follows the Biot theory, whereas the irreversible rock deformation is modeled with the Mohr–Coulomb yield criterion with associative flow rule. Fracture propagation criterion is based on the cohesive zone approach. Leak‐off is also considered. The investigation is performed numerically with the FEM to obtain the fracture opening, length, and propagation pressure versus time. We demonstrate that pumping parameters influence the fracture geometry and fluid pressures in weak formations through the viscous fluid flow and the diffusion process that create back stresses and large plastic zones as the fracture propagates. It is also shown that the product of the propagation velocity and fluid viscosity, µv that appears in the scaling controls the magnitude of the plastic zones and influences the net pressure and fracture geometry. These findings may explain partially the discrepancies in net pressures between field measurements and conventional model predictions for the case of weak porous formation. Copyright © 2014 John Wiley & Sons, Ltd.     

Reaction-induced fluid flow in synthetic quartz-bearing marbles

The reaction kinetics and fluid expulsion during the decarbonation reaction of calcite+quartz=wollastonite+CO₂ in water-absent conditions were experimentally investigated using a Paterson-type gas apparatus. Starting materials consisted of synthetic calcite/quartz rock powders with variable fractions of quartz (10, 20, and 30 wt%) and grain sizes of 10 µm (calcite) and 10 and 30 µm (quartz). Prior to reaction, samples were HIPed at 700 °C and 300 MPa confining pressure and varying pore pressures. Initial porosity was low at 2.7–6.3%, depending on pore pressure during HIP and the amount and grain size of quartz particles. Samples were annealed at reaction temperatures of 900 and 950 °C at 150 and 300 MPa confining pressures, well within the wollastonite stability field. Run durations were between 10 min and 20 h. SEM micrographs of quenched samples show growth of wollastonite rims on quartz grains and CO₂-filled pores between rims and calcite grains and along calcite grain boundaries. Measured widths of wollastonite rims vs. time indicate a parabolic growth law. The reaction is diffusion-controlled and reaction progress and CO₂ production are continuous. Porosity increases rapidly at initial stages of the reaction and attains about 10–12% after a few hours. Permeability at high reaction temperatures is below the detection limit of 10^–21 m² and not affected by increased porosity. This makes persistent pore connectivity improbable, in agreement with observed fluid inclusion trails in form of unconnected pores in SEM micrographs. Release of CO₂ from the sample was measured in a downstream reservoir. The most striking observation is that fluid release is not continuous but occurs episodic and in pulses. Ongoing continuous reaction produces increase in pore pressure, which is, once having attained a critical value (P_crit), spontaneously released. Connectivity of the pore space is short-lived and transient. The resulting cycle includes pore pressure build-up, formation of a local crack network, pore pressure release and crack closure. Using existing models for plastic stretching and decrepitation of pores along with critical stress intensity factors for the calcite matrix and measured pore widths, it results that P_crit is about 20 MPa. Patterns of fluid flow based on mineralogical and stable isotope evidence are commonly predicted using the simplifying assumption of a continuous and constant porosity and permeability during decarbonation of the rock. However, simple flow models, which assume constant pore pressure, constant fluid filled porosity, and constant permeability may not commonly apply. Properties are often transient and it is most likely that fluid flow in a specific reacting rock volume is a short-lived episodic process.Editorial responsibility: J. Hoefs     

Monotonic and cyclic characteristics of K0-Consolidated saturated soft clay under a stress path involving a variable confining pressure

In certain field conditions such as offshore projects under wave loads or embankments under traffic loads, both the vertical and horizontal stresses are variable. However, previous investigations rarely considered the variation in horizontal stress. To better understand the characteristics of natural saturated soft clay, a series of monotonic and cyclic triaxial tests with a K₀-consolidation state were carried out under a variable confining pressure (VCP) stress path. The development of axial strain, pore water pressure and effective stress path is analysed. The results show that with the increase in η (the ratio of the variation in the mean effective principal stress to that of the deviatoric stress), the undrained shear strength (q_f) decreases continuously. The pore water pressure generation is slightly improved under a stress path with increasing confining pressure. Based on the test results, a unified formula was established to predict the pore water pressure under VCP stress paths. The unique p–q–e relationship of normally consolidated clay in monotonic VCP triaxial tests was also demonstrated. Under VCP stress paths, the amplitude of the pore pressure increases, and the effective stress path tilts more sharply to the right. Moreover, a unified formula was established that can provide a good reference for predicting effective stress paths under cyclic VCP triaxial tests.

