Course Outline
A variety of interesting speakers have been recruited to lecture on topics relating to our broad subject "Fluid Flow in the Earth." The following are brief abstracts highlighting their lecture material.
New thermal evidence for crustal-scale lateral flow and the influence of seamounts: examples from the east flank of the Juan de Fuca Ridge and west flank of the East Pacific Rise by Andy Fisher, UC Santa Cruz.
Basaltic outcrops that penetrate through otherwise thick and continuous sediment on the eastern flank of Juan de Fuca Ridge may be important for the exchange of fluids, solutes and heat between basement and the overlying ocean. Basement temperatures in the vicinity of outcrops were calculated by co-locating seismic reflection profiles and multi-penetration heat flow stations, and combining these data with regional swath mapping and ocean drilling results. Two basement
outcrops, located on 3.5 Ma crust and separated by 40 km along a direction parallel to the ridge axis, demonstrate contrasting thermal conditions associated with discharge and recharge. A separate study was conducted of thermal conditions within 20-25 Ma seafloor offshore of the Nicoya Peninsula, Costa Rica. Once again, co-located heat flow and seismic data allow mapping of conditions in basement beneath thick sediments. We crossed three crustal boundaries having thermal conditions indicative of an abrupt, shallow transition in hydrothermal activity. Over much of the EPR-generated flank, heat flow is only 20-30% of that expected for normal seafloor, indicating extremely efficient advective heat loss.
Geometry of River Networks by Peter Sheridan Dodds, Columbia University and Lamont-Doherty Earth Observatory
River networks stand as an archetypal example of branching networks, an important sub-class of all network structures. The work I will present combines analytic results, numerical simulations of simple models and measurements of real river networks. Along the way, I'll review some theories regarding the dynamic origin and evolution of river network form, a problem that remains unsolved. A major stumbling block has been the supposition of that certain river network characteristics are
"universal", and I'll endeavor to explain why subtle, theory-confusing variations in network parameters appear to be inevitable.
Talk details: we focus on scaling laws which are central to the description of river networks. To simplify matters, we consider large river networks that may be considered to be effectively flat. Starting from a few assumptions about network architecture, we derive all known scaling laws showing that only two scaling exponents are independent. Having thus simplified the description of networks, we pursue the precise measurement of real network structure and the further refining
of our descriptive tools. We address the key issue of universality, the possibility that scaling exponents of river networks take on specific values independent of region. We find that deviations from scaling are significant enough to preclude exact, definitive measurements. Importantly, geology matters as the externality of basin shape is shown to be part of the reason for these deviations. This implies that theories that do not incorporate boundary conditions are unable to produce realistic river network structures. We also extend a number of scaling laws to incorporate fluctuations
about simple scaling. Going further than this, we find we are able to identify joint probability distributions that underlie these scaling laws. We generalize a well-known description of the size and number of network components (Horton's laws) as well as a description of network architecture (Tokunga's law), how these network components fit together. Both of these generalizations demonstrate that the spatial distribution of network components is random.
Polar Ocean Freezing: Dynamics, Thermodynamics and Instabilities by John Wettlaufer, Applied Physics Lab, University of Washington, and Yale University
Discussions of floating ice as an odd oceanographic phenomenon date to the time of Christ. The two-phase structure of sea ice has been evident to scientific observers since at least the time of Zubov. Nonetheless, a first principles description of the underlying causes of its structure has appeared only relatively recently. Because the mechanical, thermal and electromagnetic behavior of sea ice depend on the spatio-temporal distribution of its phase fraction, predictive models are of broad importance. The mechanisms that give rise to this structure influence the detailed heat and mass balance of the fluids that straddle it. The two-phase evolution of the material involves instabilities with characteristic scales that vary over may orders of magnitude, and can be described as the basic tenets of bulk and interfacial thermodynamics, equilibrium and nonequilibrium statistical mechanics, and hydrodynamics. In this talk, the leading order processes and their geophysical implications are discussed. Although much is known, fundamental questions and practical problems remain today, as
demanding as at any time in the recent past. Moreover, the dynamics and thermodynamics of sea ice may form a test bed for a host of problems in geodynamics.
