The vitality of the current Earth science research community is manifestly evident in the numerous strategic planning, Grand Challenges, and science vision documents that have been produced over the past decade (a list of key documents is presented in Appendix A). Any attempt at comprehensive assessment of new research opportunities across the discipline would quickly become unwieldy, and the finite expertise of any committee would result in some oversights. The committee on New Research Opportunities in the Earth Sciences (NROES), informed by personal knowledge, myriad documents produced by workshops and community organizations, and both solicited and contributed input from many researchers and program managers (see Appendix B) has attempted to identify specific areas in the basic Earth science research scope of the Division of Earth Sciences (EAR) of the National Science Foundation (NSF) that are particularly poised for rapid progress during the next decade.
Seven primary topics involving complex dynamic geosystems that can only be fully quantified by interdisciplinary approaches are highlighted in the following sections organized by scale and disciplinary participation related to the EAR Deep Earth Processes and Surface Earth Processes sections: (1) the early Earth; (2) thermo-chemical internal dynamics and volatile distribution; (3) faulting and deformation processes; (4) interactions among climate, Earth surface processes, tectonics, and deep Earth processes; (5) co-evolution of life, environment, and climate; (6) coupled hydrogeomorphic-ecosystem response to natural and anthropogenic change; and (7) interactions of biogeochemical and water cycles in terrestrial environments. These address a range of grand challenge-scale fundamental topics of both curiosity-driven and strategic Earth Science. Key to many of these topics and to many other Earth science applications are geochemical approaches to geochronology by exploiting the variety of stable and radiogenic isotopes that exist in nature to provide relative and absolute dating of geological materials and events. The expanding demand for accurate sample dating for many of the research opportunities motivates consideration of restructuring EAR-supported geochemical facilities that must simultaneously promote innovation of methodologies, training of next-generation geochemists, and servicing the burgeoning demands for what are seldom routine sample dating analyses.
Much of Earth’s present-day structure and significant parts of its history can be traced back to events that occurred within the first few hundreds of million years after its formation. Understanding the processes involved in Earth accretion and early chemical differentiation is essential for establishing the initial thermal conditions of the dynamical systems of the interior, the volatile content of the planet, and the origins of the continents that have led to the current Earth system. Recent progress on understanding the early Earth has been substantial, yet we have only begun the task of resolving the timing, nature, and interrelationships of the most decisive events, including cataclysmic
impacts; magma oceans; segregation of the core; early forms of continents, oceans, and the atmosphere; the onset of plate tectonics; and, of course, the origin of life. Because Earth grew and differentiated rapidly, the energy available to the Earth system during its early history was far higher than today, permitting whole sets of physical and chemical processes without counterparts in the modern Earth. The overarching challenge here is to understand how Earth transitioned from its formative state into the hospitable planet of today (see Box 2.1). Lessons learned from the early Earth will help us interpret the processes occurring in the hundreds of extrasolar planetary systems now being discovered by astronomers.
Accretion of Earth
The birthplace of Earth was a protoplanetary accretion disk, a cloud of gas and dust surrounding the early Sun. Modern astronomy provides a glimpse of what this environment may have been like, in the form of debris disks that surround young stars, some of which have been imaged by the Hubble Space Telescope (see Figure 2.1). Accretion disks are subject to instabilities driven by powerful gravitational and electromagnetic forces that collect dust particles into planetesimals, typically 1-kilometer-sized objects that were the fundamental building blocks of Earth and the other terrestrial planets. Once a sufficient density of planetesimals developed in the nebular cloud, increasingly violent collisions began to dominate the accretion process, forming an ever-smaller number of growing planetary embryos that swept up most of the remaining nebular debris.
Although much effort has been directed toward understanding accretion from the perspective of solar system dynamics, many related processes that were important for early Earth have not received the same attention. Accretion models, for example, often assume that colliding planetesimals simply adhere, ignoring effects like fragmentation, spin and precession, melting, vaporization, condensation, and differentiation (Chambers, 2004). There is mounting evidence for these processes, many of which bear directly on the final chemistry and structure of the accreting body (Halliday, 2004).
Geochemical and cosmochemical observations provide important constraints on the timing and the mechanisms of accretion and segregation of the core, although several interpretations are possible. For example, in the Hafnium-Tungsten (Hf-W) system, the excess radiogenic 180W in the silicate Earth relative to chondritic meteorites has been interpreted as rapid accretion or alternatively as incomplete mixing of the impactor with the growing Earth (Halliday, 2008; Rudge et al., 2010). Similar interpretations have come from other short-lived isotope systems, such as146Sm-142Nd (O’Neil et al., 2008), which also have implications for the earliest crust. The fusion of geochemistry and geophysics offers many promising avenues for better understanding formative processes that governed the early history of the Earth.
Based on isotopic evidence from meteorites, what originated as occasional planetesimal collisions soon began to run away, leaving a small number of rapidly growing planetary embryos. Improved chronological methods reveal that melting and differentiation occurred within a few million years of the formation of the first solids, probably driven by collisions and assisted by now-extinct radioactive heat producers such as26Al and60Fe. Accordingly, the assumption that Earth formed by a continuous influx of small particles made of pristine solar system condensates has given way to a much more dramatic model, in which Earth was assembled by a relatively small number of traumatic collisions involving larger objects, some of these already having differentiated interiors and well as their own internal dynamics (Canup and Asphaug, 2001). Future progress on the processes and timing of Earth’s growth in the coming decade will rely on a diversity of approaches, including:
• Application of new isotope techniques for dating methods
• Closer integration of isotope geochemistry with astrophysical approaches to planetary formation
• More comprehensive and more realistic dynamical models of the accretion process
• Evolutionary studies of the chemistry and physics of planetesimal-sized objects and planetary embryos
Earth’s interior and surface environments are profoundly influenced by our position in the Solar System and interaction with the Moon and other planets. The Moon stabilizes the orientation of Earth’s spin axis and promotes climate stability, in stark contrast to, for example, Mars. Gravitational interactions between Earth and other planets, particularly Jupiter, cause small variations in the eccentricity, obliquity, and precession of Earth. While small, these variations are likely partly responsible for ice ages.
The discovery of hundreds of planets orbiting other stars, so-called extrasolar planets, provides a new opportunity to understand whether the architecture of our solar system and presence of an Earth-like planet in the habitable zone are common. Planets around young stars may also offer a window into the earliest Earth. Limited information, however, is available about these planets, and in the best cases we know their mass, radius, eccentricity, and temperature and are able to detect some gases.
Continued exploration of our own solar system has led to new, unexpected discoveries: an active dynamo on Mercury, eruptions on Enceladus (see Figure B2.1), and methane lakes on Titan. These discoveries provide new opportunities to test our understanding of the basic processes that govern planetary evolution and interactions between Earth systems, particularly the interior, the geodynamo, surface environments, and the atmosphere.
Understanding these new discoveries and further exploration of our solar system are activities typically supported by the National Aeronautics and Space Administration (NASA). Nevertheless, there are opportunities to better understand Earth systems and the earliest Earth made possible by exploring other planetary objects. Collaboration between NASA and NSF in supporting such projects can only be positive.
FIGURE B2.1 Ice geysers erupt on Enceladus, the bright and shiny inner moon of Saturn. This image presents a backlit view of the moon’s southern limb, where icy plumes were discovered by the NASA Cassini spacecraft mission in November 2005. Cryovolcanism is evidence that the 500-km-diameter Enceladus has active internal tectonics. SOURCE: NASA Jet Propulsion Laboratory/Space Science Institute.
FIGURE 2.1 Hubble Space Telescope (HST) images showing young stars surrounded by dust rings thought to be the birthplaces of planets like Earth. Top: The planet Fomalhaut b orbits the star Fomalhaut (25 light-years away in the southern constellation Piscis Australis) near a ring of dust similar to the Kuiper Belt that may contain bodies ranging from dust grains to objects the size of dwarf planets. Bottom: Light reflected off a debris disk in cross section around the young star AU Microscopii, HD197481. SOURCES: Top: NASA, the European Space Agency (ESA), and Z. Levay. Bottom: NASA, ESA, and J. Graham.
Response to the Moon-Forming Impact
Although conclusive evidence is still lacking that ever-larger impacts dominated the later stages of Earth’s growth, the global dynamical and thermal implications of this process are not in doubt. Once Earth reached an appreciable mass, the enormous amounts of kinetic and gravitational potential energy released by large impacts dictate widespread melting, with regional and possibly global magma oceans extending to considerable depths (Tonks and Melosh, 1993).
The compositional similarity of Earth’s mantle and the Moon and realization of the importance of large impacts in the early Solar System, together with the large angular momentum present in the Earth-Moon system, have led to the theory that the Moon formed as the result of a late cataclysmic impact of a Mars-sized object with the growing Earth (Wetherill, 1990). Particle-based simulations of this giant collision (Canup, 2004; see Figure 2.2) predict that much of the preexisting layered structure of Earth was obliterated
FIGURE 2.2 Two time slices in the animation of a glancing impact of a Mars-sized planetary embryo into the proto-Earth. The silicate mantles of both objects are shown in yellow, whereas their metallic cores are shown in red. The first image is slightly before the impact; the second is about two orbital rotations of the proto-Earth following impact. SOURCE: Reprinted from Canup (2004), with permission from Elsevier.
and a substantial portion of the impacting material was thrown back into orbit, creating a post-impact accretion disk surrounding the proto-Earth, complete with its own silicate vapor atmosphere. These simulations also predict that the Moon consists primarily of material from the impacting object, and not material from proto-Earth. This computational model is challenged by remarkable similarity in oxygen isotopes found between lunar and Earth rocks, raising questions about the partitioning of material during impact.
Despite its widespread acceptance, direct evidence of a Moon-forming giant impact—the smoking gun in Earth’s early history—remains elusive. Similarly, our understanding of the events accompanying giant impacts and their consequences for the chemical and physical modification of the early Earth remain sketchy. Further delineation of the Moon-forming event and its consequences for Earth are high priorities for the coming decade.
Terrestrial Magma Oceans
Magma oceans, an almost inevitable consequence of large planetary impacts given the energies involved, were first proposed to explain the plagioclase-dominated crust of the Moon (Warren, 1985), and differentiation in an early magma ocean on Mars is thought to be responsible for the range in source compositions of Martian meteorites (Borg and Draper, 2003). As is the case for a moon-forming impact, indisputable evidence for magma oceans and their associated early atmosphere on Earth remains elusive, although there is indirect evidence from abundance patterns of the elements affected by core formation (Kleine et al., 2004), plus some isotopic evidence for early mantle differentiation and atmosphere formation that are indicative of a magma ocean environment (Moynier et al., 2010). What is more certain, however, is that terrestrial magma oceans and the early atmosphere provided highly dynamical environments in which a wide variety of chemical and physical processes were active, ranging from shock-wave heating to fracturing and fragmentation, turbulent convection, percolation, mixing, and a host of possible redox reactions. Understanding the evolution of a terrestrial magma ocean requires answers to such basic questions as:
• What is the relationship between impact and magma ocean sizes?
• What is the lifetime of a magma ocean, and how is it coupled to the early atmosphere?
• Does a terrestrial magma ocean crystallize from the bottom up or from the top down?
• Was there a deep-mantle abyssal magma ocean?
• Do deep melts rise or sink in the early mantle?
• What sequence of crystals form in a cooling magma ocean?
• As a magma ocean crystallizes, is it stably stratified, or will it overturn?
• How did metals and silicates mix and then segregate in magma oceans?
• What was the nature of mantle dynamics following magma ocean solidification?
Providing answers to these questions will probably require geodynamical modeling constrained by improved understanding of the petrology of melts and element partitioning at high pressures and temperatures, in parallel with interpretations of present-day seismic images of mantle heterogeneity in the deep mantle, where the chances are best of finding relics of this process still preserved. In addition, many of the issues raised by these questions are linked together, requiring cross-disciplinary expertise. For example, separation of immiscible liquids (in this case, iron from silicate melts) with greatly different densities happens rapidly in a low-viscosity magma ocean, whereas buoyancy-driven segregation of silicates depends on the environmental conditions. Because the moon’s interior spans a small range of pressures, the crystallization sequence of a silicate lunar magma ocean is reasonably well understood (Shearer, 2006). As is the case for many shallow layered mafic intrusions on Earth, buoyancy-driven separation of lower-density Ca- and Al-rich plagioclase from denser Mg- and Fe-rich silicates occurs on the Moon. On Earth, however, the greater range of internal pressures introduces the likelihood of liquid-solid density crossovers (Mosenfelder et al., 2007; Stixrude et al., 2009), so magma oceans may stabilize at both the top and the base of the mantle (Labrosse et al., 2007), as shown in Figure 2.3, significantly complicating their evolution.