     

Pressure dependence and anisotropy of P-wave velocities in ultrahigh-pressure metamorphic rocks from the Dabie–Sulu orogenic belt (China): Implications for seismic properties of subducted slabs and origin of mantle reflections

The compressional wave velocities (Vp), pressure derivatives (Vp′) and anisotropy (A) of three types of eclogites and country rocks from the Dabie–Sulu ultrahigh-pressure (UHP) metamorphic belt, China, have been measured under confining pressures up to 800 MPa. Type-1 eclogites, which are coarse-grained and subjected to almost no retrograde metamorphism, experienced recovery-accommodated dislocation creep at peak metamorphic conditions (in the diamond stability field). Type-2 eclogites are fine-grained reworked Type-1 materials that experienced recrystallization-accommodated dislocation creep under quartz/coesite boundary conditions during the early stage of exhumation. Type-3 eclogites are retrogressed samples that were overprinted by significant amphibolite facies metamorphism during a late stage of exhumation within the crust. Type-1 eclogites are richer in Al₂O₃ and MgO but poorer in SiO₂ and Na₂O than Type-2 and Type-3 eclogites. Anisotropy of Type-1 and Type-2 eclogites is generally low (<4%) because volumetrically important garnet is elastically quasi-isotropic, while Type-3 eclogites can exhibit high anisotropy (>10%) due to the presence of strongly anisotropic retrograde minerals such as amphibole, plagioclase and mica. The transition of the pressure dependence of velocity from the poroelastic to elastic regimes occurs at a critical pressure (P_c), which depends mainly on the density and distribution of microcracks and in turn on the exhumation history of rocks. The Vp–pressure relationship can be expressed by Vp=a(lnP)²+blnP+c (P≤P_c) and Vp=V₀+DP (P≥P_c), where P is the confining pressure, a and b are constants describing the closure of microcracks below P_c, c is the velocity when P is equal to one (MPa), V₀ is the projected velocity of a crack-free sample at room pressure, and D is the intrinsic pressure derivative above P_c. When data are curve-fit, pressure derivatives and anisotropy as functions of pressure are determined. The average Vp of the eclogites in the linear regime is 8.42+1.41×10⁻⁴P for Type-1, 7.80+1.58×10⁻⁴P for Type-2, and 7.33+2.04×10⁻⁴P for Type-3, where Vp is in km/s and P in MPa. The decrease in V₀ and increase in D from Type-1 to Type-3 eclogites are attributed to a decrease in garnet content and an increase in retrograde minerals. The NE–SW trending, NW-dipping, slab-like high Vp anomaly (8.72 km/s at a depth of 71 km) which extends from the Moho to at least 110 km beneath the Dabie–Sulu region, can be interpreted as the remnant of a subducted slab which is dominated by Type-1 eclogites and has frozen in the upper mantle since about 200–220 Ma. Such relic crustal materials, subducted and preserved as eclogite layers intercalated with felsic gneiss, garnet–jadeite quartzite, marble and serpentinized peridotite, could be responsible for regionally observed seismic reflectors in the upper mantle.     

Self‐similar solution of a plane‐strain fracture driven by a power‐law fluid

This paper analyses the problem of a hydraulically driven fracture, propagating in an impermeable, linear elastic medium. The fracture is driven by injection of an incompressible, viscous fluid with power‐law rheology and behaviour index n?0. The opening of the fracture and the internal fluid pressure are related through the elastic singular integral equation, and the flow of fluid inside the crack is modelled using the lubrication theory. Under the additional assumptions of negligible toughness and no lag between the fluid front and the crack tip, the problem is reduced to self‐similar form. A solution that describes the crack length evolution, the fracture opening, the net fluid pressure and the fluid flow rate inside the crack is presented. This self‐similar solution is obtained by expanding the fracture opening in a series of Gegenbauer polynomials, with the series coefficients calculated using a numerical minimization procedure. The influence of the fluid index n in the crack propagation is also analysed. Copyright © 2002 John Wiley & Sons, Ltd.     

