Projects

Current Projects

NANIA: Modelling complexity in earthquakes (2004-2008)

The project is funded as part of an EPSRC consortium in novel applications of computing in Complexity. The aim of our component is to elucildate the thermodynamics of earthquake populations, notably the connection between self-organised criticality and the hypothesis of maximum entropy production. Current runs are being made on a cellular automaton for in-plane shear, but future work will concentrate on the problem of fault localisation in antiplane shear.

Damage mechanics (2004-2007)

This NERC studentship aims to characterise the effect of localisation on post-yield damage mechanics. Experiemntal acoustic emission data (location, moment tensor and P-wave velocity inversion) are being used to elucidate the effect of localisation on the effective elsatic properties. The project is a collaboration with the rock phsyics group at the GFZ in Potsdam, and also with the non-destructive testing group in the School of Civil Engineering at Edinburgh.

Effect of clay content on the permeability of sandstone (2002-2006)

This studentship aims to characterise the effect of different clay content on the permeability response under stress and failure. A mean field model for 1-dimensional axial permeability is being developed based on microstructural observations of a siote of tests with control variables of ultimate strain, confining pressure and clay content.

Recently-completed projects

COFFERS Calibration of oilfield faults and fractures from flow rate statistics (2001-2005)

This project was funded by the Industry Technology Facilitator and 8 Oil companies to investigate the potential of direct inversion from flow rate data (production adn injection wells) for the geomechanical and structural response of structurally-complex reservoirs. Our role was to carry out the statistical inversions. Teh results were interpreted and confirmed through deterministic physical modelling by VIPS Ltd and Reservoir Dynamixcs Ltd.

VOLCALERT: Innovation in forecasting volcanic eruptions (2001-2004)

The project is funded by the EU fifth framework, and is co-ordinated by Dr Chris Kilburn of the Benfield Greig Hazard Research Centre at University College London. Our role is to quantify the degree of predictability in precursory signals such as seismic event rate, using analysis of the dynamical phase space, and statistical methods to extract a parsimonious model from the data.

Scaling properties of fluid flow in fractured rocks (1999-2004)

This is a joint project co-ordinated by Professor Philip Meredith of University College London under the NERC "Micro to Macro" special topic. The aim of this project is to produce controlled amounts of microcrack damage in initially virtually 0% porosity micro-granite, in order to measure the transition across the percolation threshold due to microcracking. Thermal cracking will be used as a device for producing controlled amounts of damage. The mechanical and hydraulic properties of the rocks containing different amounts of damage will then be investigated under constant strain rate, and constant stress (creep) conditions. Our role will be to carry out the time-dependent creep tests.

Time-dependent deformation of sandstones (1999-2003)

The aim of this University-sponsored PhD project wAs to determine the rate constants for the deformation of porous sandstones at very low strain rates. Experiments WEre carried out under flow-through conditions to quantify changes in permeability and fluid chemistry during deformation. Samples were monitored using a MISTRAS acoustic emission system to quantify the seismic component of microcrack damage, and the rates of chemical and physical changes compared in order to determine the relevant rate processes, and test hypotheses for deformation rates as a function of strain rate in different chemical environments.

SALTRANS: Methods for assessing salt intrusion and transport in heterogeneous and fractured aquifers (2000-2003)

This project aims to quantify the rates of ingression of salt water due to fluid withdrawal strategies for potable water. The project is funded by the EU fifth framework, and is co-ordinated by Professor Brian Berkowitz of the Weizmann Institute in Rehevot, Israel, and includes industrial end-users of the results.
SALTRANS Home

DGLab-Corinth: Deep Geodynamic Laboratory, Corinth (2000-2003)

This project is part of a cluster of European projects based around drilling directly into active normal faults in carbonate rocks around the Gulf of Corinth, Greece. This is the most actively deforming area of Europe, and the work will complement a surface-based investigation of geodetic and seismic deformation. In this project, co-ordinated from the Institut de Physique du Glope in Paris, controlled tests will be carried out downhole to characterise the stress field and hydraulic regime in and around the fault core. Our role in this project is to quantify the mechanical and hydraulic properties of the recovered core samples in and around the fault. We will also carry out control experiments on more accessible analogue cores from local quarries.
Corinth Rift Lab Home (click on 'DGLab')

FFF-Corinth: Faults, fractures and fluids, Corinth (2000-2003)

As part of the same cluster as the DGLab project, the main aim is to quantify the role of faults and fractures on fluid flow. This project is co-ordinated by Isabelle Moretti of the Institut Francais du Petrol, Paris. Our contribution will be to carry out numerical simulations of the poroelastic response of the earth to transient deformations measured in the pore pressure field downhole, in response to surface strain and seismic events.
Corinth Rift Lab Home (click on '3F')

Fracturing mechanisms: from laboratory experiments to long-term crustal processes (2001-2003)

This project, funded under the European Union Marie Curie Fellowship Scheme, is to develop multiscale cellular automaton computational models for faulting and fracturing processes in the laboratory and in the earth. A new stochastic approach is taken where the geometry of the faulting is not specified a priori by material heterogeneity, but is a truly emergent process that depends on the growth and interaction of individual faults. The method has been applied to produce synthetic earthquake catalogues, fault growth models and, most recently, to explain the evolution of multiple deformation bands.

Geometry and mechanics of faults on Mars (1999-2002)

This project was co-ordinated by Professor Rich Schulz of the Mackay School of Mines, Nevada. The aim was to quantify the geometry of faults on Mars from recent satellite images, and to suggest mechanical explanations for the forms and scaling relationships. In particular Martian fault scarps are very well preserved due to lack of erosion, and show very similar scaling relations to those on Earth, the main difference being the very long maximum length relative to the planetary radius for the Martian faults. Our role was in assisting in the interpretation of the results and in the modelling of the processes involved.

