PALEO-FLUID FLOW INDICATORS Peter Eichhubl

Stanford Rock Fracture Project Stanford University, Stanford, CA 94305

e-mail: eichhubl@pangea.stanford.edu

Abstract Criteria are described that allow reconstruction

of paleo-fluid flow direction and dynamics in exhumed flow systems, emphasizing indicators that can be mapped in the field based on outcrop and hand sample observations. In fractured rock, direction and velocity of fluid flow may be derived based on composition and size of entrained rock fragments. Layering of internal sediment and the entrapment of rock fragments in breccia and fracture constrictions may provide information on flow direction. In porous rock, flow direction may be inferred based on the shape of alteration fronts consisting of lobes and cusps. Other direction indicators for porous media flow are elongate concretions, oxidation and reduction tails, and asymmetric alteration haloes around preferred fluid conduits such as joints and faults. The applicability of these criteria for reconstructing regional flow patterns and conditions depends on their availability and thus on rock type, depositional, tectonic, and hydrologic conditions. Oxidation/reduction patterns in red beds are particularly well suited for natural analogue studies of fluid flow in faulted and fractured sandstone reservoirs. Introduction

Understanding processes of fluid flow in the subsurface is of broad interest in economic and petroleum geology, in shallow and deep groundwater hydrogeology, and in structural mechanics. Subsurface flow processes can be investigated through: 1.) in-situ monitoring of flow through pump tests and surface or borehole-based geophysical and geochemical methods, 2.) physical and numerical modeling, and 3.) outcrop studies of exhumed paleo-flow systems. While all three approaches provide complementary information, outcrop studies can provide information that is difficult to gain from in-situ and modeling experiments. Specifically, the paleo-hydrologic approach allows assessment of structural effects on fluid flow on a sub-seismic or outcrop scale in addition to the regional scale. Sub-seismic structural heterogeneities can affect flow properties significantly even on a regional scale (Antonellini and Aydin, 1994; Matthai et al., 1998; Taylor et al., 1999; Jourde et al., 2001), especially when they form sets or arrays. In addition, paleo-hydrologic studies are well suited to address the temporal evolution of flow systems over time scales

that may be significant in hydrocarbon migration studies, long-term reservoir performance, and environmental impact assessments of groundwater flow and contamination.

This study is aimed at providing criteria that allow reconstruction of flow direction and flow rate in exhumed paleo-flow systems. Emphasis is placed on criteria that can be observed in the field and in hand sample as opposed to methods that require extensive sampling for compositional analyses. Several of the criteria are also expected to be applicable to fluid migration studies that are based on core samples. Examples presented in this study come from a variety of locations, with several examples taken from red sandstone of Jurassic age exposed in Valley of Fire State Park SE Las Vegas, Nevada, and in the Moab-Arches National Park area in Utah.

Paleo-flow indicators will be discussed separately for flow in fractured and porous rock. Despite this separation, it is recognized that flow in many fluid reservoirs is a result of both porous-media and fracture flow. While discussion of these paleo-flow indicators addresses primarily their usefulness in inferring direction and rate, a complete reconstruction of paleo-fluid flow regimes has to include information on fluid composition as obtained from fluid inclusion analyses and mineralogical, elemental, and isotopic analyses of cements and alteration products, as well as information on relative and absolute timing of fluid flow. Description of such an integrated approach to paleo-fluid flow reconstruction is beyond the scope of this study.

Indicators for fluid flow in fractured rock

Large volumes of fracture or fault cement require advective mass transfer and are thus direct evidence for fluid flow along fracture or fault conduits (Cox et al., 1986; Eichhubl and Boles, 2000a). The composition of fluid inclusions and fracture or fault cement may indicate the fluid source, e.g. by entrapment of hydrocarbon inclusions and by the stable and strontium isotopic composition of cement with possible inferences about parent fluid composition. In addition to fluid compositional criteria, several macroscopic to microscopic textural features can be applied to constrain the direction and dynamics of fluid flow.

Stanford Rock Fracture Project Vol. 12, 2001 C-1

Provenance of fluidized material in faults and clastic dikes

Clastic or sandstone dikes are common vestiges of fluid flow that may indicate direction and distance of flow by association of the injected material with an underlying source unit, either through an obvious spatial connection or by compositional comparison of injected and source material (Fig. 1a) (Newsom, 1903; Thompson et al., 1999). In hydrothermal and epithermal settings, breccia and pebble dikes have been documented to extend over several hundred meter in height, with a clearly recognizable source bed for the entrained fragments (Farmin, 1934; Morris and Lovering, 1979; Holmes and Kennedy, 1983). Provided the upward transport direction of clastic material can be ascertained based on its composition, the size of the entrained fragments provides a minimum estimate of fluid flow velocity through an empirical fluidization equation (McCallum, 1985; Brown, 1990; Nichols et al., 1994; Eichhubl and Boles, 2000b). In contrast to the well known Stokes equation, the fluidization equation accounts for particle interaction in dense slurries. The fluidization equation quantifies the minimum fluid velocity necessary to get sediment into suspension. A higher velocity will be necessary for upward transport of suspended clastic material. Despite providing only a minimum estimate of flow rate, this approach is largely independent of assumptions about flow conduit geometry (Eichhubl and Boles, 2000b). The fluidization equation is not applicable to poorly sorted clastic material for which an effective mean diameter cannot be defined (Gaskell, 1992).