Transient solution for a plane‐strain fracture driven by a shear‐thinning,power‐law fluid

This paper analyses the problem of a fluid‐driven fracture propagating in an impermeable, linear elastic rock with finite toughness. The fracture is driven by injection of an incompressible viscous fluid with power‐law rheology. The relation between the fracture opening and the internal fluid pressure and the fracture propagation in mobile equilibrium are described by equations of linear elastic fracture mechanics (LEFM), and the flow of fluid inside the fracture is governed by the lubrication theory. It is shown that for shear‐thinning fracturing fluids, the fracture propagation regime evolves in time from the toughness‐ to the viscosity‐dominated regime. In the former, dissipation in the viscous fluid flow is negligible compared to the dissipation in extending the fracture in the rock, and in the later, the opposite holds. Corresponding self‐similar asymptotic solutions are given by the zero‐viscosity and zero‐toughness (J. Numer. Anal. Meth. Geomech. 2002; 26 :579–604) solutions, respectively. A transient solution in terms of the crack length, the fracture opening, and the net fluid pressure, which describes the fracture evolution from the early‐time (toughness‐dominated) to the large‐time (viscosity‐dominated) asymptote is presented and some of the implications for the practical range of parameters are discussed. Copyright © 2006 John Wiley & Sons, Ltd.     

Determining Pressure with Daughter Minerals in Fluid Inclusion by Raman Spectroscopy: Sphalerite as an Example

Abstract: Raman frequency of some materials, including minerals, molecules and ions, shifts systematically with changing pressure and temperature. This property is often used as a pressure gauge in high pressure experiments with the hydrothermal diamond anvil cell (HDAC). Since the system of fluid inclusion is similar to that of HDAC, it can also be used to determine the internal pressure of fluid inclusions. Sphalerite is a common daughter mineral. In this study, the frequency shift of the 350 cm?1 peak of sphalerite has been studied from 296 to 523 K and from 0.07 to 2.00 GPa using the HDAC. The global slope of the isotherms (?n350/?p)T is 0.0048 in the studied pressure range. No significant variation of the slopes with temperature has been observed. The correlation between the frequency shift of the 350 cm?1 peak of sphalerite and pressure and temperature is constrained as P=208.33(?np)350+3.13T?943.75. This relationship may be used to estimate the internal pressure of the sphalerite-bearing fluid inclusions.     

Numerical modeling of fluid‐driven fractures in cohesive poroelastoplastic continuum

In this article, we investigate the main parameters that influence the propagation of a fluid‐driven fracture in a poroelastoplastic continuum. These parameters include the cohesive zone, the stress anisotropy, and the pore pressure field. The fracture is driven in a permeable porous domain that corresponds to weak formation by pumping of an incompressible viscous fluid at the fracture inlet under plane strain conditions. Rock deformation is modeled with the Mohr–Coulomb yield criterion with associative flow rule. Fluid flow in the fracture is modeled by the lubrication theory. The movement of the pore fluid in the surrounding medium is assumed to obey the Darcy law and is of the same nature as the fracturing fluid. The cohesive zone approach is used as the fracture propagation criterion. The problem is modeled numerically with the finite element method to obtain the solution for the fracture length, the fracture opening, and the propagation pressure as a function of the time and distance from the pumping inlet. It is demonstrated that the plastic yielding that is associated with the rock dilation in an elastoplastic saturated porous continuum is significantly affected by the cohesive zone characteristics, the stress anisotropy, and the pore pressure field. These influences result in larger fracture profiles and propagation pressures due to the larger plastic zones that are developing during the fracture propagation. Furthermore, it is also found that the diffusion process that is a major mechanism in hydraulic fracture operations influences further the obtained results on the fracture dimensions, plastic yielding, and fluid pressures. These findings may explain partially the discrepancies in net pressures between field measurements and conventional model predictions. Copyright © 2012 John Wiley & Sons, Ltd.     