Self-organisation of fluid flow, chemical reactivity and rock strength in porous reservoir rocks (1999-2001)

This grant was provided by NERC under the ROPA scheme, following previous experimental work carried out for Mobil US on the diagenetic properties of carbonate rocks. The aim of the project was to elucidate the processes at work during reactive transport in carbonates, achieved by injecting acidic solutions into porous carbonates (see image on rock physics home page). The acid injection mimics some oil recovery and waste disposal strategies, and also serves as an analogue of a chemical self-sealing process, since the reaction product has a greater molar volume than the original calcite. The width and style of the reaction front depends on the Damkola number Da, a dimensionless measure of the ratio of the flow rate to the reaction rate. At high Da we observe "wormholing" and a loss of integrity of the specimen at the injection end. At low Da we observe the production of a self-sealing reaction front which is eventually breached by hydraulic fracture, marked by a plume in the reaction front.

MUSSLE: Management and understanding of reservoirs through stress-sensitive relative permeabilities (1999-2001)

This Edinburgh Rock Mechanics Consortium project measured and characterised the effects of stress on relative permeabilities of oil-water mixtures. Our role was to assist with the experimental programme at Heriot-Watt University, and to provide a full characterisation of the core samples and pore fluids, using a range of techniques including NMR, MRI, C-T scanning, Ion probe analysis, High-Performance Liquid Chromatography, thin section point counting etc.

Fault diagenesis in sandstone (1996-2000)

The aim of this NERC PhD project was to quantify the reactivity of synthetic fault gouges made up of crushed sandstones. The work programme included three elements: (a) chemical "batch" tests to quantify the reactivity of quartz powders in an agitated solution of water; (b) tests under fault normal compaction, and (c) tests under triaxial strain with a component of shear-enhanced compaction. The results show systematic differences in solubility as a function of particle size, temperature, normal and shear stress. The normal stress experiments produce a lithified rock from the original powder. Chemical analysis of the pore fluid shows systematic changes associated with the dissolution of fine particles and the precipitation of cements. The inferred precipitation is consistent with the observed lithification of the synthetic gouge.

STADIA: Stress and diagenesis as fault sealing mechanisms (1998-2000)

The main aim of this industry-funded project was to elucidate the mechanisms of fault sealing during the cataclastic deformation of porous sandstones. The project is a collaborative effort within the Edinburgh Rock Mechanics Consortium (ERMC - a joint effort between our group and Professor Brian Smart of the Department of Petroleum Engineering, Heriot-Watt University). The work programme involved the measurement of permeability changes in rock core tested under a variety of conditions, including standard 3.8 cm diameter core tested in a Hoek cell, 10 cm diameter core tested in our "big rig", and core tested under direct shear conditions at Heriot-Watt. The results show a distinct difference between the smaller and larger core, with the larger samples producing zones of deformation bands similar to those seen in field outcrop. Microstructural properties of the fault gouge have been used to make accurate predictions of the evolution of permeability during deformation.

SCALFRAC: Scaling properties of fracture systems (1997-2000)

This was a major European Union project aiming to quantify the influence of individual fractures and fracture populations on transport properties of joints. Our role was to (a) produce finite element models of fracture populations growing in heterogeneous media, (b) derive network models for fluid flow and contaminant dispersion in fractures, and (c) measure experimentally the permeability of a mated fracture at very low aperture. The modelling programme exploits the broad bandwidth of scales that can be achieved using large parallel supercomputers on-site, and the experimental programme is based on a new rig designed to fracture rock slabs in a pure mode I tension. The fracture modelling so far has provided a useful explanation for the occurrence for T and Y junctions in joint sets, and the fluid flow modelling an explanation for the occurrence of anomalous hydrodynamic dispersion up to a finite transient time. The laboratory results show distinctly lower permeabilities at low aperture than would be expected from the standard cubic law (Poisueille parallel plate assumption), due to the roughness of the fracture surfaces, even when they are mated.

Long-range correlations in oilfield well pressure and earthquake sequences (1997-2000)

This NERC Connect project, carried out in collaboration with BP, had the aim of quantifying the nature of correlations observed between well pairs and earthquakes at long-range, i.e. further away then might be expected from linear pore pressure or stress diffusion. The results confirmed long-range triggering of earthquake sequences with a correlation length of 10-20 km, and a maximum triggering distance of 150 km. Thus the recent events in Athens and Taiwan are unlikely to be causally related to the Izmit earthquake in Turkey. The oilfield well pressure data also showed correlations up to distances of 20 km, distances far too large to be due to Darcian flow between the relevant injector and producer wells. A first-order non-linear poroelastic model has been developed to explain these observations, producing a non-linear diffusive wave that is a plausible mechanism for poroelastic stress transfer over long but finite distances.

Transmissibility of fault damage zones (1997-1999)

The aim of this project was to quantify appropriate transmissibility multipliers for fault damage zones, using a combination of detailed field mapping, stochastic modelling of fault architecture, and calibration in laboratory tests. The project was carried out under the Teaching Company Scheme (TCS) in collaboration with Alastair Beach Associates (ABA) Limited, a geological consultancy based in Glasgow. A scaling technique was developed for predicting the transmissibility of deformation bands, and validated in the laboratory. The results are being used in the core business of ABA Ltd, and the programme was nominated for a TCS award.