Stratification and sorting of internal sediment

Deposits of rock fragments and internal sediment in fracture and fault openings have been described in Mississippi Valley-Type lead-zinc and other fracture-controlled base metal deposits (Sass-Gustkiewicz et al., 1982; Sangster, 1988; Tschauder et al., 1990; Qing and Mountjoy, 1994), in hydrothermal systems (Hsu, 1983), and in hydrocarbon systems (Kerans, 1988; Eichhubl and Boles, 1998, 2000b). Stratification and sorting would be indicative of fluidized entrainment and transport of sediment and excludes cavity infill by cavity collapse and gravitational settling of fragments. Grain size of entrained sediment can provide estimates of flow velocity based on the fluidization equation. Cross-bedding of laminated internal sediment may indicate the flow direction in a fracture network (Fig. 1b) (Sass-Gustkiewicz et al., 1982; Sangster, 1988).

Kinetic sieving

Rock fragments may be entrapped in constrictions of fracture systems or between larger

breccia components during fluidized transport, a process that is referred to as kinetic sieving in material processing. The direction of fluid flow can be inferred based on the arrangement of entrapped fragments with respect to the fracture geometry (Fig. 1c).

“Snow-on-roof” texture

Detrital fines or precipitate that forms suspended in fluid may settle on the bottom of fracture cavities or on top of breccia fragments. Such deposits, also referred to as “snow-on-roof” texture, are common in Mississippi Valley-Type deposits (Sangster, 1988). Barton et al. (1971) inferred flow rates of ore-forming fluids at the Creede mining district based on the size of settled hematite flakes. Lateral fluid flow through fracture and breccia systems may accumulate smaller rock fragments along the side of larger fragments (Fig. 1d).

Indicators for fluid flow in porous rock Lobate alteration fronts (roll fronts)

Information on paleo-fluid flow conditions in porous rock may be obtained from mineral reaction or alteration fronts (Fletcher and Hoffmann, 1974; Bickle and McKenzie, 1987; Bickle and Baker, 1990; Blattner and Lassey, 1990; McCaig et al., 1995). Alteration fronts represent sharp or diffuse boundaries between two rock volumes of different mineral assemblages. These assemblages may or may not be paragenetic or in thermodynamic equilibrium. The term “front” implies that this boundary is non-stationary and migrated through the rock body due to diffusive and/or advective mass transfer. Such alteration fronts need to be distinguished from migrating reaction boundaries due to an increase in temperature such as the conversion of smectite to illite or opal-A to opal-CT and quartz during prograde burial diagenesis of sedimentary sequences.

In basinal settings, the probably best studied examples of alteration fronts that have been attributed to fluid flow are roll fronts in sandstone-type uranium deposits (e.g. Dahlkamp, 1993). Roll fronts are typically lobate, with the convex side pointing in the direction of paleo-flow. Upper and lower boundaries are frequently stratigraphically controlled by over- and underlying units of lower permeability. Roll-front uranium deposits result from redistribution and precipitation of uranium minerals at the tip of the lobe where oxidizing fluids of meteoric origin that infiltrate the formation are in contact with reducing basinal fluids.

Alteration fronts due to precipitation or dissolution of iron oxides are common in the Mesozoic red beds of the Colorado Plateau. These alteration fronts may form by the migration of

Stanford Rock Fracture Project Vol. 12, 2001 C-2

reducing fluids dissolving ion oxides of early-diagenetic red beds (bleaching) or by oxidizing fluids precipitating iron oxides (reddening) in units with primary reducing conditions or in red beds that have been subsequently bleached (Shawe, 1976). In some cases, the spatial association of reddening or bleaching with faults and joints allow unambiguous distinction between oxidizing and reducing fronts (Foxford et al., 1996). Frequently, however, it is not clear a priori if the boundary resulted from the movement of bleaching or oxidizing fluids (Fig. 1d), particularly for boundaries that extend over regional scales and that are not associated with specific faults or joint zones.

Such oxidation and reduction fronts are typically irregular, composed of lobes and cusps (Figs. 1e-f, 2a-b). Irregularities in alteration fronts may result from depositional variation in permeability, with preferred fluid flow across courser grained laminae of cross-bedded sandstone (Fig. 2c). Other irregularities do not follow any apparent heterogeneities in permeability but may form where primary compositional variations locally impede the advancement of a reaction front.