Erratum: Properties of a pair of fracture networks produced by triaxial deformation experiments: insights on fluid flow using discrete fracture network models

Results of a series of deformation experiments conducted on gabbro samples and numerical models for computation of flow are presented. Rocks were subjected to triaxial tests (σ1 > σ2 = σ3) under σ3 = 150 MPa confining pressure at room temperature, to generate fracture network patterns. These patterns were either produced by keeping a constant confining pressure and loading the sample up to failure (conventional test: CT), or by building up a high differential stress and suddenly releasing the confining pressure (confining pressure release test: CPR). The networks are similar in overall density but differ primarily in the orientation of smaller fractures. In the case of CT tests, a conjugate fracture set is observed with one dominant fracture zone running at about 20° from σ1. CPR tests do not show such a conjugate pattern and the mean fracture orientation is at around 35° from σ1. Discrete fracture network (DFN) methodology was used to determine the distribution of flow and hydraulic head for both fracture sets under simple boundary conditions and uniform transmissivity values. The fracture network generated by CT and CPR tests exhibit different patterns of flow field and hydraulic head configurations, but convey approximately the same amount of flow at all scales for which DFN models were simulated. The numerical modelling results help to develop understanding of qualitative differences in flow distribution that may arise in rocks of the same mineralogical composition and mechanical properties, but under the influence of different stress conditions, albeit at similar overall stress magnitude.

     

Modeling spatial fracture intensity as a control on flow in fractured rock

Spatial fracture intensity (P ₃₂, fracture area by volume) is an important characteristic of a jointed rock mass. Although it can hardly ever be measured, P ₃₂ can be modeled based on available geological information such as spatial data of the fracture network. Flow in a mass composed of low-permeability hard rock is controlled by joints and fractures. In this article, models were developed from a geological data set of fractured andesite in LanYu Island (Taiwan) where a site is investigated for possible disposal of low-level and intermediate-level radionuclide waste. Three different types of conceptual models of spatial fracture intensity distribution were generated, an Enhanced Baecher’s model (EBM), a Levy–Lee Fractal model (LLFM) and a Nearest Neighborhood model (NNM). Modeling was conducted on a 10 × 10 × 10 m synthetic fractured block. Simulated flow was forced by a 1% hydraulic gradient between two vertical x–z faces of the cube (from North to South) with other boundaries set to no-flow conditions. Resulting flow vectors are very sensitive to spatial fracture intensity (P ₃₂). Flow velocity increases with higher fracture intensity (P ₃₂). R-squared values of regression analysis for the variables velocity (V/V _max) and fracture intensity (P ₃₂) are 0.293, 0.353, and 0.408 in linear fit and 0.028, 0.08, and 0.084 in power fit. Higher R ² values are positively linked with structural features but the relation between velocity and fracture intensity is non-linear. Possible flow channels are identified by stream-traces in the Levy–LeeFractal model.     

Evidence from mineral assemblages for infiltration of pelitic schists by aqueous fluids during metamorphism

Prograde mineral assemblages and compositions have been predicted for pelitic schist in the 10 component system Na₂O–K₂O–CaO–MnO–FeO–MgO–Al₂O₃–SiO₂–CO₂–H₂O for three cases of prograde metamorphism and fluid-rock interaction: (1) increasing temperature (T) at constant pressure (P) and constant pore fluid volume (1%) without infiltration (no-infiltration case); (2) increasing T at constant P accompanied by sufficient fluid infiltration that fluid composition is at all times constant (large-flux case); and (3) increasing T at constantP accompanied by a timeintegrated fluid flux f 10⁴ cm^{3 cm 2} (intermediate-flux case). Stable mineral assemblages and compositions were calculated by solving a system of non-linear equations that specify mass balance and chemical equilibrium between minerals and fluid. The model pelitic system includes quartz, muscovite, plagioclasc, chlorite, ankerite, siderite, biotite, garnet, staurolite, andalusite, kyanite, sillimanite, K-feldspar, and a coexisting, binary H₂O–CO₂ fluid. Specifically, prograde thermal metamorphism was modelled for Shaw's (1956) average low-grade pelite and for a moderate range of bulk rock compositions at P=3, 5, and 7 kb and initial fluids with Xco ₂ ^o =0.02–0.40. The model predicts a carbonate-bearing mineral assemblage for average pelite under chlorite zone conditions composed of quartz, muscovite, albite, chlorite, ankerite, and siderite. The mineral assemblages predicted for the noinfiltration case are unlike those typically observed in regional metamorphic terranes. Simulations of metamorphism for the large-flux and intermediate-flux cases, however, reproduce the sequence of mineral assemblages observed in normal Barrovian regional metamorphic terranes. These results suggest that regional metamorphism of pelitic schists is typically associated with infiltration of significant quantities of aqueous fluid.     

