Seismology [S]

S42A  ACC:09   Thursday

Spatial and Temporal Fault Zone Evolution and Fluid Flow: Integrating Models, Tools, and Observations I


Presiding: A do Nascimento, Universidade Federal do Rio Grande do Norte; R Lunn, Univ. of Strathclyde; Z Shipton, Univ. of Glasgow

S42A-01  

Statistical characterisation of fault zone properties: fault core thickness in deformation band faults.

* Shipton, Z K (zoe.shipton@ges.gla.ac.uk), Department of Geographical and Earth Sciences, Gregory Building University of Glasgow, Glasgow, G12 8QQ, United Kingdom
Lunn, R J (rebecca.lunn@strath.ac.uk), Department of Civil Engineering, University of Strathclyde, Glasgow, G4 0NG, United Kingdom
Bright, A M (aileen.bright@gmail.com), Trinity College, College Green, Dublin, 2, Ireland
Kirkpatrick, J D (james.kirkpatrick@ges.gla.ac.uk), Department of Geographical and Earth Sciences, Gregory Building University of Glasgow, Glasgow, G12 8QQ, United Kingdom

Fault zone architectures, and therefore fluid flow properties, are spatially heterogeneous. For many faults, flow through the fault zone will be controlled by a few tortuous higher-permeability pathways, the locations and frequency of which are governed by the fault zone architecture. Such detailed architectural geometry cannot be resolved for subsurface faults. Consequently, the key to prediction of subsurface bulk fault zone hydraulic properties is a statistical characterisation of the likelihood and frequency of such connected pathways. The fault core thickness of deformation band faults is highly variable, and regions of small (or zero) thickness are likely to play a significant role in fault hydraulics. We demonstrate here that along strike thickness variation can be well described by a spatially correlated random field. The thickness variation along small displacement faults is different from that along larger offset faults, with a transition at a displacement of 1m. This correlates to the development of through going slip-surfaces within the fault core. These data suggest that, at least for deformation band faults, the statistical properties of the fault zones can be linked to the fault zone development. Our data are from a small number of sites in a specific lithology and to enable such statistics to be used to make predictions in subsurface faults, a large number of similar datasets must be gathered and compared.


S42A-02  

Temporal Evolution of Permeability in an Active Normal Fault

* Fairley, J P (jfairley@uidaho.edu), Dept of Geological Sciences University of Idaho, PO Box 443022, Moscow, ID 83844-3022, United States
Zakrajsek, J R (zakr2268@uidaho.edu), Dept of Geological Sciences University of Idaho, PO Box 443022, Moscow, ID 83844-3022, United States

It is widely accepted that the distribution of permeability in a fault evolves over time as a response to cumulative slip, changes in effective stress, and the precipitation and dissolution of minerals in or near the fault plane, but supporting evidence for this belief comes largely from examination of inactive faults exposed at the surface. This situation has recently begun to change, thanks to the efforts of some investigators to monitor seismic activity resulting from perturbations to fluid pressure from fault boundaries (e.g., reservoir filling or emptying) or deep injection. In a complimentary effort, we have been monitoring discharge temperature from 18 fault controlled hydrothermal springs in the Alvord Basin of southeast Oregon, USA, for evidence of permeability-driven changes. In several springs, spectral analysis and cross-correlation of temperature signals provide strong circumstantial support for perturbations to discharge temperatures driven by unstable convection; these events imply permeability within some areas of the fault on the order of 10-11m2, based on modeling studies by earlier investigators. In other cases, we have identified in our data temperature excursions that may result from physical changes in flow path permeability or from other mechanisms such as unstable convection. Although more work is necessary to determine the origin of some of the observed discharge temperature changes, passive monitoring for temporal permeability evolution in active fault zones is a potentially useful technique that can provide data for fault/fluid flow simulations, and complements active perturbation studies currently underway in other fault systems.