In addition to the energy acquired from impacts, the segregation of the core released enormous amounts of gravitational potential energy into the Earth system. Isotope evidence generally points to early core formation (Yin et al., 2002), which is consistent with the magma ocean hypothesis, wherein growth of the core essentially kept pace with growth of the mantle. There are several theories on how the core formed that are compatible with large impacts and the existence of magma oceans. One theory assumes that impacting cores fell through the magma ocean as large metal masses, directly merging with Earth’s core (Halliday, 2006). Another assumes that dispersed metal rained down through the magma ocean, collected at its base,
FIGURE 2.3 Schematic evolution of progressive crystallization of surface and basal magma oceans (yellow) following Earth accretion and core formation, based on the assumed deep-mantle density crossover between melt and solid, leading to upward segregation of melts in the upper mantle and downward migration of melts in the lower mantle. Core-forming metals are shown in orange; solid mantle is shown in gray, with circulation indicated by arrows. SOURCE: Labrosse et al. (2007). Reprinted by permission from Macmillan Publishers Ltd.
then descended through the underlying crystalline mantle by several possible mechanisms, including fracture propagation, large metal diapirs, or metal-silicate plumes (Ricard et al., 2009). The measured abundances in the mantle of moderately siderophile elements such as nickel (Ni) and cobalt (Co) indicate that some degree of chemical equilibration between core-forming metals and mantle silicates took place, possibly at elevated pressure and temperature conditions (Chabot et al., 2005; Wood et al., 2006). Additional geochemical and petrological constraints, better resolution of its timing and duration, and a fuller picture of the possible dynamics are needed to constrain the core segregation process.
Early Earth's Surface Environments
Evidence indicates that the accretion and major differentiation of Earth, including core formation, were largely complete within about the first 100 megayears (Myr). The ensuing 500-Myr time interval, the Hadean Eon, is often referred to as the geological dark age, because there is little preservation of this interval in the rock record. Yet it remains a crucial stage in Earth’s history because the transition to a habitable surface environment occurred during this time.
There are few solid constraints on the Hadean Earth and a host of first-order questions. Heat produced during accretion and core formation, together with the higher concentrations of heat-producing radioactive elements, point to a hot, possibly water-deficient, mantle. The consensus view is that Earth’s initial atmosphere, composed mostly of hydrogen, was lost very early, perhaps during a T-Tauri phase of solar activity or through hydrodynamic escape to space aided by the strongly ultraviolet-emitting young Sun (Catling, 2006). As for the early composition of the secondary atmosphere, there is far too little in the way of direct evidence, although the decisive events in Earth’s early history point to some plausible scenarios. One possible consequence of the Moon-forming impact is rapid evolution from a hot silicate atmosphere to a steam-dominated greenhouse atmosphere (Zahnle et al., 1988), and once the magma ocean solidified, liquid water could stabilize at the surface with carbon dioxide and methane dominating the climate (Kasting and Ono, 2006). A key unknown here is the capacity of the mantle to sequester water, possibly in the presence of early whole-mantle convection.
Clues from the Early Crust
Evidence for the earliest chapters in Earth’s history comes from a variety of sources, including the bulk composition of Earth and the Moon, the angular momentum of the Earth-Moon system, traces of short-lived radioactive isotopes in meteorites and terrestrial rocks, terrestrial and lunar patterns of element abundances, and perhaps most importantly, the oldest crustal rocks and minerals. The discovery of increasingly old crustal rocks (see Figure 2.4) provides a few tantalizing clues on the state of Earth’s surface in the late Hadean
FIGURE 2.4 Images of Earth’s oldest crustal rocks. Left: The 4.28-Ga “faux-amphibolite” from the Nuvvuagittuq supracrustal belt in northern Quebec. Right: Zircons from Jack Hills Australia with ion-probe spots labeled by U-Pb ages. SOURCE: Left: O’Neil et al. (2008). Right: Wilde et al. (2001). Reprinted with permission from AAAS.
and the earliest Archean. In terms of preservation, the most diverse suite of ancient crustal rocks is found in the Isua terrane in Greenland, with ages as great as 3.8 Ga (Appel et al., 2001). These are moderately metamorphosed but contain evidence to suggest that plate tectonic processes, liquid water oceans, and perhaps life forms were present. Still older are the Acasta gneisses from north-central Canada, dated around 4 Ga (Bowring and Williams, 1999). The only known Earth materials that are unequivocally older are small zircon grains that have been removed from their parent rock, transported by fluvial systems, and deposited in sedimentary rocks of a younger age (see Box 2.2). Advances in microanalytical techniques, especially ion microprobes, have established ages around 4.3 Ga for the oldest of these. The overarching inference from these oldest crustal materials is that by the late Hadean and certainly by its end, Earth’s surface environment was rather equable, perhaps not dramatically different from the present (Mojzsis et al., 2001; Wilde et al., 2001), so that some of the conditions for sustaining life were already in place. Other critical elements are
Earth’s Oldest Solids: Hadean Zircons
The oldest known terrestrial solids are zircon crystals. Zircons are extremely resistant to both chemical and physical destruction and hence have the potential to survive billions of years of reprocessing after their formation. Fortunately, they also carry a range of mineralogical, geochemical, and isotopic tracers that document their age and environment of formation.
The oldest known zircons come from the Jack Hills region of Western Australia, where they are found in metamorphosed rocks originally deposited in a fan-delta setting (see Figure B2.2; Spaggiari et al., 2007). Although they span a range of ages, many of the Jack Hills zircons formed in the Hadean Eon (>3.8 Ga), and the oldest among them crystallized <250 million years after the birth of the solar system (e.g., Compston and Pidgeon, 1986). Because they provide a unique window into the early Earth, the Jack Hills crystals have been intensively studied in the past decade. A generally consistent story emerges from analyses of the trace element and isotopic composition of the zircons as well as the assemblage of mineral inclusions trapped within them (e.g., Wilde et al., 2001; Cavosie et al., 2005; Watson and Harrison, 2005; Trail et al., 2007; Hopkins et al., 2008; Harrison, 2009). The zircons appear to be igneous and formed at relatively low temperatures, suggesting crystallization from magma at or near water saturation. Inclusion mineralogy and oxygen isotope data indicate the magma may have formed from melting of a felsic protolith that had interacted extensively with an early hydrosphere, possibly an ocean. Geobarometry and thermometry of the inclusions and the zircons themselves suggest crystallization in an unusually cool geothermal gradient. These observations evoke an environment remarkably similar to the conditions under which modern granites form in subduction environments. Thus, it has been argued that within just a few hundred million years of the formation of the planet, a stable siliceous crust, an active hydrosphere, and a form of plate tectonics with marked similarities to the current regime had already been established.
Further advances in this field may come from identification of new localities where extremely old rocks and detrital minerals occur. This task will require application of a variety of geochemical and petrological methods, especially in geochronology. The magnitude of the undertaking is illustrated by the work invested to identify the oldest zircons from the Jack Hills. Ion microprobe analyses of more than 100,000 individual zircons were required to identify the ~100 crystals with ages >4.2 Gyr (Holden et al., 2009).
FIGURE B2.2 Jack Hills: a 4.06-billion-year-old Jack Hills zircon with mineral inclusions that characterize the parent magma’s protolith and melting/crystallization conditions. SOURCE: Hopkins et al. (2008). Reprinted by permission from Macmillan Publishers Ltd.
more problematic, however, particularly oxygen, which does not appear to have been abundant then. This raises several fundamental questions, such as:
• What is the critical oxygen concentration for early life forms?
• What was the role of the late heavy bombardment near 3.9 Ga on the terrestrial environment?
• At what time did the earliest continental crust stabilize?
• When did plate tectonics initiate, and what environmental effects did this transition have?
The Hadean Mantle and Core
Most of the major questions posed for the early surface environment also involve the composition and dynamics of the Hadean mantle, and some of these also involve the early state of the core. For example, the thermal and compositional stratification of the mantle following the major phase of core segregation (and magma ocean solidification) constitute the “initial conditions” for subsolidus mantle convection. In the same way, conditions in the core inevitably changed once the major differentiation had occurred. Evidence for these transitions can be found in the context of the search for ancient rocks and minerals described previously. Geobarometry and geothermometry techniques can infer mantle temperatures and pressures, and magnetized samples provide information about the nature of the early geodynamo and also on the energetics of the Hadean deep Earth.
An Early Earth Initiative
This suite of topics involving the early Earth emerges as a major research opportunity because there have been significant advances in theory, observations, and modeling capabilities across all of the related areas but little coordination of the research agenda. Developing a community focus on these topics and coordination of the interdisciplinary approaches is likely to accelerate progress, much as has been the case for studies of the present-day deep Earth system. The complexity and energetics of the early Earth are distinct from today, and disciplinary approaches need to be informed by the geosystems perspective that an interdisciplinary context can provide. An Early Earth initiative could build on existing community organizations and funding programs, but distinct focus is required to catalyze coordinated momentum in this arena.
The elucidation of plate tectonics over the past 50 years has provided a general framework for understanding shallow Earth structures, kinematics, and processes and for relating observations of the present Earth to those preserved in the geological record. The quests to fully quantify three-dimensional plate dynamics and to determine how distribution of materials at Earth’s surface evolves with the internal dynamic system remain primary goals of the Earth sciences. The dynamic configuration, thermal and chemical fluxes, and driving forces within Earth’s interior are all of central importance to understanding our planet’s evolution, but these must be deduced from observations made at the surface. An improved understanding of thermo-chemical internal dynamics and volatile distribution within Earth also has important societal implications for the mitigation of volcanic and earthquake hazards and for the discovery and development of mineral and geothermal resources.
Making progress has required parallel maturation of a suite of disciplines that bring key information to light: seismology to image elastic and anelastic properties and material heterogeneity throughout the interior, mineral physics to characterize thermo-elastic properties, phase equilibria, electronic transitions, and transport properties of Earth materials over the full pressure-temperature range of the interior, geodynamics to quantify dynamic behavior of deep thermo-chemical systems and their surface manifestations, geomagnetism to probe the flow field of the outer core material and to constrain temporal evolution of the geodynamo, geochemistry to define internal chemical variability and timing of fractionation events, and geology to decipher the history of crustal formation and plate tectonics recorded by surface rocks. As observational, laboratory, and modeling capabilities of these disciplines have expanded, the prospects for major advances in our understanding of Earth’s internal
dynamics have increased, and a concerted interdisciplinary effort over the next decade holds the promise of significant impact on fundamental questions such as:
• How long has plate tectonics been in operation, as we see it today?
• What is the style of mantle convection and material flux between the upper and lower mantles?
• How is chemical heterogeneity distributed in the mantle, how and when was it created, and what is its role in the dynamic circulation?
• What is the volatile budget of the deep Earth?
• How have the core and geodynamo evolved over time?
• What are the driving forces of plate tectonics and internal circulation?
• When and how did the continents form?
Specific topics for which there are clear opportunities for making progress in the next decade include (1) appraisal of geochemical heterogeneities in the deep mantle and their relationship to the dynamic system, (2) quantification of volatile fluxes and their distribution in the mantle, and (3) determination of core evolution. All three topics are central to determining the thermo-chemical evolution of Earth. Progress is being made in these areas by concerted disciplinary and interdisciplinary efforts. Breakthrough advances that resolve outstanding issues will require enhanced resolution of fine-scale structures in the interior beyond what can now be achieved, and efforts to attain higher resolution from seismological, geodynamical, and mineral physics approaches will need to be undertaken.
Quantification of Geochemical Heterogeneities and Their Role in Mantle Dynamics
Earth’s mantle comprises an immense convective system that circulate heat, volatiles such as water and carbon dioxide, silicate melts, former lithospheric material, and a host of other chemical and isotopic species between the interior and the surface. Throughout Earth’s history chemical differentiation has produced continental and oceanic crust, much of which has been subducted or delaminated, generating compositional and isotopic mantle heterogeneity. Some chemical heterogeneities have remained sequestered in the interior for billions of years, while others have rapidly recycled to the surface. This multicomponent transport constitutes the primary interaction of the deep Earth with the ocean, atmosphere, and crust over geological timescales. The internal convective engines provide strain energy for earthquakes, heat for volcanic activity, and power for the core geodynamo. Determining the magnitude, spatial distribution, and temporal variability of geochemical heterogeneities and pinpointing the locations of internal reservoirs where they are sequestered are key to understanding how the deep interior contributes to Earth’s evolution (NRC, 2008).
A profound task is to fully understand the configuration of global circulation in the mantle and its capacity to sequester chemical heterogeneities in reservoirs. Evidence from mantle-derived isotopes has long been interpreted as favoring layering of the mantle, while most geodynamic interpretations and some seismic interpretations favor mantle circulation that is at least partially continuous from top to bottom, with the transition zone providing some degree of resistance. Reconciling geochemical evidence favoring isolated mantle reservoirs, seismic evidence for down-welling slab material in the lower mantle, and geodynamic models that tend to favor extensive, although possibly intermittent, circulation remains at the heart of this long-standing controversy (Kellogg et al., 2004; Lay, 2009; Olson, 2010).