Uranium roll fronts illustrate that reaction fronts are not necessarily perpendicular to the direction of fluid flow, with the top and bottom part of a lobe forming a “lateral” front. The rate of front migration and thus the orientation of the front is a function of advective and diffusive mass transfer and of reaction kinetics. Local compositional homogeneities may not affect the direction and velocity of fluid flow but may retard reaction rates thus impeding the advance of the reaction front. An oxidation front would be impeded by a locally increased organic matter content whereas a reduction front would be impeded by a locally higher content in oxides. Local compositional variations may pin an advancing alteration front leading to the formation of cusps while the remainder of the front continues to advance forming a lobe. Lateral gradients in the rate of front migration may ultimately lead to rotation of the reaction front into parallelism with respect to the flow direction resulting in a lateral front. Similar to lobes of uranium roll fronts, lobes along oxidation-reduction contacts in red beds are thus expected to point in the direction of fluid flow whereas cusps are expected to point in the upstream direction of flow. An advanced stage of cusp formation may result in “ball and pillar” structures (Fig. 2d). The facing direction of lobes and cusps may thus allow distinction between oxidizing and reduction fronts. Lobate reaction fronts may also result from a positive feedback between flow and permeability increase due to the dissolution of pore cement (Lichtner, 1996).

Offset reaction fronts along fluid flow barriers Alteration fronts may provide a general sense of

fluid direction, i.e. within a 180! range with respect to the alteration front, unless clearly developed cusps and lobes provide additional constraints on fluid direction. In the absence of cusps and lobes, the 180! range in flow direction can be narrowed down using offsets of alteration fronts along fluid flow barriers as additional geometric constraint. Possible barriers to fluid flow are small cataclastic faults or deformation bands (Fig. 2e) that are typically lower in permeability by 1-2 orders of magnitude compared to the host sandstone (Antonellini and Aydin, 1994). The resulting disturbance in flow direction and flow velocity would result in offsets of the alteration front along deformation bands (Taylor and Pollard, 2001). Assuming that reaction fronts move at the velocity of the formation fluid (“non-reactive” front), Taylor and Pollard (2001) derived an exact solution for the angular relations between flow direction and the trace of the reaction front both outside and inside deformation bands. Because the migration rate of an alteration front will depend on both flow rate and reaction kinetics, a less exact but probably more realistic approach is based on the conservative assumption that fluid flow necessarily crossed the alteration front and deformation bands offsetting the front from the altered to the unaltered side. A best-fitting regionally averaged flow direction may be determined graphically for multiple sets of deformation bands with variable attitudes.

Offsets in reaction fronts along faults or deformation bands due to an impedance of fluid flow across fault rock have to be distinguished from offsets of reaction fronts due to fault slip. Both cases can be distinguished by comparing the offset of the reaction front with independent markers of fault slip such as offset sedimentary features. In addition, fault slip after formation of an alteration front would cause mechanical mixing of altered and unaltered rock fragments along the fault segment that separates altered from unaltered rock. For faults that offset a reaction front due to decreased fault rock permeability, the alteration front would follow the fault, with the alteration contact being either contained within the fault zone or within the host rock parallel to the fault zone but without mixing of altered and unaltered host rock fragments in the fault zone.

Asymmetric alteration haloes along joints and faults

The asymmetry of alteration haloes around joints or faults may provide directional information on fluid flow within the host rock. Alteration haloes around faults or joints that act as fluid conduits are expected to be symmetric in width if diffusive or advective transport of species into the adjacent host rock takes place at the same rate on both sides of the fluid

Stanford Rock Fracture Project Vol. 12, 2001 C-3

conduit. Provided material properties are the same on both sides of the fluid conduit as would be expected for joints and faults of small displacement, symmetry in alteration width requires that gradients in concentration and in fluid pressure within the adjacent host rock are equally symmetric. Where differences in host rock properties and significant thermal gradients across the fractured rock volume can be ruled out, asymmetric alteration haloes are indicative of either a fluid compositional or fluid pressure gradient across the fluid conduit. In the absence of differences in host rock composition on both sides of the conduit, a fluid compositional gradient within the host rock requires preferred advection of fluid into host rock on one side of the fluid conduit. Asymmetric alteration haloes are thus indicative of preferred advective transport in the direction of the alteration asymmetry due to fluid pressure gradients within the host rock.