Experimental Studies on Permeability of Intact and Singly Jointed Meta-Sedimentary Rocks Under Confining Pressure

Three different types of permeability tests were conducted on 23 intact and singly jointed rock specimens, which were cored from rock blocks collected from a rock cavern under construction in Singapore. The studied rock types belong to inter-bedded meta-sandstone and meta-siltstone with very low porosity and high uniaxial compressive strength. The transient pulse water flow method was employed to measure the permeability of intact meta-sandstone under a confining pressure up to 30 MPa. It showed that the magnitude order of meta-sandstone’s intrinsic permeability is about 10^?18 m². The steady-state gas flow method was used to measure the permeability of both intact meta-siltstone and meta-sandstone in a triaxial cell under different confining pressures spanning from 2.5 to 10 MPa. The measured permeability of both rock types ranged from 10^?21 to 10^?20 m². The influence of a single natural joint on the permeability of both rock types was studied by using the steady-state water flow method under different confining pressures spanning from 1.25 to 5.0 MPa, including loading and unloading phases. The measured permeability of both jointed rocks ranged from 10^?13 to 10^?11 m², where the permeability of jointed meta-siltstone was usually slightly lower than that of jointed meta-sandstone. The permeability of jointed rocks decreases with increasing confining pressure, which can be well fitted by an empirical power law relationship between the permeability and confining pressure or effective pressure. The permeability of partly open cracked specimens is lower than that of open cracked specimens, but it is higher than that of the specimen with a dominant vein for the meta-sandstone under the same confining pressure. The permeability of open cracked rock specimens will partially recover during the unloading confining pressure process. The equivalent crack (joint) aperture is as narrow as a magnitude order of 10^?6 m (1 μm) in the rock specimens under confining pressures spanning from 1.25 to 5.0 MPa, which represent the typical ground stress conditions in the cavern. The in situ hydraulic conductivity measurements conducted in six boreholes by the injection test showed that the in situ permeability of rock mass varies between 10^?18 and 10^?11 m². The lower bound of the in situ permeability is larger than that of the present laboratory-tested intact rock specimens, while the upper bound of the in situ permeability is less than that of the present laboratory-tested jointed rock specimens. The in situ permeability test results were thus compatible with our present laboratory permeability results of both intact and jointed rock specimens.     

Indentation of a free‐falling lance penetrometer into a poroelastic seabed

A solution is developed for the build‐up, steady and post‐arrest dissipative pore fluid pressure fields that develop around a blunt penetrometer that self‐embeds from freefall into the seabed. Arrest from freefall considers deceleration under undrained conditions in a purely cohesive soil, with constant shear strength with depth. The resulting decelerating velocity field is controlled by soil strength, geometric bearing capacity factors, and inertial components. At low impact velocities the embedment process is controlled by soil strength, and at high velocities by inertia. With the deceleration defined, a solution is evaluated for a point normal dislocation penetrating in a poroelastic medium with a prescribed decelerating velocity. Dynamic steady pressures, P_D, develop relative to the penetrating tip geometry with their distribution conditioned by the non‐dimensional penetration rate, U_D, incorporating impacting penetration rate, consolidation coefficient and penetrometer radius, and the non‐dimensional strength, N_D, additionally incorporating undrained shear strength of the sediment. Pore pressures develop to a steady peak magnitude at the penetrometer tip, and drop as P_D=1/x_D with distance x_D behind the tip and along the shaft. Peak induced pressure magnitudes may be correlated with sediment permeabilities, post‐arrest dissipation rates may be correlated with consolidation coefficients, and depths of penetration may be correlated with shear strengths. Together, these records enable strength and transport parameters to be recovered from lance penetrometer data. Penetrometer data recorded off La Palma in the Canary Islands (J. Volcanol. Geotherm. Res. 2000; 101 :253) are used to recover permeabilities and consolidation coefficients from peak pressure and dissipation response, respectively. Copyright © 2004 John Wiley & Sons, Ltd.     

A new fracture criterion for brittle rocks

A general form of a “fracture function” for isotropic brittle materials is expressed in terms of the three invariants of the stress tensor. The coefficients in the function are determined by use of the small number of experimental data under specific conditions. This function is applicable to an estimate of the fracture condition of brittle rocks under a general stress state i.e., σ₁ ≠ σ₂ ≠ σ₃. The application of this function is attempted for the data of three brittle rocks i.e., Dunham dolomite, Mizuho trachyte, and Westerly granite, reported by previous workers. For the first two, this criterion gives a good estimation of the effect of the intermediate principal stress σ₂ on failure. For the last, the fracture strength at high confining pressure is estimated by use of the several data obtained under very low confining pressures, and the agreement with experimental data is also satisfactory.     