S42A-03  

Insights Into the Hydromechanical Properties of Seismogenic Faults From a Study of Fluid- Induced Seismicity

* Talwani, P (pradeep@sc.edu), University of South Carolina; Department of Geological Sciences, 701 Sumter Street, Columbia, SC 29208, United States

The study of seismicity associated with lake-level fluctuations, fluid injections in bore holes, snow melt and anomalous rainfall reveals the following. The seismicity is associated with the diffusion of fluid-pore pressures in saturated, critically stressed fractures with seismogenic permeability, ks (5x10**(-16) to 5x10**(-14) m2). The likelihood of occurrence of seismicity increases with an increase in the excess pore pressure, p, and its time derivative, dp/dt, and decreases with and increase in the rate of fluid flow through the fractures. For fractures with ks, there is miniscule (Darcian) fluid flow, resulting in undrained conditions, and the delayed seismicity is associated with a delayed increase in dp/dt. For fractures where the permeability exceeds ks, there is an increase in the rate of fluid flow, which is nonlinear, resulting in drained conditions, a delayed decrease in dp/dt, and no seismicity. Thus fluid-induced seismicity maps regions of elevated fluid pressures and not regions of fluid flow - an observation relevant to the secondary recovery of hydrocarbons, waste disposal and mining of thermal energy (Hot Dry Rock projects).


S42A-04  

3D modeling of fault-zone architecture and hydraulic structure along a major Alpine wrench lineament: the Pusteria Fault

* Bistacchi, A (andrea.bistacchi@unimib.it), Dipartimento di Geologia, Universita' di Milano Bicocca, Piazza della Scienza 4, Milano, 20126, Italy
Massironi, M (matteo.massironi@unipd.it), Dipartimento di Geoscienze, Universita' di Padova, Via Giotto 1, Padova, 35137, Italy
Menegon, L (luca.menegon@unipd.it), Dipartimento di Geoscienze, Universita' di Padova, Via Giotto 1, Padova, 35137, Italy

The E-W Pusteria (Pustertal) line is the eastern segment of the Periadriatic lineament, the > 600 km tectonic boundary between the Europe and Adria-vergent portions of the Alpine Collisional Orogen. The lithospheric-scale Periadriatic lineament is characterized by a transcurrent polyphase activity of Tertiary age, and is marked by an array of calcalkaline to shoshonitic magmatic bodies. At the map scale, the western edge of the Pusteria line is characterized by a complex network of generally transcurrent brittle fault zones, interconnected by a full spectrum of transtensional and transpressional features related to releasing and restraining bands respectively. An older ductile/brittle sinistral activity can be recognized in some segments of the fault thanks to their relationships with a strongly tectonized Oligocene tonalite/diorite body (Mules tonalitic "lamella"), emplaced along the Pusteria line, and minor related dikes. A late dextral activity involved the whole Pusteria Fault network and is consistent with the Eastward escape of a major lithospheric block of the Eastern Alps towards the Pannonian basin. During its polyphase activity, the fault network developed a complex architecture, showing different kinds of damage and core zones. Here we report the first results of a detailed mapping project in which, in addition to a traditional structural geology work, the spatial distribution of fault rocks in core zones and the degree and characteristics of fracturing (e.g. joint spacing and number of joint sets) in damage zones are taken into account. As regards the quantitative characterization of damage zones, a new description schema, partly inspired by engineering geology classifications, is proposed. The results of this work are implemented in a 3D structural model (developed with gOcad), allowing the study of the complex relationships among the various structural, mechanical and lithological parameters which concur in the development of the fault-zone architecture. Qualitative inferences about the hydrogeological behavior of damage zones (showing different patterns and degrees of fracturing) and of core zones (showing different kinds and thicknesses of fault rocks) are straightforward and have been implemented in a 3D permeability model. The model highlights the architecture of relatively more permeable blocks (usually corresponding to damage zones) and sealing boundaries (e.g. continuous clay-bearing fault gouge horizons). This 3D permeability structure model will provide the basis for future more quantitative studies on the permeability structure of crustal-scale fault zones.


S42A-05  

Seismic Activity Induced by Water Wells Exploring a Fractured Aquifer in the Parana Basin, Brazil.