Quantifying the nature and dynamical influence of deep Earth chemical heterogeneities will require an interaction of multiple Earth science subdisciplines, including geodynamics, petrology, mineral physics, geochemistry, and seismology. New opportunities naturally arise from these interactions. For example, improved resolution of mantle seismic heterogeneity provides better constraints on candidate reservoirs and places limits on their compositions and geodynamic behavior. A dramatic example of a recent interdisciplinary advancement on this topic is provided by the discovery of two huge lower-mantle provinces with distinctive material properties (see Figure 2.5). These are the Southern Pacific and African Large Low Shear wave Velocity Provinces (LLSVPs) with several thousand-kilometer dimensions extending upward from the core-mantle boundary hundreds of kilometers (e.g., Ni et al., 2005; Wang and Wen, 2007). First detected
FIGURE 2.5 Pattern of S-wave velocity anomalies (dVs) at the core-mantle boundary for model S20RTS (Ritsema and van Heijst, 2000). Red areas have lower than average S-wave velocity, and blue areas have higher than average S-wave velocity. The green curves outline the 20 percent of the core-mantle boundary area with the lowest S-wave velocities, and this corresponds to the two LLSVPs beneath southern Africa and the south-central Pacific that have been characterized by seismic tomography and waveform modeling studies over the past two decades. SOURCE: Reprinted from Thorne et al. (2004), with permission from Elsevier.
by global seismic tomography, over the past decade these LLSVPs have been found to have abrupt lateral margins; stronger reductions of S-wave velocity than P-wave velocity, indicating anomalously high incompressibility; and anomalously high density—all suggestive of hot, chemically distinct material (Garnero et al., 2007; Garnero and McNamara, 2008; Trønnes, 2009).
Geodynamical modeling (see Figure 2.6) suggests that such massive hot dense piles of material can be localized by mantle circulation, with their margins possibly serving as loci for thermal boundary layer instabilities that rise through the mantle as well as accumulation zones for dense partially molten material right above the core-mantle boundary (e.g., Nakagawa and Tackley, 2004; McNamara and Zhong, 2005). Mineral physics experiments and theory now allow thermal and chemical heterogeneity of these provinces to be estimated based on predictions of elastic parameters (e.g., Murakami et al., 2004; Mao et al., 2006; Duffy, 2008; Ohta et al., 2008; Shim, 2008). Next-generation experimental facilities will provide the ability to characterize textures throughout the mantle pressure-temperature (P-T) range, such as crystal-liquid wetting angles and shape-preferred orientations—features that provide direct constraints on mantle evolution. The locations of large igneous provinces (LIPs) reconstructed for plate motions suggest that the deep-mantle LLSVPs may have persisted for at least 300 My, constituting a long-term connection between deep dynamics and surface geology (Torsvik et al., 2006; Burke et al., 2008). Many questions about the composition and dynamics of these huge chemical heterogeneities remain to be resolved, and petrological and geochemical investigations of surface materials are needed to evaluate possible deep compositions, but their discovery has driven models for mantle evolution in totally new directions.
Disciplinary advances underlying the progress in characterizing deep chemical reservoirs include improved global seismic data sets accumulated from fixed and portable seismic networks; improved three-dimensional (3D) waveform modeling and imaging capabilities for resolving complex, deep structures; improved resolution of 3D thermo-chemical convection models enabled by faster computers and enhanced numerical codes; novel 3D petrographic analyses for lower-mantle conditions enabled by 3D x-ray tomography with nanoscale resolution; greatly expanded experimental determinations of deep-mantle properties enabled by synchrotron radiation facilities; and greatly improved molecular dynamics models implemented on fast computer networks. The rapid accumulation of new data, models, and properties positions the community
FIGURE 2.6 Seismic tomography indicates that the present-day lower mantle is dominated by large low-velocity provinces beneath southern Africa and the south-central Pacific, plus high-velocity regions beneath the Pacific Rim, as shown in Figure 2.5. The evolution of these structures with time is critical to deciphering the origin and composition of mantle reservoirs and their fluxes. This figure shows a simulation of whole-mantle convection with thermal and chemical heterogeneity and reconstructed plate motions since 450 Ma. Left: Calculated mantle structure at 230 Ma with reconstructed plate boundaries in black. Right: Present-day mantle structure with continent outlines in black from the same simulation. Positive and negative temperature anomalies are shown in yellow and blue, respectively; dense chemical heterogeneity is shown in green; the core-mantle boundary is shown in pink. This simulation predicts that a Paleozoic Gondwana LLSVP split to form the African and Pacific structures. It illustrates how plate and continent reconstructions can be combined with seismic tomography, LIPS paleo-reconstructions, and geodynamical modeling to trace the evolution of present-day mantle structure into the deep past. SOURCE: Zhang et al. (2010).
to integrate the separate advances into new understanding of thermo-chemical convection throughout the upper and lower mantles, including effects of the subducted lithosphere, deep chemical piles, and thermo-chemical plumes.
While near-term progress can be anticipated based on the improved data, analysis techniques, and facilities that support research on the deep Earth system, final resolution of many of the key issues will require a significant improvement in high-resolution observational, theoretical, experimental, and modeling capabilities. On the observational end, the primary challenge is the big step to fully 3D seismic imaging with short scale-length resolution. This is achieved in the shallow oil exploration industry using very fine wavefield sampling that is not approached by current global seismic networks or even large-scale deployments of continental-scale arrays such as the EarthScope transportable array (e.g., Rost et al., 2008). There is a need for moderate aperture (~100 km) dense (100 to 200 stations) broadband arrays deployed in multiple locations around the world that can provide high-resolution imaging of specific regions of the deep mantle within the large-scale framework structures that can be imaged by existing global networks. An “Array of Arrays” concept is being developed in the seismological community as a means to achieve the high-resolution capabilities essential to resolving detailed structures in boundary layers, in deep subducting slabs, and in deep plumes as well as for improving models of statistical heterogeneity of small-scale structures that cannot be deterministically imaged. This undertaking will require strong international partnerships.
Advances in theoretical and computational capabilities for 3D seismic processing, for ab initio mineral physics calculations of material properties, and for multiscale 3D spherical geodynamics are all required to take a big step forward in resolving fine-scale structures and dynamics. Access to massive computational resources is also needed for dealing with the complexity of high-resolution seismological imaging and modeling, theoretical mineral physics, and especially global geodynamics calculations. These global geodynamics calculations will include fine-scale boundary layers on thermal and chemical boundaries, phase changes including iron (Fe) spin-state transitions, and partial melting effects, with long-time evolution. Improved experimental resolution of high P-T elasticity and
transport properties will also be required, which will likely involve establishing new NSF-supported analysis nodes on large U.S. Department of Energy (DOE) high-energy facilities. The overall scope of facilities needed to make the next large steps in understanding the deep Earth thermo-chemical dynamic system will likely require major instrumentation initiatives and interagency partnerships. While the EarthScope project is completed over the next decade, planning efforts will need to be undertaken throughout the decade to achieve the capabilities needed for resolving key deep Earth system controversies.
Quantification of Mantle Volatile Fluxes
The stored quantity and flux of water into and out of the mantle are critical factors for sustaining life, facilitating plate tectonics (by making faults weak and lowering the viscosity of the mantle), and creating volcanism. As the universal solvent, the flux of water is intimately connected to most geochemical and volatile cycles and hence to the weathering of continents and the formation of mineral and ore deposits. A basic understanding of the dynamic Earth cannot be achieved without quantitative knowledge of the distribution and behavior of water and the feedback between the water cycle in the solid Earth and the climate system. Yet the sign of the net flux of water between Earth’s interior and the near-surface hydrosphere is not even known (e.g., Hirth and Kohlstedt, 1996; Bercovici and Karato, 2003; Ohtani et al., 2004; Hirschmann, 2006; Olson, 2010).
Likewise, the deep interior plays a critical role in the global carbon cycle, and carbon can also alter physical properties of the mantle yielding feedbacks between carbon cycling and mantle dynamics (see Figure 2.7). As is true for H2O there are great uncertainties in the distribution and flux of carbon (e.g., Dasgupta and Hirschmann, 2010). Most of Earth’s carbon is stored in rocks, with much of that carbon in the mantle. Most mantle carbon is stored in high-pressure minerals, and volcanic processes provide the mechanisms for transferring some of this carbon to the atmosphere, while subduction provides the main mechanism for its return to the interior. Because the mantle carbon reservoir is thought to be large, resolving the internal component of the global carbon cycle is vital to interpreting the record of long-term climate changes. The broad scope of understanding Earth’s carbon cycle from crust to core will require the expertise of geologists, physicists, chemists, and microbiologists. For example, discoveries of microbial life deep in the crust beneath both the oceans and continents indicate a rich subsurface biota that by some estimates may rival all surface life in total biomass. The subduction of tectonic plates and volcanic outgassing are primary vehicles for carbon fluxes to and from deep within Earth, but the processes and rates of these
FIGURE 2.7 Cycling of volatile molecules through Earth’s mantle is known to have an important role in regulating the level of carbon dioxide in the atmosphere. Mineralogists have discovered that many high-pressure minerals, such as the wadsleyite form of olivine present in the transition zone, can contain large amounts of water as hydrogen dissolved into their crystal structures. Current research points to a large fraction of our planet’s volatile budget being locked up inside the solid Earth (Kellogg et al., 2004). SOURCE: Figure provided by R. Dasgupta and M. Hirschmann.
fluxes—as well as their variation throughout Earth’s history—remain poorly understood. For example:
• Is biologically processed carbon represented in deep Earth reservoirs?
• What are the physical and chemical processes that govern carbon’s distribution in Earth?
• How do carbon's elemental character and behaviors impact its various roles in the Earth system?
The current opportunity to improve our understanding of volatile fluxes in the interior also derives from improvements in high-resolution imaging of internal structures and material properties with seismology and magnetotellurics, especially in regions of both active and ancient subduction, in new petrological and volcanic constraints on subduction zone volatile fluxes, in high-resolution 3D geodynamical modeling capabilities for subduction zones with volatile transport and mineralogical reactions, and in mineral physics characterization of the myriad hydrous phases, dehydration processes, and influence of volatiles on rheology and the elastic properties imaged by seismology. Concerted community efforts to study subduction zones such as GeoPRISMs bring together diverse research communities that can address the volatile budget and flux problem, and large-scale studies of upper-mantle structure such as those conducted under the Continental Dynamics and EarthScope programs now regularly cast interpretations of seismic models in terms of coupled thermal, volatile, and chemical heterogeneities rather than solely thermal models (see Figure 2.8). With great expansions of seismological databases that can be anticipated over the next decade, in parallel with improved characterization of rheological and elastic attributes that reflect volatile presence and abundance, significant progress on mapping volatile distributions and resolving volatile fluxes can be anticipated with sustained research investment.
Quantification of Core Evolution
Our knowledge of Earth’s core has advanced greatly over the past few decades, albeit with continued surprises time and again (e.g., Nimmo, 2007). As the core cools, the inner core grows by solidification of iron at its surface accompanied by a depletion of the light alloy component. It was originally assumed that this would lead to a relatively homogenous inner core
FIGURE 2.8 Images of Vp/Vs seismic wave velocity ratio variations in the mantle wedge beneath Central America. Low-velocity regions in the wedge may involve either fluids extracted from the slab or regions of partial melting caused by fluid-assisted reduction of melting temperature extending upward from the slab/wedge interface. SOURCE: Syracuse et al. (2008).
structure, possibly with some thin surface transition zone. Seismology demonstrated that the inner core has both small-scale and large-scale heterogeneities that appear to reflect dynamical processes. Early characterization of the heterogeneity demonstrated the presence of anisotropic structure closely aligned with the rotation axis, but it is now recognized that there are hemispherical patterns in the inner core structure as well as changes in anisotropic pattern with depth (e.g., Ishii and Dziewonski, 2002; Song, 2007). Parallel improvements in seismological constraints on outer core structure indicate that there is a region above the inner core that has reduced velocity gradients indicative of transitional properties (Zou et al., 2008). Greatly expanded geodynamo simulation capabilities have also explored thermal, electromagnetic, and dynamic coupling of the inner and outer core regimes, seeking constraints on the inner core growth mechanism and outer core energy budget (see Figure 2.9). Coupling between the mantle and outer core and gravitational interaction between the mantle and inner core have been explored with improved geodynamic simulations constrained by orbital observations. Paleomagnetic observations have documented Earth’s early magnetic field behavior back to at least 3 billion years ago (see Box 2.3), providing valuable constraints on geodynamo variations linked to inner core growth (Tarduno et al., 2007). All of these approaches to quantifying core structure and history are building an observational database on which major synthesis of core evolution should be viable over the next decade.