Evolution of fluid pressure and fracture propagation during contact metamorphism

Abstract Rock fracture enhances permeability and provides pathways through which fluids migrate. During contact metamorphism, fluids contained in isolated pores and fractures expand in response to temperature increases caused by the dissipation of heat from magmas. Heat transport calculations and thermomechanical properties of water-rich fluids demonstrate (1) that thermal energy is a viable mechanism to produce and maintain pore fluid pressure (P_f) in a contact metamorphic aureole; (2) that the magnitude of P_f generated is sufficient to propagate fractures during the prograde thermal history (cause hydrofracture) and enhance permeability; and (3) that P_f-driven fracture propagation is episodic with time-scales ranging from years to thousands of years. Because P_f dissipation is orders of magnitude faster than P, _f buildup, P_f oscillations and cyclical behaviour are generated as thermal heating continues. The P_f cycle amplitude depends on the initial fracture length, geometry and the rock's resistance to failure whereas the frequency of fracture depends on the rate of heating. Consequently, oscillation frequency also varies spatially with distance from the heat source. Time series of fluid pressures caused by this process suggest that cyclical fracture events are restricted to an early time period of the prograde thermal event near the intrusive contact. In the far field, however, individual fracture events have a lower frequency but continue to occur over a longer time interval. Numerous fracture cycles are possible within a single thermal event. This provides a provisional explanation for multiple generations of veins observed in outcrop. P _f cycling and oscillations may explain several petrological features. If pore fluids are trapped at various positions along a pressure cycle, the large amplitude of P_f variations for small fractures may account for different pressures recorded by fluid inclusions analysed from a single sample. P_f oscillations, during a single thermal episode, also drive chemical reactions which can produce complex mineral textures and assemblages for discontinuous reactions and/or zoning patterns for continuous reactions. These can mimic polymetamorphic or disequilibrium features. Temporal aspects of fracture propagation and permeability enhancement also constrain the likely timing of fluid flow and fluid-mineral interactions. These data suggest that fluid flow and fluid-mineral reactions are likely to be restricted to an early period in the prograde thermal history, characterized by high P_f coincident with relatively high temperatures, fracture propagation and consequent increases in permeability. This early prograde hydration event is followed by diffusional peak metamorphic reactions. This relationship is evident in the complex mineralogical textures common in some metamorphosed rocks.     

Semi-brittle behavior of wet olivine aggregates: the role of aqueous fluid in faulting at upper mantle pressures

The role of aqueous fluid in fracturing in subducting slabs was investigated through a series of deformation experiments on dunite that was undersaturated (i.e., fluid-free) or saturated with water (i.e., aqueous-fluid bearing) at pressures of 1.0–1.8 GPa and temperatures of 670–1250 K, corresponding to the conditions of the shallower regions of the double seismic zone in slabs. In situ X-ray diffraction, radiography, and acoustic emissions (AEs) monitoring demonstrated that semi-brittle flow associated with AEs was dominant and the creep/failure strength of dunite was insensitive to the dissolved water content in olivine. In contrast, aqueous fluid drastically decreased the creep/failure strength of dunite (up to ~?1 GPa of weakening) over a wide range of temperatures in the semi-brittle regime. Weakening of the dunite by the aqueous fluid resulted in the reduction of the number of AE events (i.e., suppression of microcracking) and shortening of time to failure. The AE hypocenters were located at the margin of the deforming sample while the interior of the faulted sample was aseismic (i.e., aseismic semi-brittle flow) under water-saturated conditions. A faulting (slip rate of ~?10^?3 to 10^?4 s^?1) associated with a large drop of stress (Δσ?~?0.5 to 1 GPa) and/or pressure (ΔP?~?0.5 GPa) was dominant in fluid-free dunite, while a slow faulting (slip rate?<?8?×?10^?5 s^?1) without any stress/pressure drop was common in water-saturated dunite. Aseismic semi-brittle flow may mimic silent ductile flow under water-saturated conditions in subducting slabs.