Rocks and Fluids: Prediction and Monitoring in Various Geological Settings Including Fault Zones II
Presiding: M M Haney, Colorado School of Mines; B Tichelaar, Shell International Exploration and Production
S43A-01 13:35h
Fluid flow in fault zones: evidence from hydrogeological and geological studies
Many fault zones have strong effects on fluid flow. Fault zones normally consist of two major hydrogeological units: a fault core, primarily made of breccia or gouge, and a fault damage zone, primarily consisting of fractures of various sizes. Active faults commonly have great effects on the transport of crustal fluids. For groundwater, for example, the effects of fault slip during earthquakes include changes in the yield of springs, water table, and stream flow. Similar effects occur in hydrothermal systems. Dramatic changes in hot springs and wells in geothermal fields occurred during two M6.6 earthquakes on strike-slip faults in the South Iceland zone. Similarly, significant changes occurred in the groundwater system associated with the Storagurra reverse fault in North Norway during an M4 earthquake in 1996. When a fault slips during an earthquake, all the pores and small fractures that meet with the slip plane become interconnected so that the fault may suddenly develop a very high hydraulic conductivity. Fluid transport in fault zones is also controlled by the current stress field. This is mainly because fractures are sensitive to changes in the stress field and deform much more easily than circular pores. In many fault zones, the majority of fractures in the damage zone is oriented subparallel to the main fault plane, in which case the current stress field may have strong effects on the permeability of the fault zone. When the maximum principal compressive stress is at a high angle to the fault strike, many fractures in the damage zone tend to close and fluid transport is reduced. When, however, the maximum principal compressive makes a small angle with the fault strike, fractures in the damage zone tend to be open and fluid transport is enhanced. The best evidence for palaeo-fluid flow, particularly in deeply eroded, inactive fault zones, are networks of mineral veins. Here we present field examples of faults and mineral veins in layered sedimentary rocks from the Bristol Channel Basin, UK, and volcanic rocks from the Husavik-Flatey-Fault, an active transform fault in North Iceland. The host-rock lithologies studied include: (1) mudstones (Upper Triassic) with numerous faults and gypsum veins at Watchet, Somerset Coast (Southwest England); (2) limestone and shale layers (Lower Jurassic) dissected by faults with calcite veins near Kilve, Somerset Coast, and at Nash Point, Glamorgan Coast (South Wales); and (3) basaltic lava flows (Upper Tertiary) in the damage zone of the Husavik-Flatey-Fault with numerous veins of quartz, chalcedony and zeolites. In all the study areas, the mineral veins are related to the faults, indicating that geothermal water was transported along the then-active faults into the host rocks. At Watchet, water that got access to anhydrite nodules in the mudstones lead to local fluid overpressure build-up due to the volume change during the transformation to gypsum, resulting in the development of gypsum veins. At Kilve, calcite veins occur almost exclusively in the cores and damage zones of (mostly normal) faults, and there is evidence that the veins were injected as hydrofractures (fractures generated by internal fluid overpressure) from the fault planes into the limestone layers. Similarly, at Nash Point, calcite veins are related to (mostly strike-slip) faults. Some veins were injected into the limestone layers of the fault damage zones directly from the fault planes. In the Husavik-Flatey-Fault, the mineral veins were generated at the time when the damage zone supplied fluids to surface geothermal fields.