The committee also anticipates major developments in understanding of Earth’s core through static high-pressure experiments and density functional theory calculations. With newly developing high P-T techniques allowing direct access to core conditions, novel experimental probes especially well suited for Fe and its alloys, and advances in theoretical techniques for treating transition metals, the time is ripe for a renaissance in studies that will provide improved understanding of the thermal evolution, seismic structure, growth
FIGURE 2.9 Fluxes of heat and light elements (Si, C, O, S, etc.) from the mostly solid inner core into the molten outer core provide much of the power for the geodynamo and also influence the rate of inner core growth and the thermo-chemical evolution of the core as a whole. New interpretations of these fluxes center on the significance of the seismic F-layer above the inner-core boundary (ICB), which appears to be depleted in light elements compared to the overlying outer core, and the observed dichotomy between eastern and western hemispheres of the inner core. This diagram shows one interpretation, the so-called inner core translation instability, in which the inner core dynamics resemble that of a continental glacier. Freezing on the western side of the ICB releases light elements in buoyant plumes into the outer core, while melting on the eastern side of the ICB releases iron-rich liquid, forming the dense F-layer. SOURCE: Reprinted from Alboussiere et al. (2010) with permission from Macmillan Publishers Ltd., and from Monnereau et al. (2010) with permission from AAAS.
New Opportunities in Rock Magnetism in the 21st century
Rock and mineral magnetism constitute the essential connection between geomagnetic records of the past and the answers to the Grand Challenges (NRC, 2008) to understand the origin and evolution of Earth and the other planets. One such broad challenge is: How strong or weak have the internal geomagnetic and planetary magnetic fields been over the past 4.5 billion years? In particular, what can we learn from magnetism of ancient rocks that can illuminate the intertwined record of the geomagnetic field during the first billion years of Earth’s existence and the formation and growth rate of the solid inner core? In the past decade the rock and paleomagnetic community have proven the feasibility of extracting reliable values of paleomagnetic intensity from 1-billion-year-old single silicate crystals containing magnetite grains that have been protected from subsequent chemical alteration. Figure B2.3 shows an example of magnetic signals now being studied using current scanning Superconducting Quantum Interference Device (SQUID) sensors. But to advance the science to more precise and higher (temporal) resolution records of paleointensity, the use of submicrometer-sized, datable zircons and oxide exsolution structures from the early Archean period is needed.
There are two essential requirements for such progress, and both can be within reach if collaborative and focused efforts are now initiated on two fronts. One is the development of novel SQUID and non-SQUID sensors (e.g., spin-exchange relaxation free or Spin Exchange Relaxation-Free [SERF] method-based) that are capable of measuring submillimeter samples, and signal enhancement techniques for the very small magnetic signals that such scanning techniques will deliver. The second requirement is inherently linked to the first and involves “ground truthing” magnetic measurements that are based on submillimeter samples. Because these samples are single crystals, there are a number of rock magnetic effects that must be examined in order to ensure that they are accurate recorders of Earth’s magnetic field. These effects include remanence anisotropy due to crystallographic alignment of magnetic oxides within the silicate host, magnetostatic interactions between inclusions, and subsolidus exsolution structures within the oxides. Measuring the importance of these effects will require the use of instruments capable of imaging magnetism at scales of 10 to 1,000 nm, such as transmission electron microscopes and magnetic force microscopes. Ultimately these kinds of studies will allow researchers to select only those samples that can be confidently used for reconstruction of geomagnetic paleointensity for such ancient times.
FIGURE B2.3 Example of current scanning SQUID microscopy with submillimeter (~100 micrometer) resolution (A) showing geomagnetic reversal stratigraphic dating of alternating polarities recorded by magnetite crystals in submillimeter layers of a seafloor manganese nodule (B). The nodule is only 35 cm thick, and the alternating magnetizations can be fit to a known polarity reversal timescale (C). SOURCE: Reprinted from Oda et al. (2011) with permission of Geological Society of America.
mechanism, magnetic field generation, and dynamic behavior of the core.
The next decade offers the potential for building on and integrating the recent observational, computational, and experimental advances noted above into a robust model for evolution of the inner and outer cores. The thermo-chemical evolution of the core dynamic system is manifested not only in the geomagnetic field but also in the thermal history of the planet; the rate of inner core growth is determined by how rapidly the core cools, which is controlled by the mantle. Thus, a bounty of fundamental results can be harvested by developing a quantitative understanding of core evolution. The NSF’s Cooperative Studies of the Earth’s Deep Interior (CSEDI) program is structured to support interdisciplinary coordination on this topic, and community organizations such as the Cooperative Institute for Dynamic Earth Research (CIDER) enhance communications across the disciplines and training of graduate students in the diverse arena of core studies.
Plate tectonics provides a first-order description of how Earth’s surface shifts with time, with the motions near plate boundaries largely involving seismic or aseismic faulting and elastic or anelastic rock deformation. Plate motions driven by mantle flow concentrate stresses on faults at plate boundaries, powering the cycle of frictional stress accumulation, elastic and anelastic strain deformation, and slow or abrupt (earthquake) fault displacement and stress and strain release. Ground motions caused by elastic waves and surface deformations produced during rapid earthquake faulting constitute one of nature’s greatest hazards, with tremendous annual loss of life and damage on a global basis. The impact of earthquakes can be staggering; hundreds of thousands of fatalities in moderate-size events like the 2010 Haiti (magnitude Mw 7.0) earthquake or immense events like the 2004 Sumatra (Mw 9.2) earthquake and tsunami, and hundreds of billions of dollars in damage as in the 2011 Japan (Mw 9.0) earthquake. Since 2004 there have been more great earthquakes around the world than in any 6.5-year period in seismological history (back to 1900), and burgeoning population growth near plate boundaries will place ever-increasing populations and built infrastructure at risk throughout this century. Efforts to understand how faults accumulate and release stress and strain and the nature of the resulting ground motions constitute major scientific challenges highlighted in community planning documents from seismologists (Lay, 2009), geodesists (UNAVCO, 2008), geodynamicists (Olson, 2010), GEOPrisms (MARGINS Office, 2010), and the EarthScope program (Williams et al., 2010). The 2008 NRC report Origin and Evolution of Earth highlighted the question of whether earthquakes, volcanic eruptions, and their consequences can be predicted, as one of 10 Grand Challenges in the Earth sciences.
Earthquake science is intrinsically interdisciplinary and deals with complex multiscale dynamical systems spanning the microscale processes of friction and fluids in fault zones to the macroscale processes of elastic and anelastic crustal deformations and elastic waves in the crust and near-surface environment. Geologists provide a framework for studying deformation near plate boundaries by documenting the style and timing of faulting over geological time, and by examining exhumed faults to study frictional characteristics and evolution of fault gouge. Seismologists use the elastic wave energy radiated from dynamic fault ruptures to estimate the size of earthquakes and to determine details of the rupture process, along with quantifying seismic wave propagation and ground shaking effects. Geodesists measure deformations of the rock around a fault zone both before (interseismic), during (coseismic), and after (postseismic) an earthquake, as well as stable sliding of some aseismic faults. Rock mechanics researchers determine frictional mechanisms and theory for rupture nucleation and arrest to guide understanding of the frictional instabilities associated with earthquakes and stable sliding. The collective scientific understanding of earthquake faulting from these endeavors feeds into earthquake engineering and emergency response efforts to mitigate the impacts of fault ruptures.
The earthquake cycle notion provides a basic framework for understanding deformation near a fault that is loaded by large-scale plate motions. Once an earthquake has occurred and the postseismic period of stress and strain transients has ended, the earthquake cycle begins anew with interseismic frictional locking of the fault and onset of fault zone strain accumulation. Geological, seismic, and geodetic data are used to evaluate the size and frequency of large earthquakes in a particular region. A catalog of historical behavior of a fault is then used to assess how large and how often fault ruptures can be expected statistically. Geodetically determined rates of strain accumulation can be evaluated relative to total plate motions and stress drop determinations for prior events on the fault to anticipate where and how much future strain release will occur. Determining the statistical likelihood of earthquakes in a region is of particular interest to society because engineering building codes are guided by the probability of experiencing various levels of ground shaking within the lifetime of a building. However, this earthquake cycle model is only useful to the extent that we can fully understand how deformation accumulates and how faults fail. The nonlinearity of frictional instabilities, the influence of dynamic and static stress perturbations by other earthquakes, and the complexity of stress heterogeneity from prior ruptures and non-uniformity of strain accumulation all add uncertainty to forecasting future earthquake occurrence.
Recent Advances—The Wide Range of Slip Velocities
The scientific view of how faults slip has evolved dramatically in the past decade. Developments in space geodesy—particularly the Global Positioning System (GPS) and Interferometric Synthetic Aperture Radar (InSAR)—allow the interseismic deformation phase of the earthquake cycle to be imaged with unprecedented temporal (GPS) and spatial (InSAR) sensitivity. Prior to the development of large GPS arrays, measurements of deformation in fault zones were either unavailable or sporadic. As GPS resolution improved to the several millimeter level, it became clear that overall strain accumulation does not always follow a simple linear model (see Figure 2.10). Instead, observations show that steady accumulation of deformation in the volume adjacent to a plate boundary can be punctuated by abrupt changes in sign, indicating non-seismic slip of portions of megathrust faults. These “slip reversals” were first observed at multiple stations in the Cascadia region within the past decade and have been termed either “slow slip” or “slow earthquakes.” In Cascadia the slow slip events occur about every 14 months (Miller et al., 2002) and are thought to involve intermittent shearing displacement of the down-dip region of the megathrust in a transition zone from unstable to stable sliding, partially relaxing strain in the upper plate. It is plausible that this is a primary mechanism that helps load and initiate large earthquake ruptures on the shallower unstable sliding regime of the plate boundary.
Just as slow earthquakes were first being recognized, seismologists made an additional discovery, classifying a new kind of seismic signal: non-volcanic tremor (Obara, 2002). Tremor consists of long-duration trains of weak ground motions that do not have easily identifiable body wave arrivals. Soon after slow slip events were discovered, it was shown that the slow slip episodes in Cascadia correlate with periods of enhanced seismic tremor, with the term episodic tremor and slip (ETS) being used to describe the combined phenomena
FIGURE 2.10 Comparison of GPS observations of upper plate displacement and seismic tremor activity levels for the Cascadia subduction zone. Cyan dots represent daily location solutions for the East-West component of the Victoria GPS station, with the overall eastward trend (green) representing the upper plate deformation caused by convergence between the frictionally locked (not slipping) shallow megathrust fault between the Juan de Fuca and North American plates. Every ~14 months the trend of the GPS locations reverses direction for ~2 weeks away from the secular trend, which is inferred to result from the deeper portion of the megathrust fault slipping slowly, relaxing some strain in the upper plate even while the shallow portion of the fault is still not slipping. Blue lines represent hours of non-volcanic tremor in each 10-day window recorded in the same region. There is a positive correlation between the times when the GPS displacement has a reversal and periods of strong tremor activity. SOURCE: Reprinted from Rubinstein et al. (2010) with permission from Springer Science+Business Media. Modified and extended from Rogers and Dragert (2003) with permission from AAAS.
(Rogers and Dragert, 2003). Efforts to establish the direct relationship between tremor and slow slip are under way, with possibilities including heterogeneous frictional conditions on the deep megathrust as well as activations of multiple faults due to fluid motions and changes in strain produced by the slow slip. ETS has now been reported in many other subduction zones but with variable manifestations. In southern Mexico the slip events are as much as five times larger than in Cascadia and less frequent (Kostoglodov et al., 2003); tremor has also been observed in the Mexican subduction zone (Payero et al., 2008). Episodic slip events have been reported in the New Zealand Hikurangi subduction zone (Douglas et al., 2005; Wallace and Beavan, 2006), while tremor was elusive (Delahaye et al., 2009) until recently observed (Kim et al., 2011). The diversity of fault slip processes has been particularly well documented in Japan, where there is high density of both geodetic and seismic instrumentation. For example, borehole tiltmeters in the Nankai Trough have been used to detect slip events that were much too small to be identified on GPS receivers. The migration of both tremor and slip on the fault zone interface was subsequently imaged, with migration speeds of ~10 km per day (Obara et al., 2004), comparable to observations in Cascadia. Earthquakes depleted in short-period radiation (low-frequency earthquakes, or LFEs) have been identified near the down-dip edge of the unstable megathrust zone (Katsumata and Kamaya, 2003), and it currently appears that tremor involves superposition of many small LFEs (or even normal earthquakes).