Formation and Decomposition of Natural Gas Hydrates in Geological Settings by Bruce Buffett, University of British Columbia
Large volumes of natural gas (mainly methane) are trapped along deep continental margins and in permafrost by icy solids known as gas hydrates (or clathrates). Speculations about massive release of methane from gas hydrates into the atmosphere have raised concerns about the possible consequences for global climate changes because methane is a strong greenhouse gas. However, relatively little is known about how gas hydrates form and dissociate in nature. I summarize the conditions required for thermodynamic stability of gas hydrates and use simple models to appraise several commonly proposed mechanisms for hydrate formation. Observations from recent ODP Legs are used to test the predictions of the models. I conclude by examining the potential role of gas hydrates on global climate, focusing specifically on methane release from the world's oceans and from under continental ice sheets during periods of deglaciation.
Flow focusing on the Continental Slope: Implications for Slope Stability and Crustal Fluid Flow by Peter Flemings, Pennsylvania State University
When low permeability sediment rapidly and unevenly buries a permeable layer, fluid overpressures are elevated where overburden is thin, reduced where overburden is thick, and flow focusing occurs along the permeable layer. A basic result is that aquifer pressures converge on the least principal stress and when this happens, the excess pressures generate enhanced permeability, drive fluid expulsion, and in some cases can initiate slope failure and landslides. Flow focusing and associated slope instability is demonstrated for the New Jersey continental slope and the Gulf of Mexico; examples of fluid expulsion on the sea floor and deep basinal fluid migration are explored. Flow
focusing is present in a huge range of geological systems: it regulates crustal fluid flow; it generates slope instability; and it is expressed in sea-floor seeps and hydrothermal vents. Flow focusing may control accretionary prism deformation, the hydrogeology of glacial systems, and may be present on Mars.
Infiltration-Driven Contact Metamorphism by John Ferry, Johns Hopkins University
The principal evidence for infiltration of rocks by chemically reactive fluids during contact metamorphism is: (1) significant progress of decarbonation reactions in rocks that record chemical equilibrium with CO2-poor fluids and (2) changes in oxygen isotope composition greater than can be explained by metamorphic reactions. Infiltration of reactive fluids drives many commonly observed metamorphic mineral reactions, especially in carbonate rocks. Interpretation of the spatial
distribution of both the mineral products of infiltration-driven reactions and mineral d18O with transport theory leads to the identification of: (a) the mechanism of mineral-fluid reaction; (b) the
geometry, amount, and source of reactive fluid flow; (c) the networks for flow at mm to km scales; and (d) the geologic controls on those flow networks.
Two case studies will be compared and contrasted. Reactive fluid flow during contact metamorphism in the Mt. Morrison pendant, eastern Sierra Nevada, California, produced wollastonite in calcite-matrix quartz sandstone but no changes in d18O at the present level of exposure. The mapped distribution of wollastonite indicates that flow was primarily upward, parallel to steeply-dipping bedding, with time-integrated flux in the range 1700-3700 mol fluid/cm2 rock and a magmatic source. Flow was focused into near-vertical sheet-like channels at m to km scales by lithologic contacts and pre-metamorphic faults, dikes, and fold hinges. Reactive fluid flow during contact
metamorphism in the Beinn an Dubhaich aureole, Isle of Skye, Scotland, produced periclase in dolomite marbles and changes in d18O up to 14". The mapped distribution of periclase and isotope alteration indicates that flow was primarily upward, parallel to the vertical contact between marble and a pre-metamorphic dike, with time-integrated flux less than 400 mol/cm2 and a magmatic source. At both locations reactive fluid flow was pervasive at the grain scale although the exact nature of the mm-scale flow network(s) is uncertain. Vertical fluid flow explains the development of d18O alteration in the Beinn an Dubhaich aureole and its absence of exposure in the Mt. Morrison pendant. The difference in time-integrated flux between the locations is correlated with pluton size. An overview of 12 other aureoles indicates that results for the Mt. Morrison pendant and Beinn an Dubhaich aureole may be generalized to contact metamorphism worldwide with a few significant
exceptions. In particular, flow networks at the m-km scale are controlled by a variety of pre-metamorphic planar geological structures. Flow in vein networks is not important.