S43A-02 13:50h
Developing a Methodology for Measuring Stress Transients at Seismogenic Depth
The dependence of crack properties on stress means that crustal seismic velocity exhibits stress dependence. This dependence constitutes, in principle, a powerful means of studying transient changes in stress at seismogenic depth through the repeat measurement of travel time from a controlled source. While the scientific potential of this stress dependence has been known for decades, time-dependent seismic imaging has yet to become a reliable means of measuring subsurface stress changes in fault-zone environments. This is due to 1) insufficient delay-time precision necessary to detect small changes in stress, and 2) the difficulty in establishing a reliable in-situ calibration between stress and seismic velocity. These two problems are coupled because the best sources of calibration, solid-earth tides and barometric pressure, produce weak stress perturbations of order 102-103 Pa that require precision in the measurement of the fractional velocity change dlnv of order 10-6, based on laboratory experiments. We have thus focused on developing a methodology that is capable of providing this high level of precision. For example, we have shown that precision in dlnv is maximized when there are Q/π wavelengths in the source-receiver path. This relationship provides a means of selecting an optimal geometry and/or source characteristic frequency in the planning of experiments. We have initiated a series of experiments to demonstrate the detectability of these stress-calibration signals in progressively more tectonically relevant settings. Initial tests have been completed on the smallest scale, with two boreholes 17 m deep and 3 meters apart. We have used a piezoelectric source (0.1ms source pulse repeated every 100ms) and a string of 24 hydrophones to record P waves with a dominant frequency of 10KHz. Recording was conducted for 160 hours. The massive stacking of ~36,000 high-SNR traces/hr leads to delay-time precision of 6ns (hour sampling) corresponding to dlnv precision of 3 × 10-6. We find that barometric pressure fluctuations are easily observed in the delay time data with a SNR of 1000. Also, while lower in amplitude, diurnal and semidiurnal solid-earth-tidal components are also observed. We have also conducted preliminary tests at the Richmond Field Facility permits cross-borehole recordings at a distance of 30 m, and depths to 70 m, using the same equipment. The dominant frequency in this case was 1KHz. While only very short time segments have thus far been analyzed, the preliminary data show that we are able to attain the same high precision (dlnv of order 10-6 ) as in the first experiment. The third and most tectonically relevant experiment is being conducted at the Parkfield site of EarthScope's SAFOD drill hole, performing a cross-hole experiment at approximately 2km depth using both the SAFOD pilot hole as the source hole, and a geophone string in the main hole. The cross-hole distance is approximately 400m. Making use of a specially designed 750Hz 18-element piezoelectric source, we expect to obtain stress-induced temporal changes in dlnv along this path, which if confirmed, would demonstrate the ability to measure KPa-level stress variations at near-seismogenic depth.
S43A-03 14:05h
Quantitative analysis of slip-induced dilation with application to fault zone fluid migration
A new quantitative model for calculating the amount of fluid migration during fault slip is developed. This model is based on observations from laboratory rock mechanics experiments of porosity evolution during frictional sliding and geophysical data that demonstrate the existence of propagating slip pulses that experience transient dilation during earthquakes. The nature of dilation during slip is dependent on the frictional behavior, that is whether the fault experiences seismic or aseismic slip. The transient dilation that occurs during seismic slip is more efficient in transporting fluids along the fault than the semi-permanent, but relatively reduced dilation that occurs during aseismic slip. For aseismic faults, fluid migration rates are primarily dependent on the existing (static) permeability structure of the fault rock. For the seismic case, both the static and dynamic (i.e, during seismic slip) permeability are important. Applications of these models to estimate leakage using fault parameters derived from seismically active faults that trap hydrocarbons in the Cuisiana Field, Columbia, and aseismic faults at Eugene Island Block 330 Field, Gulf of Mexico, are grossly consistent with observed hydrocarbon volumes and inferred along-fault migration rates. Results of this model have broad implications for understanding the effects of faults on charge/migration, compartmentalization, and providing permeable pathways for the leakage of reservoir fluids on exploration (geologic) and production time scales.
S43A-04 14:20h
Quantitative Integrated Evaluation in the Mars Basin, Gulf of Mexico
Today's exploitation of hydrocarbons in the Deepwater Gulf of Mexico requires a subtle, sophisticated class of opportunities for which uncertainties must be quantified to reduce risk. The explorer is often faced with non-amplitude supported hydrocarbon accumulations, limitations of seismic imaging, and uncertainty in stratigraphy and hydrocarbon kitchens, all in an environment of still-maturing technology and rising drilling costs. However, many of the fundamental Exploration processes that drove the industry in the past in the Gulf of Mexico still apply today. Integration of these historically proven processes with each other and with new technologies, supported by a growing body of knowledge, has provided a significant new methodology for wildcat and near-field Exploration. Even in mature fields, additional opportunities are seldom characterized by unambiguous attributes of direct hydrocarbon indicators or amplitude support. Shell's Quantitative Integrated Evaluation process relies upon visualization of integrated volume-based stratigraphic models of rock and fluid properties, and by relating these properties to measured and predicted seismic responses. An attribute referred to as the Differential Generalized Attribute, which summarizes the differences between multiple scenario response predictions and actual measured data, can then be used to distinguish likely scenarios from unlikely scenarios. This methodology allows competing scenarios to be rapidly tested against the data, and is built upon proprietary knowledge of the physical processes and relationships that likely drive vertical and lateral variation in these models. We will demonstrate the methodology by showing a portion of the Mars Basin and describing the integrated capability that is emplaced at the Exploration phase, and matured throughout the Appraisal, Development and Production life cycle of a basin discovery.