Conventional earthquakes involve large amounts of energy release in small amounts of time, with rupture spreading over the fault at very high velocities of several kilometers per second (see Figure 2.11). The ETS and LFE observations make it clear that fault slip occurs on an immense variety of temporal scales that appear to scale differently than for fast ruptures. Some of the large slow slip events have the equivalent strain release of large conventional earthquakes (e.g., Kostoglodov et al., 2003). In some regions, where the total seismic slip budget falls very short of the total plate tectonic convergence budget, such as the Marianas and Tonga subduction zones, the entire megathrust may be fail-
FIGURE 2.11 Relationship between the duration and magnitude of regular earthquakes (thick blue line) and low-frequency earthquakes (LFEs, red), very low-frequency earthquakes (VLFs, orange), and slow slip episodes (SSEs, green) in the Nankai trough off the coast of Japan and episodic tremor and slip (ETS, light blue) in the Cascadia subduction zone off the coast of the U.S. Pacific Northwest. Pink dots are silent earthquakes; black symbols are slow events. See original source for further explanation. SOURCE: Reprinted from Ide et al. (2007) by permission from Macmillan Publishers Ltd.
ing in slow slip or stable sliding processes, as appears to be the case along North Island, New Zealand. This indicates the importance of understanding the full range of frictional processes that appear to play a huge role in plate motions. The simple earthquake cycle model that has been invoked for decades needs to be expanded to accommodate these new observations. Much of the effort thus far has focused on categorizing these events—where and when they occur and how big they are. And while slow slip appears to be related to a frictional behavior intermediate between that of steady sliding and stick-slip earthquakes, theoretical developments are needed in order to make advances in our understanding of these new observations of fault slip. Laboratory studies of rock mechanics spanning the full range of fault slip velocities play a key role in quantifying the observations.
Rapid progress in this area can be sustained because of major observational facilities such as EarthScope being deployed with sufficient station density to capture the full spectrum of fault behavior. Organized community efforts such as GeoPRISMS and the Southern California Earthquake Center (SCEC) draw together the interdisciplinary communities working on faulting and earthquake processes at all scales, and the level of research excitement and activity provides a clear opportunity for major advances on this topic during the next 10 years.
Recent Advances—Dynamic Fault Modeling
Significant breakthroughs are also being made in our understanding of seismic radiation from earthquakes. Some of the most exciting quantifications of global earthquake ruptures in the past decade have come from innovative use of regional arrays, showing the expansion of fault rupture for the recent immense earthquakes in Sumatra, Chile, and Japan (e.g., Ishii et al., 2005; Lay et al., 2010a). Global seismic network data are now used to estimate slip distribution for all major faults; geodetic data sets using GPS, InSAR, uplift, and tsunami excitation have improved constraints on fault displacements for events around the world. Faults likely to rupture at super shear velocities have been identified (Bouchon and Vallee, 2003), and the complexity of faulting beyond simple slip pulse models has been resolved (e.g., Lay et al., 2010b). However, many fundamental questions about earthquakes remain:
• How do earthquakes initiate?
• What controls the branching of rupture or the triggering of one fault by rupture of another?
• Why does a rupture stop?
• When and why do rupture speeds exceed the seismic shear velocity?
• Can rupture attributes be anticipated based on geodetic determinations of prior fault locking and strain accumulation?
To mitigate risk from earthquakes, it is first necessary to know how strongly the ground will vibrate. This is difficult to predict given both the complexity of earthquake ruptures and the wave focusing and defocusing effects and soil interactions of seismic waves. At present, ground motions during earthquakes are usually characterized by very simple measurements, such as peak ground acceleration or velocity. These data are used by engineers to estimate the strength of ground shaking expected during an earthquake of given size by using empirical relationships based on past earthquake data in a given region (size of earthquake, distance to the rupture, and local geology). This approach seems to work adequately for moderate earthquakes, but rupture finiteness and wave directionality effects for large events greatly complicate the ground motion prediction. Because large (and very large) earthquakes occur infrequently, the empirical-based seismic hazard relationships are not well constrained, and recent earthquakes have offered repeated surprises in terms of the intensity of ground shaking actually experienced. It is desirable to move forward from empirical approaches to quantitative modeling approaches.
Most seismic and geodetic models of fault slip are kinematic in nature; simplifying assumptions are made to allow the estimation of the relevant parameters (e.g., faults are planar, slip is unidirectional). Physical properties of the fault are typically not modeled because of their complexity. However, new simulations have shown the potential to bridge the gap from standard kinematic models to physics-based models (e.g., Dunham and Archuleta, 2005). Dynamic rupture modeling considers the joint stress-slip evolution during earthquake shear failure as being driven by the redistribution of
stored strain energy and can serve as the foundation for predicting both fault behavior and strong ground motion. Dynamic rupture modeling includes realistic 3D simulations of fault roughness; spatially variable frictional properties; and other effects, such as basin reverberation and focusing, soil nonlinearity, and soil-structure interactions. Quantitative modeling now provides a prospect of eliminating dependence on poorly constrained empirical models, thus linking seismic hazard analysis for the first time to physics-based concepts such as stress-time evolution.
Much of this work is currently coordinated by the SCEC, where there is a community effort to develop 3D rupture models with full 3D implementations, from finite-element codes to high-resolution 3D community crustal models (Olsen et al., 2008). These models are currently being used to predict shaking in Los Angeles, San Francisco, and other cities from ruptures on the San Andreas Fault or other regional faults. This has been an interdisciplinary effort bridging rock physics, seismology, soil mechanics, structural geology, and earthquake engineering and requires the use of today’s most powerful supercomputers because representations of faults must span spatial scales covering many orders of magnitude and because physical quantities must be calculated at all causally connected points to properly account for stress and slip evolution.
Advancing earthquake source studies to a full physics-based model of initiation, propagation, and arrest requires knowledge of the stresses on faults, how those stresses change with time, and the influence of pore fluid pressure. Resolving these questions requires improvements in computational resources and support for theoretical developments so that 3D wavefields can be computed for realistic crustal environments. Furthermore, additional ground displacement records recorded near faults during large earthquakes are needed to test the results of dynamic rupture models. While much of the San Andreas Fault has been instrumented (by other government agencies) with strong motion sensors, accelerometers do not directly record ground displacements and cannot distinguish rotations from accelerations. Combining strong motion records with GPS position estimates (in the same way that GPS is often combined with more precise gyroscopes in navigation systems) would address the limitations of strong motion data. It is desirable to support collocation of strong motion sensors when GPS receivers are installed in fault zones.
The recent earthquake drills, or ShakeOuts, conducted in California (Perry et al., 2008) and since expanded1 to Nevada, Utah, Oregon, Idaho, the central United States, British Columbia, and Guam have used realistic shaking simulations to guide the responses of millions of people to scenario events. The effort has just begun, and as computers, 3D methods, and the interfaces between the scientists and engineers, scientists and first responders, and societal engagement improve, this area will greatly expand (see Box 2.4). NSF’s role includes interagency engagement with the U.S. Geological Survey (USGS) in SCEC, along with direct funding of many related basic research efforts in each component (theory, computational support, new observations) that feed into this hazard area. This work has the potential to transform probabilistic hazard analysis and to greatly enhance public preparedness for earthquake disasters.
With the recent demonstration that physics-based approaches to probabilistic seismic hazard analysis are both viable and important, the research opportunity is clear: further coordinate the interdisciplinary effort to advance understanding of dynamic failure at all scales from fault zone to remote ground shaking. This ambitious effort is under way, and sustaining it should provide major advances over the next decade.
Recent Advances—EarthScope Project
EarthScope is the first Major Research Equipment and Facilities Construction (MREFC) project conducted by the Earth sciences, receiving $200 million of NSF’s MREFC support from outside of the Directorate for Geosciences (GEO) and an increase in GEO/ EAR annual funding that provides ongoing Operations and Maintenance (O&M) support projected to continue through at least 2018. EAR has built up Earth-Scope research funding steadily by designating funds from divisional budget increases since the onset of the project. The success of EarthScope is critical to establishing a precedent for future efforts to draw MREFC funding to the discipline. The fact that the 2003-2008 EarthScope facilities construction phase was completed
Near Real-Time Analysis of Earthquakes and Volcanic Eruptions
With rapid growth of human population, society faces increasing exposure to catastrophic effects of earthquake faulting, tsunamis, and volcanic eruptions. As basic scientific investigations of these phenomena advance, a natural result is that observational and analytic procedures mature to the point where they can be robustly and rapidly applied, even while the event is under way. This exercise of scientific understanding can enable development of real-time hazard warning systems to society’s great benefit, both from early warning of imminent shaking or tsunami arrivals and by providing guidance to effective post-event emergency response activities. While continuous environmental monitoring is typically the function of mission-driven agencies, development of the fundamental understanding on which real-time warning capabilities can be based involves NSF-funded research on natural phenomena.
Early warning systems rely on continuous acquisition of data from potential source regions, and real-time telemetry of the data or local analysis products for events as they occur to central processing centers where the signals enable near real-time evaluation of the process and its hazards, launching appropriate communications about the event and its potential distributed impact. For earthquakes and volcanic eruptions, such methodologies can exploit the finite velocity with which seismic, tsunami, or air-blast waves spread from the source relative to electronic communications to warn nearby regions before the waves arrive. Automated systems that sense initial signals can also activate immediate responses locally to mitigate the impact of later-arriving signals. These strategies are exemplified by ocean-scale tsunami warning systems, such as the National Oceanic and Atmospheric Administration’s (NOAA) Pacific and Alaska Tsunami Warning Systems, and the Shinkansen (Japanese bullet train) accelerometer system for stopping trains when P wave ground motions exceed certain thresholds (in advance of later-arriving, stronger S wave and surface wave ground motions). The potential for many applications to mitigate shaking damage given from seconds to hours of lead time after occurrence of an event is just beginning to be explored.
Rapid analysis and warning of large earthquake ruptures can potentially be achieved with integrative approaches using geodetic (continuous, high-sample-rate GPS), seismic (rapid local network event location, mechanism, and finite faulting determination), and ocean measurements (water pressure and geodetic systems that detect tsunami waves offshore). Such applications are rapidly emerging, and over the next decade significant enhancements and impacts from these capabilities can be exploited. The Cascadia subduction zone is being instrumented onshore and offshore using EarthScope and American Recovery and Reconstruction Act funding. These regional seismic and geodetic networks within one venue of natural hazard exposure are valuable for advancing the basic science underlying rapid warning capabilities. Significant progress has been made in developing remote tsunami warning capabilities, but close-in tsunami warning, where warning response times of only tens of minutes are required, presents great challenges. This drives basic science efforts to establish what aspects of large offshore earthquakes can be reliably characterized in ground motion signals soon after an event initiates and the extent to which the ultimate size of the event can be anticipated early in its process. Similar challenges exist for developing rapid warning of volcanic eruptions that present hazards to air traffic. Development of seismic, geodetic, and infrasound analysis that can establish the occurrence of strong tropospheric and stratospheric blasts and ash clouds requires better understanding of explosive eruption processes and their manifestations.
Data from EAR facilities in seismology (IRIS), geodesy (UNAVCO), EarthScope, and community organizations (SCEC, GeoPRISMS) provide the means for coordinated efforts to rapidly analyze signals from active processes. Ultimately, monitoring and implementation of warning systems are the provenance of the USGS and/or NOAA (for Homeland Security), but developing and integrating the scientific approaches remain a basic science problem, as extensive fundamental understanding of the processes and the signals they generate lies at the core of all early warning strategies. Rapid analysis and quantification of earthquake and volcanic processes are also relevant to basic research on dynamic phenomena, especially as interactions between dynamical systems even at long ranges are now being broadly recognized.
on time and on budget has strongly positioned EAR for future MREFC competitions and for National Science Board and congressional support of future Earth science projects. Achieving full success of the project will involve completion of the science plan defined in the original proposal and updated in the EarthScope Science Plan for 2010-2020 (Williams et al., 2010).
The scientific rationale for following through on the EarthScope program in the next decade is compelling. Densification of geodetic and seismic observations along the plate boundary on the western coast of the United States and along the Alaska-Aleutian volcanic arc has already resulted in exciting discoveries about faulting and deformation processes described above. Seismic, geodetic, magnetotelluric, and geochemical data collected by EarthScope are progressively revealing deep crustal and upper-mantle structures under North America, unveiling as the Transportable Array sweeps eastward. Fundamental questions such as the deep configuration of the Juan de Fuca plate, the fate
of other subducted portions of the Farallon plate, deep crustal delamination processes under the Basin and Range and deep structure of the Colorado Plateau and Rio Grande rift, the detailed structure and mantle flow beneath the Yellowstone volcanic center, and the lithospheric contrasts across the Rocky Mountain front are all being vigorously addressed with hundreds of papers appearing (see Figure 2.12). Large-scale deformation of western North America is being revealed by the geodetic instrumentation with unprecedented resolution (see Figure 2.13). Prospects are good for resolving many long-standing large-scale framework questions about the driving processes for North American geological history. Further eastward migration of the Transportable Array will expose unknown structures beneath the eastern continental margin and then across Alaska, where there have been relatively few seismic instruments. Unraveling the complex history and processes of North American evolution has commenced but will require the synoptic framework structures anticipated from the full EarthScope program. As this framework emerges from the NSF-led effort, interagency coordination may help this understanding to penetrate into mission agencies such as the Department of Energy, the U.S. Geological Survey, and the Nuclear Regulatory Commission, all of which have programs impacted by earthquake hazards and continental deformation related to topics such as carbon
FIGURE 2.12 One of the goals of EarthScope is to resolve the upper-mantle structure beneath the North American continent using seismic signals and to interpret the dynamical processes by which the continent has evolved. The example shown here is the high-resolution determination of the 3D structure under the Great Basin. Source: Reprinted from West et al. (2009) by permission from Macmillan Publishers Ltd.