An example of asymmetric oxidation haloes adjacent to joints was described by Lore et al. (2001) for organic-rich siliceous mudstone associated with shallow in-situ combustion of natural gas near Orcutt, California. On a quarry face mudstone is preferentially oxidized around joints in directions facing centers of combustion alteration (Fig. 2f). It is inferred that joints acted as conduits for meteoric water or steam which entered the porous siliceous mudstone driven by convection around combustion centers. Reaction of the oxygenated surface water or steam with mudstone caused pyrite and smectite dissolution and hematite precipitation. Gradual depletion in oxidation potential of infiltrated water due to reaction with the organic rich mudstone caused the decrease in hematite precipitation with distance away from the joints.

A seemingly similar pattern of asymmetric oxidation haloes is found adjacent to small faults or deformation bands in Jurassic Aztec sandstone at Valley of Fire State Park in southeast Nevada (Fig. 3a) (Taylor and Pollard, 2001) Unlike the joint system at Orcutt, deformation bands at Valley of Fire have lower permeability compared to the adjacent host rock and thus acted as barriers to fluid flow rather than as flow conduits. The asymmetric alteration pattern is interpreted by deformation bands acting as baffles to fluid flow across the faulted sandstone, with individual deformation bands shielding adjacent oxidized sandstone from reducing fluid and bleaching.

Elongate concretions and cement columns

Mozley and Goodwin (1995) described elongate calcite concretions in an otherwise poorly consolidated sandstone adjacent to the Sand Hill fault in New Mexico (Fig. 3b). Based on the preferred orientation of these concretions parallel to the fault

they attributed their elongate shape to preferred concretion growth parallel to the direction of fault-parallel fluid flow. Compared to the ubiquitous occurrence of spherical concretions elongate concretions appear to be relatively rare. Berner (1968) suggested that the spherical growth of concretions is due to diffusive mass transfer of dissolved ions toward the concretion surface. Spherical concretions are typically also spherical when closely spaced to each other which suggests that mass transport—diffusive or advective—is not the rate limiting step for concretion growth in these cases but rather the rate of mineral precipitation on the concretion surface. Elongate concretions, however, may provide the exception where transport is indeed the rate-limiting factor controlling concretion growth. In a rapidly flowing ground water flow, such transport control would likely lead to enhanced precipitation rate on the concretion side that opposes flow as suggested by Mozley and Goodwin (1995).

Chan et al. (2000) described concretionary columns composed of hematite cement in Navajo sandstone near Moab, Utah. These columns are 10-40 cm in diameter, 1 m in length, and oriented at high angle to bedding (Fig. 3c). A higher concentration of hematite cement along the out perimeter of these concretions forms a more resistant rind. Chan et al. (2000) attributed this columns to flow of meteoric fluids across the sandstone precipitating Fe oxides in contact with reducing fluids.

Oxidation and reduction tails

In association with hematite columns, Chan et al. (2000) described streaks of hematite cement similar to Liesegang patterns on bedding planes of Navajo sandstone (Fig. 3d). These streaks or “comet tails” originate at the columns and point into regionally consistent directions. Based on the interpretation of Chan et al. (2000),. these streaks resulted from remobilization and precipitation of iron oxides around the columns in the direction of ground water flow. Similar streaks were described by Aydin et al. (1999) from Capitol Reef National Park (RFP field trip 1999).

Asymmetric reduction spots or tails around organic debris are common in association with uranium roll fronts. The brownish color of these tails results from the precipitation of the mineral jordisite (MoS2) (Squyres, 1970; Dahlkamp, 1993). Tails point in the direction of fluid flow consistent with the shape of roll fronts.

Mineral coarsening along alteration fronts

Oxidation/reduction fronts in red beds are frequently characterized by a band of locally enhanced reddening in the reduced sandstone immediately behind the front (Fig. 3e). Based on the

Stanford Rock Fracture Project Vol. 12, 2001 C-4

association of some of these fronts with conduits of reducing fluid it appears that increased reddening is characteristic of reduction fronts and may thus be used as a criterion to distinguish reduction from oxidation fronts. The increased reddening is due to an increase in mineral size of hematite. This hematite coarsening is interpreted to result from dissolution of early diagenetic hematite at the reduction front and subsequent reprecipitation behind the front. This remobilization could result from upstream diffusion of Fe2+ away from the reduction front causing local supersaturation with respect to hematite and subsequent precipitation.

Asymmetric deposition of clay and carbonate along deformation bands in sandstone

Clay cement and micritic carbonate are observed preferentially in pore space along one side of some deformation bands, but are absent on the other side (Fig. 3f). This asymmetric distribution may result from preferred cementation on the upstream side of deformation bands that act as low-permeability barriers for flow in the porous host rock. Based on laboratory experiments, Whitworth et al. (1999) explained preferred calcite precipitation on the upstream side of fault gouge by a membrane effect resulting in locally increased Ca and bicarbonate concentrations. Alternatively, higher concentrations of clay and micritic carbonate along deformation bands may result from kinetic sieving of fines that are entrained in moving pore fluid.