The significance of mud volcanoes for dewatering convergent margins, devolatilization processes and geochemical budgets by Achim Kopf, Scripps Institution of Oceanography
Mud volcanism is a phenomeno which has puzzled geoscientists for approximately two centuries. Mud diapirism and extrusion are a well-known phenomena whereby fluid-rich, fine grained sediments ascend within a lithologic succession due to their buoyancy. These processes have long been recognized to be related to the occurrence of petroleum, regional volcanic and earthquake activity, and tectonic compression (e.g. in accretionary complexes and orogenic belts). Mud extrusions are found both onshore and offshore and in various tectonic settings on Earth, however, the majority of the features known to date are located at convergent margins. Despite their abundance, mud volcanoes have only recently been acknowledged as one key mechanism in volatile transfer at active convergent margins. Main reason for this neglect may be the short-lived nature of these features, because the clay-rich, undercompacted, liquefied material easily erodes.
A compilation of previous mud volcano research shows variable geometries of the shield- and dome-shaped features (up to tens of km in diameter and several 100m in height) and a large diversity regarding the origin of the fluid and solid phases involved in extrusion. Despite the variability in appearance, deep-seated mud volcanoes have a number of features in common: (i) The connection with rapidly deposited, overpressured, thick argillaceous sequences of mostly Tertiary age as
parent beds; (ii) the incorporated fragments of underlying rocks and other structural associations; (iii) a relationship to regional tectonics and seismicity; and (iv) the presence or influx of gaseous
and liquid fluids to facilitate diapiric intrusion and extrusion. Gas (predominantly methane), water, and mud may be mobilized at subbottom depths of only a few meters, but in places can originate from several km depth (with minor contributions from crustal and mantle depth).
The possible contribution of mud extrusion to global budgets, both from quiescent fluid emission and the extrusive processes themselves, are important. In regions where mud volcanoes are abundant, like in the collision zones between Africa and Eurasia, fluid flux through mud
extrusion exceeds the compaction-driven pore fluid expulsion. This growing importance of mud volcanism as an efficient dewatering mechanisms in subduction zones has important repercussions on the backflux of water and chemical components to the hydrosphere. Also, quiescent degassing of mud volcanoes may contribute significantly to volatile budgets and, hence, greenhouse climate. Arguably, the major role of mud volcanism in global budgets would have been recognized earlier if not so few fossil examples were preserved.
Pockmarks and self-sealing fluid flow through the seafloor by Martin Hovland, Statoil, Norway
Based on this new information from visual surveys done at two methane seep sites in the North Sea, it is suggested that there may be three phases leading to the sealing and re-routing of upward-directed fluid flow. These phases are manifested by: 1) Virgin seeps, where the gas comes directly from small vents in the seafloor. 2) Oxidation of methane with sulphate by Anaerobic Oxidation of Methane aggregates (AOMs), providing sulphide to bacterial mats (i.e., Beggiatoa sp.) where the gas
accumulates in sediments before occasionally venting through holes in the mats. 3) Authigenic carbonate-cemented slabs where no visible gas is evident, but where sampling of carbonate nodules suggests that gas migrates up to the lower part of the slabs and seeps through small holes and is utilized in different ways, including anoxic oxidation and production of carbonate cements. These results, therefore, suggest that the formation of the bacterial mats may represent the first phase of
natural sealing. The formation of the carbonates, which host numerous sessile and filter-feeding organisms, represents the final phase in the natural seep sealing process. In other regions of the oceans, where other gas compositions, migration rates, and sediment conditions prevail other sealing processes will occur.