FIGURE 2.13 Example of the remarkable spatial resolution of the crustal deformation field in the western United States determined by the Plate Boundary Observatory geodetic instrumentation. SOURCE: McCaffrey et al. (2007). Reprinted with permission of John Wiley and Sons.
sequestration, geothermal energy, fracking for shalegas recovery, nuclear power plant siting, and building code development.
The economic rationale for sustaining the EarthScope project through the planned program to 2018 is equally compelling given the large investment of NSF funds in EarthScope, the superb success of the facilities in achieving the primary data collection goals to date, the exciting scientific results on first-order Earth science problems, and the excellent prospect for
sustaining the flow of discoveries and resolving long-standing questions. After 2018 any continued elements of the project will need to be carefully assessed and evaluated in terms of prospects for proportionate advances. The NSF system for MREFC programs causes particular stresses for directorates that have not had prior MREFC initiatives (the need for creation of O&M and research funds within the directorate to follow up on the infusion of MREFC capitalization funds) and the successful completion of EarthScope may ease the establishment of new EAR MREFC programs. With aspirations for major new Earth sciences facilities being articulated by multiple EAR communities, future MREFC proposals should be at least one strategy considered by EAR management.
Natural Laboratory Strategy
Research on faulting and deformation processes can be conducted over a wide range of efforts, spanning single-investigator theory and laboratory efforts to integrated field activities. It is essential to sustain the former, while the latter has become the focus of large-scale community efforts and NSF programs, exemplified by the SCEC, the Margins and Ridge initiatives, Continental Dynamics projects, and Earthscope. For the next decade several regions have been identified by GeoPRISMS as important natural field laboratories for coordinated efforts; these include Alaska and Cascadia, along with North Island, New Zealand. All of these present opportunities for increased involvement of EAR over the previous Margins program.
The Alaskan subduction zone provides a second natural laboratory to study fault zone processes. This zone is complex, with significant variations in geometry and locking and more frequent magnitude 7.0 to 8.0 earthquakes and volcanic eruptions than Cascadia (see Box 2.5). There is an existing GPS network (~150 stations) in Alaska, maintained by the Plate Boundary Observatory (PBO). The committee anticipates two significant and complementary research and instrumentation efforts in Alaska in the next decade. First, the EarthScope Transportable Array will arrive in Alaska in 2014 if the next phase of EarthScope operations is sustained, extensively increasing the on-land seismic network, which has been sparse relative to the huge tectonically active domain. Second, GeoPRISMS recently announced that the Alaskan subduction zone will be one of its primary scientific targets, which means that offshore seismic sensors will likely become available. The combination of seismic and geodetic instrumentation and the synergism with GeoPRISMS science objectives will allow unprecedented opportunities for fault zone earthquake and deformation studies. EAR collaboration with the NSF Division of Ocean Sciences (OCE) could ensure optimal usage of the scientific data collected in Alaska.
While natural laboratories in Cascadia, Alaska, and New Zealand present excellent opportunities for research on faulting processes, it is desirable to pursue an ultimate goal of instrumenting all accessible fault zones. Progress can be made by taking advantage of interdisciplinary collaborations. For example, EAR is co-sponsoring the installment of a 50-station GPS network in the Caribbean.2 EAR’s goals for this effort are to assess seismic hazards in the region. The NSF Division of Atmospheric and Geospace Sciences (AGS) is co-funding the network because the same GPS data can be used to help atmospheric scientists predict the intensification and direction of tropical storms and hurricanes. In addition to partnering within GEO and with other NSF directorates, EAR can continue collaborations to maintain networks with other government agencies that use seismic (USGS, DOE) and geodetic instrumentation (NASA, NOAA). Innovative uses of existing networks and facilities should be encouraged, including applications to hydrology and meteorology, to broaden the support base for these data collection efforts.
One of the major advances in the Earth sciences over the past decade was the recognition and verification of broad connections between climate, surface processes, and tectonics. The NRC Landscapes on the Edge (2010a) report identified research questions that center on interactions among climate, topography, hydrology and hydrogeology, physical and chemical denudation, sedimentary deposition, and rock deformation in
Volcanic eruptions provide spectacular and frequent (more than 70 different volcanoes erupt every year) reminders that Earth is a dynamic and evolving planet. Lava flows, pyroclastic flows, and ash fall are proximal hazards; gases and dust lofted into the atmosphere have global effects on climate, life, and air traffic. Volcanic hazard does not end with the eruption—lahars and landslides create hazards long after an eruption ends. Despite a long history of investigation, numerical models of volcanic processes, laboratory characterization of the properties of magmas, and real-time monitoring of active volcanoes are only now beginning to show their promise to both predict eruptions and quantitatively interpret volcanic deposits.
Volcanic eruptions are the end product of a complex set of interacting processes: melting Earth’s interior, the storage and chemical evolution of magma, the ascent of magma through the crust, and the fragmentation of magma during explosive eruptions. Several key questions remain the subject of active research. Why do volcanoes erupt in so many different ways? Can the duration and style of eruption be predicted from pre-eruption signals? Why do super-volcanoes exist? Why do earthquakes sometimes trigger volcanic eruptions? What processes govern the speed and distance traveled by pyroclastic flows?
Modern research in volcanology relies on integrating complementary approaches: remote sensing from space with InSAR and spectroradiometers; distributed high-frequency monitoring of GPS, tilt, seismic, infrasound, acoustic, and electromagnetic signals; gas sampling; measuring the rheological properties and phase equilibria of magmas in the lab; and numerical simulations of conduit processes, the multiphase dynamics of eruption columns and pyroclastic flows, and the thermal and chemical evolution of magma within the crust. Additionally, large-scale laboratory experiments offer an important opportunity for validating the new generation of numerical models for conditions and properties that are well constrained. At the present time, NSF does not support either such large-scale laboratory facilities for community use or experimental facilities for studying magma properties at relevant deformation rates and temperatures.
Monitoring of volcanoes in the United States is performed by the USGS and its volcano observatories. NSF-supported research adds to these activities by supporting complementary principal investigator–led monitoring, theoretical work, and laboratory analyses. Partnerships and collaborations between NSF and other agencies, such as the USGS, may be vital for making full use of the data and addressing questions that are beyond the primary objective of hazard assessment. Support is also needed to rapidly respond to new eruptions and to ensure that instruments are available.
FIGURE B2.5 Some of the phenomena at volcanoes that transport mass and energy to the surface and create volcanic hazards. Modern volcanology seeks a quantitative understanding of these processes and their interactions. SOURCE: Myers et al. (2008).
tectonically active mountain belts as particularly intriguing. While the feedbacks between tectonics, climate, erosion, and deposition have been the focus of field studies and numerical simulations over the past decade, elucidating connections between these processes continues to drive discoveries. Such feedbacks influence the sensitivity of landscape response to climate change and involve numerous complex interactions among climatic, geological, and geomorphological processes. Our understanding of the dynamics of landscape evolution and the linkages between climate, Earth surface processes, and tectonics across a wide range of spatial and temporal scales is ripe for substantial advances now that the advent of thermochronometric methods provides data on erosion rates over geological timescales, cosmogenic methods for dating geomorphological surfaces have matured to the point of being readily accessible to researchers across the field, and high-quality digital topography (such as LiDAR) is increasingly available for regions around the world (see Figure 2.14).
Development and elaboration of transport laws offer the potential to connect studies of active processes with their signatures in landscapes and the related sedimentary and climatic record. In addition, recent studies have highlighted the importance of regional context in sorting out controls on landscape development and evolution as competing theories are seen to have more or less explanatory power in different physiographic, tectonic, and climatic settings. For example, numerous studies have documented evidence for the operation of a so-called “glacial buzzsaw” through which efficient glacial erosion above the glacial equilibrium line altitude (ELA) limits the height of mountains (Brozović et al., 1997; Mitchell and Montgomery, 2006; Enghold et al., 2009). In contrast, glaciers in the southern Andes have the opposite effect and instead shield alpine topography from erosion and thereby enhance elevation (Thomson et al., 2010). Likewise, a recent study that reviewed global erosion rates found that, contrary to the often invoked conventional wisdom that glaciers are the most efficient erosional agents, erosion by rivers can keep up with glacial erosion in tectonically
FIGURE 2.14 Combing LiDAR data with geological observations allows the response of erosional processes in small drainage basins to rock uplift to be determined for the first time in the field at a Dragon’s Back pressure ridge along the San Andreas Fault. These types of detailed measurements were not possible prior to the advent of LiDAR mapping. (A) Airborne Laser Swath Mapping (ALSM) topography (1-m digital elevation model); (B) geology; (C) total rock uplift (~140 k.y.) inferred from distribution of geological contacts; and (D) instantaneous rock uplift rate. SOURCE: Reprinted from Hilley and Arrowsmith (2008) with permission of Geological Society of America. See original text for further explanation.
FIGURE 2.15 Relationship between glacial, fluvial, and composite landscape erosion rates and the contributing basin area, as measured by sediment yield data collected over a 20-year period. Black symbols refer to glaciated basins; gray and open symbols indicate river basins. PNW refers to river basins in the U.S. Pacific Northwest. SOURCE: Koppes and Montgomery (2009). Reprinted by permission from Macmillan Publishers Ltd.
active mountain belts (see Figure 2.15). In some regions the landscape-scale pace of erosion is correlated with hillslope steepness (or local relief; Ahnert, 1970), whereas in others it is correlated with changes in river profile steepness (Wobus et al., 2003). Like these examples, many of the key controls on landscape evolution appear to have context-dependent aspects that present challenges—and opportunities—for developing integrated global understanding of the controls on landscape dynamics. Greater understanding is needed not only to identify fundamental controls on, and theory for, landscape evolution but also to understand how different circumstances and settings influence the driving forces or dominant factor(s) and how systems interact in different regional contexts. Only then can the range and limits to the applicability of theories and the strength and consequences of interactions among processes be known.
To date, however, overarching theory has proven useful, and substantial progress has been achieved from studies of steady-state orogens. For example, recognition of the role of enhanced windward erosion and limited erosion on the leeward, rain-shadowed side of mountain ranges (e.g., Reiners et al., 2003) has confirmed predictions of modeling studies (e.g., Koons, 1990; Willett et al., 1993; Willett, 1999) and bolstered evidence of rock uplift and deformation patterns that matched the conceptual framework (e.g., Beaumont et al., 1996; Batt and Braun, 1999). Connections between climate, erosion, and the tectonically driven growth of orogenic wedges have been explored in coupled models (e.g., Whipple and Meade, 2004, 2006; Tomkin and Roe, 2007). Coupling of erosion, tectonic deformation, and patterns of rock uplift have also been explored at finer scales through the development of individual fold belts or geological structures (e.g., Wobus et al., 2003; Hilley et al., 2004; Simpson, 2004; Stolar et al., 2007). While there has been tremendous progress on such linkages, significant uncertainties and questions remain about the role of erosional processes on the dynamic development of geological structures in diverse tectonic settings.
Further elaboration and evaluation of such linkages
and the implications for landscape response to tectonic and climatic perturbation offer tremendous research opportunities. In particular, key research opportunities include:
• The role of climate and tectonics in surface processes and landscape evolution;
• Feedbacks and linkages between climate and surface processes with mountain building and decay, shoreline advance and retreat; and
• Linkages among cliamate, surface processes of erosion, transport, and sedimentation, and deep Earth lithospheric processes.
These linked research areas offer exciting new opportunities to broaden our understanding of the fundamental controls on Earth surface processes and their influences on the world’s landscapes.
Role of Climate and Tectonics in Surface Processes and Landscape Evolution
While one could hardly imagine a more striking contrast than that between the slow evolution of hard, dense tectonic plates and the fluid, rapidly changing atmosphere, the connections between the climate and tectonic systems are far deeper and more subtle than commonly imagined (NRC, 2010a). Climate, tectonics, and erosion interact over timescales ranging from individual storm events or earthquakes to millions of years over the course of the evolution of a mountain range. The importance of climate and climate variability is central to understanding both the geomorphological impacts of shallow crustal processes over short timescales and how such processes integrate up over longer timescales to influence landscape evolution.
A quantitative, process-based understanding of the linkages among climate, hydrology, geomorphological processes, ecosystems, and landscape evolution is a primary goal of research on Earth surface processes. Fundamental to achieving this goal is the development of transport laws that mathematically characterize the controls on rates of processes shaping Earth’s surface. While significant progress has been made in developing transport laws for a variety of processes (Dietrich et al., 2003), transport laws are still lacking for processes as fundamental as landslides, glacial erosion, and chemical erosion. In addition, the fundamental controls on one of the basic components of the rock cycle, the breakdown of rock into erodible debris, is poorly understood. The formulation of process laws allows quantification of the driving phenomena and thereby rigorous exploration of questions of sensitivity of landscape response to climate change and numerous feedbacks between climatic, hydrological, geological, and geomorphological processes.