S43A-05 14:35h
Overpressure generation and episodic dewatering in the Delaware basin, western Texas: The dual nature of a fault zone
This study combines numerical modeling techniques with field data to investigate overpressure development and episodic dewatering processes in the Delaware Basin and the Central Basin Platform (CBP). Low-permeability shales in the eastern Delaware Basin are characterized by elevated pore fluid pressures that are much greater than hydrostatic pressures. Observed geophysical anomalies such as low resistivity, high conductivity, and high seismic transit time are consistent with the presence of fluid-filled, fractured and mechanically weak rocks in the eastern Delaware Basin. Our new geophysical analyses also indicate that the overpressures likely extend further into the shallower western basin in Culberson County near the sulfur mineralization area. The predicted present-day gas window is located within the overpressure zone, suggesting that overpressure may be maintained by active oil-to-gas conversion. A basin hydrology model Basin2 is modified to evaluate the pore pressure increases by oil-gas conversion from the equation of state (EOS) for the CH4-CO2-H2O system. Our modeling shows that source beds in CBP have been rather shallowly buried and did not become thermally mature to generate gas directly. Overpressure and episodic dewatering processes appear to be the most plausible mechanism to move deep-basin hydrocarbons eastward into the CBP and westward into the shallow margin of the Delaware Basin. The model of long-distance fluid migration from the eastern Delaware across the fault zone into the CBP is supported by the geochemical similarities of oils from the two basins. A model that invokes episodic release of overpressured fluids by hydrofracturing processes can simultaneously provide mechanisms for achieving transient overpressure preservation and substantial, episodic fluid movement. The episodic dewatering through hydrofracturing processes thus can better explain the observed overpressure preservation, long-distance fluid migration, and related mineralization and alteration of evaporitic strata in the Permian Basin.
S43A-06 14:50h
Geological Modeling and Property Scaling in Carbonates Based on Petrophysical Logs, Core Analyses, and Sequence Stratigraphy
Properties of carbonates are particularly sensitive to changes in rock fabric over the range of scales from at least 10-6 to 102 meters. We have examined core and well logs from reservoirs and aquifers around the world including the Middle East and Florida. In all cases, the geology per se was important but undervalued in the engineering plans and specifications. We conclude that it is best to use high-resolution stratigraphic models to guide geologic correlation between boreholes using electrical borehole images, NMR logs, core description, and core analysis as well as seismic and pumping-test data if available. As required by the principle of mapping, all data must be georeferenced to a single reference frame. Sampling and observations must be at precise scales and locations. Uncertainties must remain explicit. We must preserve very small angles (ll0.5°) and determine orientations with respect to depositional dip. Statistical distribution of properties requires stationarity that is unlikely in carbonates. We have found that rigorous application of high-resolution sequence stratigraphy is key to solving real engineering problems in carbonates. It depends on accurate observation and interpretation of stratal surfaces and sequence boundaries. While the identifications are best on core, a combination of appropriate borehole logs has proven satisfactory. To extrapolate away from the borehole, we grow three-dimensional forms constrained by observed thicknesses, orientations, facies, accommodation space, and the a priori knowledge of natural occurrence. Viewing data in three-dimensions is advantageous, but the mathematical projection of three-dimensional forms proves preferable. After establishing geometries, we apply simple constitutive relations constrained within facies to produce synthetic elastodynamic logs at the scale appropriate for the sonic logs and various seismic measurements. The relations connect mineralogy, density and saturation to properties including seismic slownesses, elastic moduli, and permeability and their sensitivity to environmental conditions. Details in and among the cores and plugs are crucial to assigning properties correctly to facies within stratigraphic bodies.