Glacial Hydrology: A fluid flowing in its own solid by Neil Humphrey, University of Wyoming
The flow of water on top, inside and underneath glacial ice is an intriguing geo-science example of a fluid flowing through a solid of the same composition. Parts of the glacial hydrologic system are well
understood, these include flows nearest the surface (or input) and those nearest the glacier terminus (or the output). The remainder of the flow system is poorly known, despite considerable effort over many years, and despite the fact that the flow of ice on earth, from mountain glaciers to ice sheets is mainly controlled by water flow and water storage. Understanding the behavior of water and ice requires consideration of thermodynamics and heat transfer, non-linear rheologies and non-homogeneous and anisotropic stress and strain rate fields, as well as multi-phase flow problems. In this talk I will present examples from both the known and poorly known parts of the flow system, and will include some possible solutions to some of the nagging problems.
Subduction zone pore pressures, fluid budgets, and flow pathways: Insights from numerical modeling by Demian Saffer, University of Wyoming
At subduction zones, tectonic loading of offscraped and underthrust sediments results in rapid compaction and high rates of fluid expulsion. The resulting hydrologic and mechanical systems are
intimately linked, and probably change dramatically through time. The potential for devastating earthquakes and tsunamigenic events is strong motivation for understanding how fluid processes may control fault zone strength and material properties. Distributed deformation, relatively high rates of fluid production, and plate boundary faulting constitute ideal natural laboratory conditions for studying the interplay of fluid flow and tectonics, as well as for evaluating fluid, heat, and solute
mass balances. Despite the importance of fluids in a wide range of subduction zone processes, many key questions remain unanswered: What are pore fluid pressures and porosities at depths where earthquakes nucleate? Is fluid flow along faults steady through time, or is it transient? If flow is transient, is it related to the seismic cycle? During what percentage of the time are flow conduits open?
Because only the uppermost few kilometers of subduction zones are accessible by drilling or seismic imaging, I employ numerical models which couple fluid flow, porosity reduction, clay dehydration, and solute transport to address these questions. The results indicate that most of the incoming fluids escape via diffuse flow at the seafloor, but a significant fraction can be channeled along high-permeability conduits. Down-hole geochemical anomalies, previously interpreted as evidence for long distance flow of deeply derived fluids along fault conduits, probably reflect a dominant component of in situ diagenesis combined with some focused expulsion of fluids from depth. By comparing numerical model results with direct measurements of vent fluxes, I demonstrate that fluid mass balances can be used to estimate the spatial or temporal extent of high permeability
fault zones that may represent weakened patches maintained by high fluid pressure. This research has led to a refined conceptual model of fault behavior, where focused flow in fault zones occurs within a network of connected high permeability pods; as permeable channels shift location through time, transient flow and the associated thermal and chemical signals are introduced locally.
At a large scale, pore fluid pressures are widely thought to control the shape of accretionary complexes. Existing theoretical and conceptual models show that significantly elevated pore fluid pressures are required for accretionary wedges to maintain their observed geometry - that of a thinly tapered wedge. In general, the taper angle of an accretionary wedge and the strength of the plate boundary fault are predicted to decrease with higher pore fluid pressure. In fact, pore fluid pressure is not a static quantity, but reflects a dynamic balance between geologic processes that act to increase pore pressure, and properties of the rock that allow pressures to dissipate. I propose a refined conceptual model for accretionary wedge morphology that incorporates the fundamental causes of excess pore pressure in subduction zones. Numerical models of fluid flow demonstrate that sediment permeability and plate-convergence rate are two important controls on subduction zone mechanics through their influence on pore pressure. Low permeability and rapid convergence sustain nearly
undrained conditions and result in shallowly tapered wedge geometry, whereas high permeability and slow convergence result in steep geometry. These results are in generally good agreement with data from active accretionary complexes, but also illustrate the importance of other factors, such as incoming sediment thickness and stratigraphy. One key implication is that strain rate and hydrologic properties may strongly modify the strength of the crust in a variety of geologic
settings by controlling pore pressure.