The linkages among surface processes and climate with tectonics also have societal implications on human timescales in the role that sedimentation and erosion play in the distribution and rates of displacement of active faults. Landforms and sedimentary deposits preserve records of past earthquakes and deformation that are used to evaluate recurrence intervals for active faults and assessment of seismic hazard. Recent paleo-seismological observations of migration of deformation between fault strands over thousands of years (Dolan et al., 2007) challenge traditional views of steady fault slip due to far-field plate motions, with important implications for seismic hazards, earthquake clustering, fault growth, and fault interactions.
Feedbacks and Linkages Between Climate and Surface Processes with Mountain Building and Decay
The rugged topography of mountain environments reflects the interplay of spatially variable tectonic uplift and erosion. The consequences of rapid erosion in response to snowmelt, intense rainfall, or glacial dambreak floods are familiar to those living in mountain environments. Less widely appreciated is how rates and patterns of deformation in tectonically active mountain belts can be greatly influenced by the spatial distribution and pace of erosion by landslides, river incision, and glaciation (NRC, 2010a). Recent recognition of the strong coupling between erosion and surficial mass redistribution and deeper tectonic and structural deformation creates new opportunities for interdisciplinary research that bridge climate science, geomorphology, structural geology, and geophysics.
Precipitation and erosion induced by orogenic effects impact the distribution of deformation in mountain belts. Conversely, the size and distribution of high-elevation topography influence global,
regional, and local climates (e.g., Meehl, 1992; Wu et al., 2007). While much of the work on climate-erosion linkages in the past decade has focused on steady-state landscapes, new research opportunities in transient responses of landscapes include the buildup and tearing down of mountains, the evolution of rift zones or volcanic arcs, and the role of climate variability (ranging in scale from glacial-interglacial periods to surface response to changes in storm frequency-magnitude relationships). The response of crustal-scale processes and feedback through climate linkages is central to understanding the controls on mountain building and decay and landscape response times to climatic changes and climate variability.
Erosional unloading and sediment loading of Earth’s surface also influences the structural geology and rheology of the lower crust. While coupled tectonic-surface process models predict that the structural evolution of a mountain belt is sensitive to spatial and temporal variability in climate forcing (see Figure 2.16), the common assumptions that erosional efficiency increases linearly with precipitation, discharge, or stream power have not been demonstrated over orogenic timescales. Similarly, the role of lithological variability on long-term patterns of landscape evolution remains poorly constrained.
Research opportunities under this theme also include the influence of the global distribution of topography on climate through, for example, how the location of mountain belts impacts larger-scale climate patterns. On the global scale and over geological timescales, the positions of the continents affect ocean circulation and global climate. A prominent example is the breakup of Antarctica and Australia and opening of the Drake Passage, which led to circumpolar circulation that isolated Antarctica from warmer low-latitude waters and is implicated in the cooling climate of the Cenozoic. Conversely, climate influences the deformation and structural evolution of mountain belts and the margins of continents. For example, Northern Hemisphere glaciation in the late Cenozoic is linked to denudation and migration of deformation in the St. Elias range in Alaska (Berger et al., 2008; Chapman et al., 2008), with implications for mountain belts worldwide. In this research area of climate and orogenesis, empirical studies have lagged models. New observational studies are needed that integrate geomorphology with geochronological and
FIGURE 2.16 Example of unidirectional moisture flux and mountain-belt evolution. (a) Results of numerical model of the Southern Alps of New Zealand with moisture-laden winds arriving from the west (left). (b) The observed topography and pattern of total uplift in the Southern Alps closely match the numerical experiment shown in (a). SOURCE: Whipple (2009), modified by permission from Koons (1990). Courtesy of Macmillan Publishers Ltd.
geochemical studies to constrain timing and rates of uplift and erosion, with seismic imaging of sediment deposits adjoining mountain belts that record past conditions, and with structural geology and geophysical studies that target deeper crustal and mantle structure.
Linkages Between Climate, Surface Processes, and Deeper Earth Processes
Although it has long been recognized that lower crust and mantle processes can significantly influence landscape evolution, linkages between climate and deeper Earth processes remain largely unexplored. Climate and tectonics are fundamentally linked through the influence of sediment loading and erosion unloading on the state of stress in Earth’s interior that in turn govern tectonic motions. For example, the development of large, high-elevation plateaus holds the potential for strong climate-tectonic feedbacks through rapid, localized incision on plateau margins that receive substantial precipitation. Such localized erosion creates the potential to advect hot, low-viscosity, mid-to-lower crustal rocks to the surface in either channel flow along laterally continuous belts or localized domal uplifts (e.g., Beaumont et al., 2001, 2004; Hodges et al., 2001; Zeitler et al., 2001; Koons et al., 2002). Rapid erosion in such settings can lead to a positive feedback by drawing up highly pressurized ductile rock toward the surface, resulting in isothermal decompression that may induce partial melting that further reduces viscosity and resistance to flow. Because deformation rates can respond to surface forcing with little time lag, the pace of surface erosion can drive long-term patterns of structural deformation. The response to climate variability of such tightly coupled erosion-tectonic systems has not been explored and presents an attractive opportunity for future research.
Other examples of deeper Earth response to erosion unloading and sediment loading of Earth’s surface include the impact of sediment distribution on the distribution and magnitudes of subduction zone megathrust earthquakes, with important implications for the major human population centers located along subducting margins (e.g., Wells et al., 2003). Recent studies reveal linkages between climate and volcanic activity with increased volcanic activity during periods of deglaciation (e.g., Sigvaldason et al., 1992; Jellinek et al., 2004) that are attributed to enhanced decompression mantle melting due to glacial unloading (Jull and Mackenzie, 1996; MacLennan et al., 2002). Release of carbon dioxide associated with this enhanced subaerial volcanism during deglaciation may in turn play a significant role in modulating glacial/interglacial cycles (Huybers and Langmuir, 2009).
Global patterns of sea-level rise are directly linked to elastic deformation of the solid Earth and are another manifestation of the complex interactions between Earth’s interior and surface. There is particular concern that accelerated melting in the modern warming world could lead to collapse of the West Antarctic Ice Sheet with meter-scale rises in sea level worldwide. Highly non-uniform sea-level rise is predicted with enhanced sea-level rise around North America as a result of the interplay between changes in gravity due to the redistribution of ice/water and rock, changes in Earth’s rotation, and changes in shoreline geometry (see Figure 2.17).
FIGURE 2.17 Predicted sea-level change in meters following the collapse of the West Antarctic Ice Sheet, based on theory that includes variations in ice and ocean volume, gravity, rotation, and shoreline configurations and deformation of the crust and mantle. SOURCE: From Mitrovica et al. (2009). Reprinted with permission from AAAS.
In all of these research areas, significant opportunities exist for framing testable hypotheses to guide field studies of the interactions between climate and tectonics in landscape evolution. Of particular need are studies to evaluate temporal variability. Given the different timescales of climate variability and deep Earth processes, what are the sensitivities and lag times built into their interactions? Several challenges and science objectives under the theme of climate-landscape-tectonics interactions are primed for significant advances:
• Developing theory for the interactions between climate, topography, land cover, and the deeper Earth interior at global, regional, and local scales.
• Integrating surface processes and deep Earth studies, including petrological and seismological studies, and the record of past surface environments, to explore connections between deep Earth processes and Earth surface dynamics.
• Developing geomorphic transport laws that account for climate and the role of biota to describe and quantify river and glacial incision; landslides; and the production, transport, and deposition of sediment.
• Measuring and modeling landscape evolution under diverse and varying climatic conditions, with an emphasis on identification of physiographic signatures of climate and climate variability, and evaluation of thresholds of landscape response and the limits of landscape resilience.
• Improvement of coupling between surface process and climate models, including incorporation of feedbacks and thresholds.
All of these promising research areas will be facilitated by recent and new developments in thermochronometry, cosmogenic methods for dating geomorphological surfaces, LiDAR, satellite imagery, modeling capabilities, experimental methods, and field instrumentation.
Earth is apparently unique in the Solar System in bearing living organisms that profoundly modify planetary processes affecting the composition and properties of the atmosphere, hydrosphere, and lithosphere. The geological record has provided a compelling narrative of major changes in Earth’s climate, environment, and evolving life, played out over billions of years that has defined our planet’s life-sustaining outer shell. These interactions continue to shape the world in which we live, and our future depends on such interactions as they unfold over the coming centuries—and on our thoughtful and responsible stewardship of them. Yet to understand the future, we need to know our geochemical and geobiological past.
Earth’s environmental systems have experienced geochemical, climatic, and biotic change, with conditions in the distant past remarkably different from those of the Holocene epoch—the epoch when low and relatively stable atmospheric carbon dioxide and largely benign climatic conditions fostered human civilizations. Earth’s deep-time record provides numerous unique analogs to the emerging climate state of dramatically warmer temperatures and highly elevated greenhouse gas contents in the atmosphere. But life’s planetary habitat has undergone even more profound geochemical transformations. For example, the advent of biological oxygen production and the expansion of plants onto land are both changes that reorganized element fluxes and concentrations in the ocean, sediments, and atmosphere on a global scale. Only the deep-time geological and paleontological record can provide examples of change that rival the scale of contemporary human-induced impacts on land, biota, oceans, and climate. Thus, understanding past biosphere-geosphere behavior is a potent approach to anticipating how linked physical, chemical, and biological processes that characterize Earth’s surface may be impacted by and respond to human activity. Earth’s biogeochemical history archived in the deep-time geological record thus provides a major research opportunity to investigate the future of our planet.
Understanding recent and ongoing climate change requires a full exploration of the range of climate phenomena, rates, feedbacks, thresholds, and tipping points captured over the long “experiment” of Earth history. Studies of the deep-time record have revealed that Earth’s climate varies between two extremes. At one extreme is a cool, glaciated icehouse state associated with low greenhouse gas concentrations in the atmosphere and the state in which humans evolved, while at
the other extreme is a warm greenhouse mode apparently associated with higher atmospheric greenhouse gas levels and small-to-no ice sheets (see Box 2.6). The geological archive has been particularly important for revealing how many physical, chemical, and biological processes operated differently or were unique to past warmer and transitional states than during the present cool state (NRC, 2011a).
Our ability to characterize and interpret the deep record has increased dramatically over the past decade and continues at an accelerating pace. New tracers (proxies) of past conditions have greatly refined our ability to extract ancient records of Earth surface conditions, including temperature, atmospheric levels of carbon dioxide, the chemical composition of and oxygen availability in the ocean, regional hydroclimate, and the interrelationship and physiologies of ancient life forms. These proxy records can now be placed in an ever more refined age context stemming from successfully coordinated efforts in the geochronological community (e.g., EARTHTIME) aimed at better, higher-resolution use of traditional methods; new and emerging techniques to accurately date nontraditional materials; and extension of orbitally tuned kilo-year-scale chronometers to the deep past. No longer is poor age control the bane of studies aimed at the past. Also, these diverse data can now be brought together into the interpretative framework of small-to large-scale numerical approaches ranging from geochemical box models to global climate models such as general circulation models (GCM; NRC, 2011a).
These advancements allow development and testing of process-based hypotheses, which in turn are leading to major improvements in our understanding of the interplay of climate and life in molding and modulating one another. For example, mining of the geological record over the past several decades has documented feedbacks in the global climate system that appear unique to warmer conditions (e.g., the mid-Cretaceous and early Eocene; Zachos et al., 2001, 2008; Kiehl, 2011). Such mining has simultaneously revealed repeated periods of abrupt climate change that have, at times, led to accelerated warming, major change in regional hydroclimates, and major ecological disruption (e.g., the Paleocene-Eocene Thermal Maximum, or PETM; see Kennett and Stott, 1991; Zachos et al., 2001; Wing et al., 2005; Woodburne et al., 2009). A new coupling between highly resolved phylogeny reconstructions and the geochemical record of environmental change (see Box 2.7) is dramatically changing our understanding of the mechanisms behind Earth’s largest biogeochemical transitions. Despite such advances, understanding Earth’s spectrum of climate phenomena and the associated history of life at the temporal and spatial scales appropriate for testing specific hypotheses of mechanistic linkages and causation remains a significant challenge for nearly every major trend and event. The following discussion presents a set of deep-time research opportunities that, during the coming decade, are likely to lead to major advances in our understanding of variability in the geosphere and its interwoven interaction with the biota.
How Have Dynamics of the Global Climate System Varied in the Past?