Discussion

The usefulness of paleo-flow indicators for fluid flow reconstruction in sedimentary basins depends on their availability which will vary with rock type and hydrodynamic and tectonic setting. For instance, indicators of fluid flow in fractures that are based on the entrainment of particles require fluid flow velocities that are significantly faster (>1 m/day) than characteristic rates at which compacting sediment moves through the pore water column (<1 m/1000 yr). Such high flow rates may be encountered in certain meteoric recharge systems with effective channeling of subsurface flow such as karst and hydrothermal settings. Rapid expulsion of basinal brine requires mechanisms of fluid pressure build-up to above-hydrostatic levels and episodic fluid release (Cathles and Smith, 1983; Eichhubl and Boles, 2000b). In sedimentary basins, such conditions are expected to result from compaction disequilibrium of water-rich sediment, diagenetic dehydration reactions, and hydrocarbon maturation, provided overlying units of low permeability limit the rate of fluid discharge by porous media flow.

Alteration fronts require a chemical disequilibrium between pore fluid and rock and

sufficiently fast reaction kinetics to result in observable dissolution and precipitation patterns. Chemical disequilibrium will be the norm for basinal brine migrating through sedimentary strata of heterogeneous composition. Reaction kinetics of fluid/mineral reactions, on the other hand, are notoriously slow under diagenetic conditions and may prevent the formation of distinct dissolution and precipitation patterns. The kinetic sluggishness of fluid/mineral reactions under diagenetic conditions leads to effective reaction thresholds, however, that are probably required for the formation of distinct alteration fronts (Ortoleva, 1994; Lasaga, 1998). Such reaction thresholds may provide a kinetic switch between reaction and non-reaction resulting in a distinct alteration front as opposed to diffuse reaction boundaries. Because fluid/mineral reaction kinetics depend on flow rate, the formation of alteration fronts that can be used for reconstruction of paleo-flow direction may also depend on specific flow conditions.

The use of alteration fronts for inferring paleo-fluid flow conditions depends on their ease of observation in the field. Unlike isotopic and most other diagenetic reaction fronts, oxidation/reduction fronts can be visually identified in the field providing a record of paleo-fluid flow that can be mapped from outcrop to regional scale. The abundance of such alteration patterns in Mesozoic strata of the southwest United States is due to first-cycle red bed formation under oxic conditions during early burial in continental depositional settings (Walker et al., 1978). Clastic sequences deposited in marine settings usually lack an efficient mechanism that supplies oxygenated fluid into the sediment. However, the patterns described here based on examples of oxidation/reduction fronts in red beds are likely to be similar to reaction fronts produced by other dissolution and precipitation reactions. Reaction fronts other than oxidation/reduction fronts may require extensive sampling for petrographic and compositional analyses and may thus be more difficult to map on a regional scale. Clearly, the ease of recognizing oxidation/reduction fronts in the field make red beds an ideal natural laboratory for paleo-fluid flow studies.

Conclusions

The effect of sub-seismic structures on fluid migration in the subsurface and the temporal evolution of flow systems can be assessed based on paleo-fluid flow studies of exhumed flow systems. The direction of fluid flow in fracture systems can be inferred based on the composition of entrained fragments, cross-bedding of internal sediment, and the entrapment of fragments within constrictions of the fracture system. Flow rates in fracture systems can

Stanford Rock Fracture Project Vol. 12, 2001 C-5

be estimated based on the size of transported rock fragments using an empirical fluidization equation, contingent upon evidence for fragment transport in suspension. Evidence for fluidized transport are sorting of internal sediment, “snow-on-roof” textures, and the upward transport of rock fragments of distinct composition from an underlying source unit.

The direction of fluid flow through porous rock may be reconstructed based on the lobate/cuspate shape of alteration fronts, asymmetric oxidation and reduction tails, and asymmetric alteration haloes around joints and faults. Cementation patterns that provide information on flow direction include elongate concretions and cement columns, and the preferred deposition of pore-filling cements along faults that act as barriers to cross-fault fluid flow.

The availability of paleo-flow indicators depends on rock type and hydrologic and tectonic setting. Continental red beds such as the Mesozoic sequences of the southwest United States, displaying abundant oxidation and reduction features, provide ideal conditions for paleo-flow studies in clastic units.

Acknowledgements

This study benefited from discussions with Frank Spera, Jim Boles, James Elliot, Rick Sibson, Rick Behl, Lans Taylor, and Atilla Aydin. Field assistants were Sue Dougherty, Melissa Morse, and Nick Davatzes. The administrations of Death Valley National Park and Valley of Fire State Park (NV) are thanked for access permission. NSF grant EAR 93-04581 provided initial support. Continued support through the industrial affiliates of the Stanford Rock Fracture Project and DOE grant DE-FG03-94ER14462 are thankfully acknowledged.