* Assumpcao, M (marcelo@iag.usp.br), Universidade de Sao Paulo, IAG-USP Rua do Matao 1226 Cidade Universitaria, Sao Paulo, SP 05508-090, Brazil
Yamabe, T H (higashi@fct.unesp.br), UNESP, Faculdade de Ciencias e Tecnologia, Presidente Prudente, SP , Brazil
Barbosa, J R (jroberto@iag.usp.br), Universidade de Sao Paulo, IAG-USP Rua do Matao 1226 Cidade Universitaria, Sao Paulo, SP 05508-090, Brazil
Lopes, A E (afonso@iag.usp.br), Universidade de Sao Paulo, IAG-USP Rua do Matao 1226 Cidade Universitaria, Sao Paulo, SP 05508-090, Brazil
Balancin, L (balancin@iag.usp.br), Universidade de Sao Paulo, IAG-USP Rua do Matao 1226 Cidade Universitaria, Sao Paulo, SP 05508-090, Brazil

Shallow seismic activity, with magnitudes up to 2.9 and intensities V MM, has been observed since 2004 near some deep wells (120-200m deep) drilled in early 2003 to extract water from a fractured aquifer in the Bebedouro rural area. The wells, drilled for irrigation purposes, cross a sandstone layer about 60-80m thick and extract water from a confined aquifer located in fractured zones between basalt flow layers (Cretaceous Serra Geral Formation of the intracratonic Parana Basin). Seismic activity occurred in 2004 and 2005 as swarms of events mostly during the rainy season when the wells were not pumped. During the dry season, when the wells were pumped almost continuously, the activity was very low. A 6-station seismographic network, installed in March 2005, has located more than 1500 micro-earthquakes. The events are less than 1 km deep (most probably within the 0.5 km thick basalt layer) and cover an area roughly 1.5 km x 5 km across. Migration of the epicenters away from the two closest wells (which also have the largest water flow, more than 150 m3/h) was clearly observed in 2005 with a "seismic diffusivity" of about 0.3 to 0.6 m2/s. Some additional wells were drilled in the same area in early 2006 causing a third swarm of activity starting about one month after drilling. This last swarm also showed a clear migration of epicenters with a "seismic diffusivity" of about 1.0 m2/s. Geophysical and geothermal logging of several wells in the area showed that water from the shallow sandstone aquifer enters the well at the top, usually in waterfalls, flows down the wells and feeds the confined fractured aquifer in the basalt layer at the bottom. We propose that the earthquake swarms are induced by pore pressure diffusion in the fractured basalt layer, due to the extra pressure from the surface aquifer, reaching critically pre- stressed areas up to a few km away from the wells. During periods of continuous pumping, the reduction of pore pressure in the confined aquifer shuts down the seismic activity. This is the second clear case of earthquake activity induced by drilling of deep water wells in the Parana Basin. In the period 1977-79 a similar case occurred in Nuporanga, about 80 km away from Bebedouro, with similar hydro- geological characteristics.


S42A-06  

Multiescalar studies of fracturing mechanisms in fluvial-lacustrine basins

* Carreon-Freyre, D (freyre@geociencias.unam.mx), Centro de Geociencias, UNAM, UNAM, campus Juriquilla, Queretaro, QRO 76230, Mexico
Cerca, M (mcerca@geociencias.unam.mx), Centro de Geociencias, UNAM, UNAM, campus Juriquilla, Queretaro, QRO 76230, Mexico
Hidalgo, C (hidalgo@colpos.mx), Colegio de Posgraduado, campus Montecillo, Km. 36.5 Carretera Mex-Texcoco, Montecillo, Mex 56230, Mexico
Hernandez-Marin, M (mhmarin@vt.edu), Dept. of Geological Sciences Virginia Tech, Virginia Tech, campus Blacksburg, Blacksburg, VIR 24060, United States

Fracturing of clayey fluvial and lacustrine deposits has become a major problem in several cities of central Mexico. The available data reveals the coexistence of several factors determining fracturing at different scales. As main factors we analyze the variation in compressibility of sediments causing differential deformation and withdrawal of groundwater causing a drop in pore pressure. Compressibility depends on consolidation, a term that in soil mechanics refers to the expulsion of interstitial water, and provokes volume decrease and land subsidence. Although major volume decrease occurs in the vertical scale, consolidation of silty clayey materials generates also horizontal tensile stresses. Considering that this factor can be determining to the generation of fractures, the deformational conditions of clayey, silty and sandy sequences is analyzed integrating their stratigraphy and mechanical characteristics. A particular emphasis is made in the mineralogical heterogeneity of the clay fraction that can be related to compressibility variations and can generate micro-fracturing by differential deformation. As study case we analyze the mechanical and geological properties of two sedimentary sequences with contrasting hydraulic and mechanical behavior. Our results show that the paleoenvironmental history of sediments can be used to determine a specific type of fracturing. Thus, the fracturing in fluvial lacustrine deposits is not a random phenomenon but is highly dependent of the geological properties of materials.