Elevated pore pressures and episodic fluid flow in accretionary complexes: Results of field measurements and numerical modeling by Elizabeth Screaton, University of Florida
In accretionary complexes, rapid loading due to sedimentation and tectonic compression elevates fluid pressures and drives fluid flow. If sediment permeabilities are too low to allow fluid expulsion to keep pace with the loading, overpressures build. These overpressures are believed to control the initiation and movement of fault zones, including the decollement zone that separates overriding and subducting plates. At the same time, fault zones in accretionary complexes are believed to play a major role in focusing fluid expulsion, which requires high permeabilities. A variety of direct and indirect evidence has been used to understand how faults can be zones of both elevated pore pressures and high permeabilities. Evidence for elevated pore pressures includes mechanical arguments, observations of anomalously high porosities, and a few direct measurements of pore pressure. Evidence concerning sediment permeabilities includes observed fluid flow rates at vent sites, laboratory tests on cores, and rare in-situ hydrogeologic tests. However, our strongest evidence for high permeabilities is provided by inverse modeling of observed thermal or geochemical anomalies. Overall, evidence suggests that fluid flow must be episodic, probably related to fracture opening as pore pressures reach a critical threshold. Simulation of the evolution of pore pressures and permeability in a growing accretionary complex provides greater insight into the episodicity of fluid flow, and is consistent with previous inferences that there is likely to be a large time lag
between the initiation of accretion and the development of high-permeability pathways.
The Role of Fluid Muds in Coastal Sediment Transport by Gail C. Kineke, Dept. of Geology and Geophysics, Boston College
High-concentration suspensions of fine sediments, or fluid muds, have been investigated for several decades, often in conjunction with dredging activities or sediment dispersal from rivers with extremely high suspended-sediment concentrations. Studies have been specific to particular sites such as the Severn or Gironde estuaries, the mud banks on the Surinam coast, and the Huanghe (Allen et al., 1980; Kirby and Parker, 1983; Wells and Coleman, 1981; Wright et al., 1990). A study on
the Amazon shelf (AMASedS 1989-1991) demonstrated the critical role these suspensions play in sediment dispersal and the influence they have on a variety of processes ranging from modifying tidal wave propagation to inhibiting geochemical exchange between the water column and seabed (Kineke et al., 1996). Since the Amazon study, fluid muds have been identified and studied in a number of diverse estuarine and shelf environments that are subject to a wide range of conditions of
sediment and freshwater input as well as wave and tidal energy. With advances in instrumentation, fluid muds are being recognized as quite common, and, where present, dominate the sediment transport processes.
Examples include delivery of Eel River sediment to the mid-shelf, wave attenuation and coastal accretion on the shallow Louisiana shelf, localized areas of high deposition in estuaries, and hyperpycnal flows down submarine canyons (Blake et al., 2001; Kineke et al., 2000;
Traykovski et al., 2000).
Magma Chamber Madness by Bruce Marsh, Johns Hopkins University
The dynamic evolution (as well as the very concept itself) of magma chambers and magmatic systems in gerneral is considered in the light of both historic and present concepts of the origin of the diversity of the igneous rocks. The cooling and solidification of magmatic sheets is examined with particular emphasis on natural systems, like the Sudbury impact melt sheet, where the history of solidification is known. These insights are used to discuss how major, vertically extended magmatic systems, or mush columns, are structured and how they function to provide the critical integrated physical and chemical link between deep source regions and volcanism itself. The extensive Ferrar magmatic system in the McMurdo Dry valleys, Antarctica shows many of the
features expected in large scale mush columns.