References Antonellini, M. A., and Aydin, A., 1994, Effects of faulting

on fluid flow in porous sandstones: Petrophysical properties: AAPG Bulletin, v. 78, p. 355-377.

Aydin, A., Davatzes, N., Flodin, E., and Girbacea, R., 1999, Capitol Wash Stops, in: Fractures and Faults in Folded Rocks: The Waterpocket Monocline and the San Rafael Anticline (Utah): Proceedings of the Stanford Rock Fracture Project, 1999, p. 2-6-2-14.

Barton, P. B. Jr., Bethke, P. M., and Toulmin, M. S., 1971, An Attempt to Determine the Vertical Component of Flow Rate of Ore-Forming Solutions in the OH Vein, Creede, Colorado: Society of Mining and Geology Japan, Special Issue, v. 2, p. 132-136.

Berner, R. A., 1968, Rate of concretion growth: Geochimica et Cosmochimica Acta, v. 32, p. 477-483.

Bickle, M. J., and Baker, J., 1990, Advective-diffusive transport of isotopic fronts: an example from Naxos, Greece: Earth and Planetary Science Letters, v. 97, p. 78-93.

Bickle, M. J., and McKenzie, D., 1987, The transport of heat and matter by fluids during metamorphism:

Contributions to Mineralogy and Petrology, v. 95, p. 384-392.

Blattner, P., and Lassey, K. R., 1990, Transport of stable isotopes, kinetics, dispersive advection, and the ‘isotopic fronts’ of Baumgartner and Rumble (1988): Contributions to Mineralogy and Petrology, v. 105, p. 486-490.

Brown, K. M., 1990, The Nature and Hydrogeologic Significance of Mud Diapirs and Diatremes for Accretionary Systems: Journal of Geophysical Research, v. 95, p. 8969-8982.

Chan, M. A., Parry, W. T., and Bowman, J. R., 2000, Diagenetic Hematite and Manganese Oxides and Fault-Related Fluid Flow in Jurassic Sandstones, Southeastern Utah: AAPG Bulletin, v. 84, p. 1281-1310.

Cox, S. F., Etheridge, M. A., and Wall, V. J., 1986, The Role of Fluids in Syntectonic Mass Transport, and the Localization of Metamorphic Vein-Type Ore Deposits: Ore Geology Reviews, v. 2, p. 65-86.

Dahlkamp, F. J., 1993, Uranium Ore Deposits: Berlin, Springer-Verlag, 460 p.

Eichhubl, P., and Boles, J. R., 1998, Vein formation in relation to burial diagenesis in the Miocene Monterey Formation, Arroyo Burro Beach, Santa Barbara, California, in Eichhubl. P., ed., Diagenesis, deformation, and fluid flow in the Miocene Monterey Formation of coastal California: SEPM Pacific Section Special Publication, v. 83, p. 15-36.

Eichhubl, P., and Boles, J. R., 2000a, Focused fluid flow along faults in the Monterey Formation, coastal California: Geological Society of America Bulletin, v. 112, p. 1667–1679.

Eichhubl, P., and Boles, J. R., 2000b, Rates of Fluid Flow in Fault Systems–Evidence for Episodic Fluid Flow in the Miocene Monterey Formation, Coastal California: American Journal of Science, v. 300, p. 571-600.

Farmin, R., 1934, “Pebble dikes” and associated mineralization at Tintic, Utah: Economic Geology, v. 29, p. 356-370.

Fletcher, R. C., and Hofmann, A. W., 1974, Simple models of diffusion and combined diffusion-infiltration metasomatism, in: Hofmann, A. W., Giletti, B. J., Yoder, H. S., and Yund, R. A., eds., Geochemical Transport and Kinetics: Washington, Carnegie Institution, p. 243-259.

Foxford, K. A., Garden, I. R., Guscott, S. C., Burley, S. D., Lewis, J. J. M,., Walsh, J. J., and Watterson, J., 1996, The Field Geology of the Moab Fault, in: Huffman, A. C., Lund, W. R., and Godwin, L. H., eds., Geology and Resources of the Paradox Basin: Utah Geological Association Guidebook, v. 25., p. 265-283.

Gaskell, D. R., 1992, An Introduction to Transport Phenomena in Materials Engineering: New York, Macmillan, 637 p.

Holmes, R. W., and Kennedy, M. B., 1983, Mines and Minerals of the Great American Rift (Colorado-New Mexico): New York, Van Nostrand Reinhold, 332 p.

Hsu, K. J., 1983, Neptunic dikes and their relation to the hydrodynamic circulation of submarine hydrothermal systems: Geology, v. 11, p. 455-457.