S42A-07  

Stress Memory in Fluid-Filled Fractures - Insights From Induced Seismicity

Dura-Gomez, I (idura@geol.sc.edu), University of South Carolina; Department of Geological Sciences, 701 Sumter Street, Columbia, SC 29208, United States
* Talwani, P (pradeep@sc.edu), University of South Carolina; Department of Geological Sciences, 701 Sumter Street, Columbia, SC 29208, United States

Detailed studies of reservoir and injection induced seismicity provide an opportunity to study the characteristics of the fractures associated with fluid-induced seismicity. In 1996, we noted that the first three series of earthquakes with M greater or equal than 5.0 in the vicinity of the Koyna reservoir, occurred only when the reservoir levels had exceeded the previous maxima. In subsequent years, three more similar episodes were noted in the vicinity of the Koyna and the nearby Warna reservoir, without a single repetition in the epicentral location. This behavior was similar to Kaiser effect observed in the laboratory. A similar behavior has been observed in many cases of injection induced seismicity. At the Denver arsenal well in the 1960s and the Soultz, France, hot rock site in the 1990s, among others, seismicity only occurred when the differential pressure, (downhole well borehole pressure excess over the ambient natural pressure) reached a threshold value. These threshold values differed for different wells and depths. The seismicity stopped when the differential pressure was lowered below the threshold value. These observations show that the stress memory (associated with Kaiser effect) observed in the laboratory with small samples, is also displayed in nature where the volume of rocks is from hundreds to thousands of cu.km. The fluid-filled seismogenic fractures near the reservoirs and bore wells, associated with fluid-induced seismicity, seems to behave like a finely tuned, sensitive system that "remember" the largest stress perturbation they have been subjected to. Here we present these observations of stress memory in fluid-filled fractures associated with induced seismicity and suggest possible causes.


S42A-08  

Modeling Spatial and Temporal Fault Zone Evolution in Basement Rocks

* Lunn, R J (rebecca.lunn@strath.ac.uk), University of Strathclyde, Department of Civil Engineering University of Strathclyde John Anderson Building 107 Rottenrow, Glasgow, G4 0NG, United Kingdom
Moir, H (heather.moir@strath.ac.uk), University of Strathclyde, Department of Civil Engineering University of Strathclyde John Anderson Building 107 Rottenrow, Glasgow, G4 0NG, United Kingdom
Shipton, Z K (zoe.shipton@ges.gla.ac.uk), University of Glasgow, Department of Geographic and Earth Sciences University Avenue University of Glasgow, Glasgow, G12 8QQ, United Kingdom
Willson, J P (jonnywillson@hotmail.com), Heriot-Watt University, School of the Built Environment Heriot-Watt University Riccarton, Edinburgh, EH14 4AS, United Kingdom

There is considerable industrial interest in assessing the permeability of faults for the purpose of oil and gas production, deep well injection of waste liquids, underground storage of natural gas and disposal of radioactive waste. Deterministic prior estimation of fault hydraulic properties is highly error prone. Faults zones are formed through a complex interaction of mechanical, hydraulic and chemical processes and their permeability varies considerably over both space and time. Algorithms for predicting fault seal potential using throw and host rock property data exist for clay-rich fault seals but are contentious. In the case of crystalline rocks and sand-sand contacts, no such algorithms exist. In any case, the study of fault growth processes does not suggest that there is a clear or simple relationship between fault throw and the fault zone permeability. To improve estimates of fault zone permeability, it is important to understand the underlying hydro-mechanical processes of fault zone formation. In this research, we explore the spatial and temporal evolution of fault zones through development and application of a 2D hydro-mechanical finite element model. The temporal development of fault zone damage is simulated perpendicular to the main slip surface using Navier's equation for mechanical deformation. The model is applied to study development of fault zones in basement rocks. We simulate the evolution of fault zones from pre-existing joints and explore controls on the growth rate and locations of multiple splay fractures which link-up to form complex damage zones. We explore the temporal evolution of the stress field surrounding the fault tip for both propagation of isolated small faults and for fault linkage Results from these simulations have been validated using outcrop data.