Jourde, H., Aydin, A., and Durlofsky, L. J., 2001, Upscaling permeability of fault zones in porous

Stanford Rock Fracture Project Vol. 12, 2001 C-6

sandstone: from field measurements to numerical modeling: AAPG Bulletin (submitted).

Kerans, C., 1988, Karst-Controlled Reservoir Heterogeneity in Ellenburger Group Carbonates of West Texas: AAPG Bulletin, v. 72, p. 1160-1183.

Lasaga, A. C., 1998, Kinetic Theory in the Earth Sciences: Princeton, Princeton University Press, 811 p.

Lichtner, P. C., 1996, Continuum formulation of multiphase multicomponent reactive transport, in: Lichtner, P. C., Steefel, C. I., and Oelkers, E. H., eds., Reactive Transport in Porous Media: Reviews in Mineralogy, v. 34, p. 1-81.

Lore, J., Eichhubl, P., and Aydin, A., 2001, In-situ combustion of hydrocarbons and its impact on alteration and fracturing of siliceous mudstone, Orcutt Field, California, submitted to Journal of Petroleum Science and Engineering.

Matthai, S. K., Aydin, A., Pollard, D. D., and Roberts, S. G., 1998, Numerical simulation of departures from radial drawdown in a faulted sandstone reservoir with joints and deformation bands, in: Jones, G., Fisher, Q. J., and Knipe, R. J., eds., Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs: Geological Society London Special Publication, v. 147, p. 157-191.

McCaig, A. M., Wayne, D. M., Marshall, J. D., Banks, D., and Henderson, I., 1995, Isotopic and Fluid Inclusion Studies of Fluid Movement along the Gavarnie Thrust, Central Pyrenees: Reaction Fronts in Carbonate Mylonites: American Journal of Science, v. 295, p. 309-343.

McCallum, M. E., 1985, Experimental Evidence for Fluidization Processes in Breccia Pipe Formation: Economic Geology, v. 80, p. 1523-1543.

Morris, H. T., and Lovering, T. S., 1979, General Geology and Mines of the East Tintic Mining District, Utah and Juab Counties, Utah: United States Geological Survey Professional Paper, no. 1024, 203 p.

Mozley, P. S., and Goodwin, L. B., 1995, Patterns of cementation along a Cenozoic normal fault: A record of paleoflow orientations: Geology, v. 23, p. 539-542.

Newsom, J. F., 1903, Clastic Dikes: Geological Society of America Bulletin, v. 14, p. 227-268.

Nichols, R. J., Sparks, R. S. J., and Wilson, C. J. N., 1994, Experimental studies of the fluidization of layered sediments and the formation of fluid escape structures: Sedimentology, v. 41, p. 233-253.

Ortoleva, P., 1994, Geochemical Self-Organization: New York, Oxford University Press.

Qing, H., and Mountjoy, E. W., 1994, Origin of Dissolution Vugs, Caverns, and Breccias in the Middle Devonian Presqu’ile Barrier, Host of Pine Point Mississippi Valley-Type Deposits: Economic Geology, v. 89, p. 858-876.

Sangster, D. F., 1988, Breccia-Hosted Lead-Zinc Deposits in Carbonate Rocks, in: James, N. P., and Choquette, P. W., eds., Paleokarst: New York, Springer, 102-116.

Sass-Gustkiewicz, M., Dzulynski, S., and Ridge, J. D., 1982, The Emplacement of Zinc-Lead Sulfide Ores in the Upper Silesian District—A Contribution to the Understanding of Mississippi Valley-Type Deposits: Economic Geology, v. 77, p. 392-412.

Shawe, D. R., 1976, Sedimentary Rock Alteration in the Slick Rock District, San Miguel and Dolores Counties, Colorado: U.S. Geological Survey Professional Paper, no. 576-D, 51 p.

Squyres, J. B., 1970, Origin and Depositional Environment of Uranium Deposits of the Grants Region, New Mexico: PhD Dissertation, Stanford University, 228 p.

Taylor, W. L., Pollard, D. D., and Aydin, A., 1999, Fluid flow in discrete joint sets: Field observations and numerical simulations: Journal of Geophysical Research, v. 104, p. 28,983-29,006.

Taylor, W. L., and Pollard, D. D., 2001, Estimation of in situ permeability of deformation bands in porous sandstone, Valley of Fire, Nevada: Water Resources Research, in press.

Thompson, B. J., Garrison, R. E., and Moore, J. C., 1999, A Late Cenozoic Sandstone Intrusion West of Santa Cruz, California: Fluidized Flow of Water- and Hydrocarbon-Saturated Sediments, in: Garrison, R. E., Aiello, I. W., and Moore, J. C., eds., Late Cenozoic Fluid Seeps and Tectonics Along the San Gregorio Fault Zone in the Monterey Bay Region, California: AAPG Pacific Section Guide Book, v. GB-76, p. 53-74.

Tschauder, R. J., Landis, G. P., and Noyes, R. R., 1990, Late Mississippian Karst Caves and Ba-Ag-Pb-Zn Mineralization in Central Colorado: Part I. Geologic Framework, Mineralogy, and Cave Morphology, in: Beaty , D. W., Landis, G. P., and Thompson, T. B., eds., Carbonate-Hosted Sulfide Deposits of the Central Colorado Mineral Belt: Economic Geology Monograph, v. 7, p. 308-338.

Walker, T. R., Waugh, B., and Grone, A. J., 1978, Diagenesis in first-cycle desert alluvium of Cenozoic age, southwestern United States and northwestern Mexico: Geological Society of America Bulletin, v. 89, p. 19-32.

Whitworth, T. M., Haneberg, W. C., Mozley, P. S., and Goodwin, L. B., 1999, Solute-Sieving-Induced Calcite Precipitation on Pulverized Quartz Sand: Experimental Results and Implications for the Membrane Behavior of Fault Gouge, in: Haneberg, W. C., Mozley, P. S., Moore, J. C., and Goodwin, L. B., eds., Faults and Subsurface Fluid Flow in the Shallow Crust: American Geophysical Union Monograph, v. 113, p. 149-158.

Stanford Rock Fracture Project Vol. 12, 2001 C-7

Figure 1. a. Pebble dike at Tintic, UT. Fragments of Tintic quartzite have been transported upward over 600-1800 m (Morris and Lovering, 1979). b.Cross-bedded and graded internal sediment of dolomicrite filling fracture cavity in siliceous dolostone of Miocene Monterey Formation, Santa Barbara, CA. c. Entrapped breccia fragments within interstices of larger breccia fragments, indicative of fluid flow from left to right. Titus Canyon, Death Valley National Park, CA. d. Preferred deposition of smaller rock fragments on the left side of large fragments suggests fluid flow from left to right. Titus Canyon, Death Valley National Park, CA. e. Oxidation (dark)/reduction (light) front in sandstone at Red Rock Canyon State Park W Las Vegas, NV. The direction of fluid flow based on the shape of the front is ambiguous, either oxidizing fluid from left to right, or reducing fluid from right to left. Reduced, unstained islands within reddened sandstone are suggestive of oxidizing fluid moving left to right. f. Lobate/cuspate alteration contact at Valley of Fire State Park, NV. Cusps indicate flow of reducing fluids from right to left.

Stanford Rock Fracture Project Vo. 12, 2001 C-8

a

c

e

b

d

f

Figure 2. a. Lobate/cuspate alteration front at Rainbow Rock, W Moab, UT. The upper reduction lobe is impregnated with hydrocarbon (arrows), consistent with the lobe shape indicating reducing fluids from right to left. b. Lobate/cuspate alteration front near Upheaval Dome, Canyon Lands National Park, UT. Alteration front resulted from reducing fluids moving top to bottom. c. Cross-bedded sandstone adjacent to Moab fault, UT. Coarser grained units are preferentially bleached over fine grained units by fluid moving from right to left, entering the formation from the fault. d. "Ball-and-pillar" structure due to migration of a reduction front from top to bottom. Vicinity of Moab fault, UT. e. Offset alteration front along deformation bands at Valley of Fire State Park, NV. Offset are due to reduction in permeability of fault rock, impeding the migration of the alteration front. f. Asymmetric oxidation haloes adjacent to joints in siliceous mudstone near Orcutt, CA. Oxidizing fluids migrated into the formation along joints and within the matrix from right to left.

a

c

e

b

d

f

Stanford Rock Fracture Project Vo. 12, 2001 C-9

Figure 3.a. Asymmetric oxidation/reduction haloes adjacent to deformation bands at Valley of Fire State Park, NV, interpreted as reduction shadows next to deformation bands by reducing fluids moving left to right. b. Elongate calcite concretions along the Sand Hill fault, NM, indicative of fluid flow along the dip direction of the fault. Peter Mozley for scale. c. Concretionary columns of hematite cement in Navajo sandstone W Moab, UT. Hammer for scale. d. Comet tails around hematite columns, interpreted by Chan et al. (2000) to point on the direction of regional fluid flow. The tail pointing to the upper left overprints and earlier tail pointing to the upper right. e. Coarsening of hematite (arrow) along reduction front in red siltstone. Reduction front migrated from top to bottom as well as left to right, originating from a vertical fault to the left of the image. W Moab, UT. f. Micritic calcite (arrows) deposited along vertical deformation band (middle section of image) in sandstone, Chimney Rock/San Rafael Swell, UT. Preferred calcite deposition indicates fluid flow from left to right. Scale bar is 0.1 mm.

Stanford Rock Fracture Project Vo. 12, 2001 C-10

a

c

e

b

d

f