Contemporary climate change can be better understood through exploration of the range of climate states, rates, feedbacks, and tipping points captured over Earth’s history. The current glacial state provides an important baseline against which future climate change can be assessed. Understanding a world characterized by ice sheets at both poles and atmospheric carbon dioxide partial pressure (pCO2) up to 30 percent less than present-day levels, however, captures only a small part of known climate variability. At current rates of concentration, by the year 2100 greenhouse gas concentrations will approach atmospheric values inferred for the greenhouse climates of the Paleogene (Kiehl, 2011; NRC, 2011a). Critical insights into how Earth’s systems have functioned in such a high CO2 environment are archived in the records of past warm periods and major climate transitions. For example, deep-time studies reveal past periods of anomalous tropical and polar warmth that were associated with major changes in ocean and atmospheric circulation, including at times marine anoxia and acidification, and intensification of the hydrological cycle that included both increased rainfall in some areas and increased drought in others (e.g., Wilson and Norris, 2001; Pagani et al., 2006). Consequences for marine and terrestrial ecosystems were dramatic. Intervals of abrupt climate change documented by the deep-time geological record—most notably, past hyperthermals—reveal how changes in
CO2-Climate Linkages Through Earth History
Warmer greenhouse conditions that have dominated Earth history have been typically associated with CO2 levels in the atmosphere elevated over those of present-day carbon dioxide partial pressure (pCO2; 392 ppmv) and those of cooler icehouse periods (see Figure B2.6, top). The widespread continental ice sheets of icehouse times have been rare during warm periods, with the exception of transient glaciations (e.g., Ordovician [~440 Ma]). The climate linkage between radiative forcing, Earth surface temperatures, and high-latitude continental glaciation is clearly delineated in the history of the buildup of the Antarctic and Northern Hemisphere ice sheets (see Figure B2.6, bottom). For example, the buildup of the East Antarctic Ice Sheet was initiated by the coupled effects of long-term decrease in atmospheric pCO2 across a climate threshold and orbital climatic preconditioning (Pälike et al., 2006). During the Early Pliocene warming (3.5 to 3.2 Ma)—a time associated with CO2 levels that may have been comparable to current levels (Pagani et al., 2010)—sea level was 15 to 25 m and possibly 36 m higher than at present day (Wardlaw and Quinn, 1991; Shackleton et al., 1995; Naish et al., 2009).
Such deep-time records also reveal how long-term (millennial timescale) and short-term (operating on a subcentury timescale) feedbacks have interacted to influence climate and sea-level dynamics under rising levels of atmospheric CO2 and other greenhouse gases and provide insight into the duration over which elevated greenhouse gas levels have persisted in the atmosphere—both issues of direct societal relevance. For example, studies of long-term equilibrium sensitivity of surface temperatures to rising atmospheric CO2 levels indicate temperature has been enhanced during times of higher atmospheric CO2 due to the switching on of long-term positive feedbacks (Royer et al., 2007; Pagani et al., 2010). Feedbacks such as changes in ice sheet volume, distribution and composition of terrestrial biomes, and greenhouse gas release from soils, tundra, and ocean sediments typically operate on timescales much longer than that of humans and are projected to become increasingly more relevant on human timescales (decades) with continued global warming (Hansen and Sato, 2001; Hansen et al., 2008).
FIGURE B2.6 Top: Atmospheric pCO2 and continental glaciation over the past 800 million years. Vertical white and gray bars indicate the timing and extent of continental ice sheets (after Crowley, 1998; Evans, 2000). CO2 trends are inferred from mineral and biological proxies. Plausible ranges of CO2 estimated using the GEOCARB III model are also plotted (Berner and Kothavala, 2001). All data have been adjusted to the Gradstein et al. (2004) timescale. Bottom: Global compilation of deep-sea benthic foraminifera 18O isotope records from 40 Deep Sea Drilling Program (DSDP) and Ocean Drilling Program (ODP) sites (Zachos et al., 2001) updated with high-resolution records for the Eocene through Miocene intervals (Billups et al., 2002; Bohaty and Zachos, 2003; Lear et al., 2004). Much of the post-Oligocene 18O variability (~70 percent) reflects changes in Antarctic and Northern Hemisphere ice volume, which is represented by white and gray horizontal bars (e.g., Hambrey et al., 1991; Wise et al., 1991; Ehrmann and Mackensen, 1992). The dashed bars represent periods of ephemeral ice or ice sheets smaller than present, whereas the solid bars represent ice sheets of modern or greater size. The evolution and stability of the West Antarctic Ice Sheet (e.g., Lemasurier and Rocchi, 2005) remain uncertain and could affect estimates of future sea-level rise. SOURCE: Caption adapted from Jansen et al. (2007). Diagram courtesy of Linda Sohl and Mark Chandler.
greenhouse gas concentrations can abruptly and profoundly influence climate and life (McElwain et al., 2005; Schaller et al., 2011).
Deep-time geological records and the genomes of living organisms are also rich archives of Earth’s deep-time history. Mineral and biological environmental indicators (proxies) record the interaction, feedbacks, and responses of physical, chemical, and biological processes under the full range that the Earth system has experienced (see Figure 2.18). A major challenge is to develop reliable proxy records of atmospheric gases, surface temperatures, precipitation, relative humidity, and marine and terrestrial productivity at a variety of temporal scales from millions to thousands of years to address the multiple scales at which the processes act. Opportunities for research exist in the development and calibration of new and existing proxies, the construction of precise and accurate long-and short-term proxy records at the requisite spatial and temporal resolution dictated by the hypotheses being tested, including next-generation paleoclimate-data and model-model comparisons.
How Have Climate, Life, and Biogeochemical Cycling Interacted Through Time?
The deep-time geological record documents the magnitude over which the physical, chemical, and
biological attributes of the ocean, continents, and atmosphere have varied over the history of Earth. There is little debate that microbial life and plant life have played a fundamental role in the evolving atmospheric concentrations of O2 and CO2, but the specifics of this interplay remain highly controversial. Tectonically driven changes in degassing and continental weathering are also fundamental, especially with respect to geologically transient but biologically devastating greenhouse gas increases such as during mass extinction and ecosystem reorganization events, notably those at the end-Permian, end-Triassic, Cretaceous-Paleogene, and PETM. However, it remains to be understood how new life forms changed the nature of elemental cycling (O, C, N, S) or how long-term changes in geochemical cycling have influenced the evolution of new life forms. Similarly, the oceans have fluctuated from periods of minimal oxygenation to conditions comparable to the well-ventilated ocean bodies of today. In addition to gradual, long-term shifts in baseline conditions, the oceans have at times experienced rapid perturbations that have led to transient states in ocean chemistry and circulation. These in turn have contributed to major climate change, ocean acidification and hypoxia, and consequent large-scale biotic impact.
In the tropics, integrated paleoclimate and paleoecology studies can address the fundamental question of how hot the tropics will become, and to what extent
Molecular Geobiology Data Revolution
Armed with modern capabilities in macromolecular sequencing, the structures and processes of entire microbial communities can now be characterized. New advances allow determination of tens of billions of bases per run, and this scale of capacity is jump-starting the field of environmental genomics. Genomic data derived from environmental RNA reveal microbial dynamics on scales of minutes, while data derived from DNA allow characterization of geobiological evolution over billions of years. The field is poised to address challenges facing humanity, including increasing soil fertility to aid in feeding the world’s growing population, providing novel approaches to managing Earth’s resources and waste disposal and attenuating the impacts from human land use and climate change in the critical zone.
This emerging revolution offers unprecedented insight into the microbial communities that mediate Earth’s elemental cycles. With our growing ability to identify the biological diversity of microbes irrespective of whether they can be cultivated, it can now be identified where these microbes are located in relation to each other and to Earth materials, and their activity and geochemical roles can be tracked over space and time. Never before has it been possible to obtain such information without having microbes in culture, and never before have so many data been collected. But this is just the tip of an immense “iceberg” in a data revolution that is beginning to show its full weight, as the “meta-omics” world (meta-genomics, proteomics, transcriptomics) becomes readily accessible. The availability of inexpensive sequencing has moved studies at the interface between geochemistry and molecular biology to a new level. Nearly limitless amounts of molecular (sequence) data can now be collected, allowing the genetic complement of nearly any environment to be seen nearly instantaneously. Billions of base pairs can be “harvested” and analyzed to provide a DNA snapshot of the biodiversity and gene diversity of an environment, while monitoring of RNA and protein expression provides new avenues for probing geobiological dynamics in near real time. From this perspective come unprecedented baselines and records of change in the face of recent environmental perturbation.
Accompanying these extraordinary opportunities is the reality that we still have a long way to go to realize the promises that new “omics” approaches hold for transforming the fields of geobiology and geochemistry. The explosion in sequencing has unveiled staggering genetic diversity, but these new vistas are matched by a widening gap between gene sequencing data and our understanding of the data’s biochemical, ecological, and geochemical function. Much critical and fundamental work is needed, including (1) annotating and identifying new genes, (2) sorting out the implications of genetic diversity within microbial taxonomic units, and (3) filling the dearth of reference strains and genomes needed to test hypotheses generated via genomic and metagenomic approaches.
The sheer volume of data available at relatively low cost increasingly pushes analytical challenges into the realm of computer science—one of the major challenges of the next few years. Added to these computational challenges will be interfacing the omics data with geochemical/geological data—two data sets that are fundamentally different in terms of definition and quantification. Bringing the two fields together will ultimately allow each to make predictions about the other: omics approaches open entirely new avenues for probing geochemistry, while the geochemical community (organic and inorganic) can provide a rich context in which to understand molecular geomicrobiology. Integrating these communities has vast potential for transformative cross-disciplinary breakthroughs, including new advances at very fine temporal scales.
Among the emerging research questions and opportunities empowered by new computational, nanoscale, and DNA-based approaches are the following:
• What regulates cellular and subcellular agents in complex environmental systems?
• How does biodlversity relate to ecosystem function, and resiliesce, and how does it respond to environmental perturbation and specifically climate change?
• What can the genetic record tell us about the history of life and its planetary habilat?
• How can we integrate genomics and the geological record to probe the emergence of metabolic processes and their impacts on the evolving geochemical states of Earth?
ocean chemistry will be perturbed, as atmospheric CO2 continues to rise (NRC, 2011a). Such changes may have dire effects on tropical ecosystems, with the potential for severe declines in diversity over large areas. The penultimate deglaciation of the Late Paleozoic Ice Age is the only archival record of the tropical floral response to climate change associated with the end of a glacial epoch. How Arctic ecosystems will respond if sea ice disappears permanently—or if the Greenland ice sheet retreats significantly—can be examined through the lens of past warm periods, such as the mid-to-late Cretaceous and the early Cenozoic, when the Arctic was ice-free and supported lush rainforests, warm swamps with aquatic floating plants, and warm-water fauna.
The forcings that led to past oceanic perturbation, the rates of change and recovery, the importance of thresholds, and the connections between oceanic
FIGURE 2.18 Cenozoic pCO2 for the period 0 to 65 million years ago. Data are a compilation of marine (C isotopic composition of alkenone biomarkers and boron isotopic compositions of foraminifera) and lacustrine mineralogical records. The dashed horizontal line represents the maximum pCO2 for the Neogene (Miocene to present) and the minimum pCO2 for the early Eocene, as constrained by calculations of equilibrium with Na-CO3 mineral phases (vertical bars, where the length of the bars indicates the range of pCO2 over which the mineral phases are stable) that are found in Neogene and early Eocene lacustrine deposits (Lowenstein and Demicco, 2006). SOURCE: Zachos et al. (2008).
change and biological crises all require further investigation to be properly understood. Greatly improved dating, refinement and further development and calibration of proxy records of regional and global climate, and appropriately resolved databases will permit researchers to reconstruct past changes in Earth’s surface environments, including the atmosphere, oceans, and soil systems, as well as greenhouse gas burdens (see Box 2.8). These reconstructions will permit characterization of past climates and will give insights into anthropogenic impacts. Furthermore, opportunities for new research arise from new techniques, allowing the interaction between organisms and the environment to be examined directly in living forms through molecular means and in deep time by integration of phylogenies with proxy records of environmental and climate change. This involves assessment of the origin of clades of organisms (both by phylogenetic and phylogenomic methods) and delineation of the nature of environmental feedbacks that may allow elucidation of the cause-and-effect conundrum of biotic evolution and major climate change. These connections resonate with anthropogenic effects in which the biological innovations that make humans what they are have clearly resulted in changes in carbon dioxide and other greenhouse gas concentrations, climate, and ecosystem function.
What Are the Trends and Milestones in the Interaction and Co-evolution of Life and the Environment?
The deep time record has revealed events and trends of enormous magnitude and import well outside the scale of human experience. Some of these events have been the subject of long-standing inquiry, such as the origin of life, and others are relatively newly discovered, such as the bolide impact at the Cretaceous-Paleogene boundary. Opportunities exist for new research at the interface between mechanistic studies of biological processes such as proteomics, the discovery of new types of and spectacularly preserved fossils, and the application of highly accurate dating techniques linking disparate environments and processes.
A mechanistic understanding of the origin of life remains a vexing challenge and one of the great opportunities of this century. Recently, the exploration of extreme modern environments, such as hydrothermal vents, coupled with metagenomics (e.g., Grzymskia et al., 2008), phylogenomics (Delsuc et al., 2005),
Proxies for Reconstructing Past Climates
Reconstructing past climates rests on our ability to indirectly infer temperature, precipitation, atmospheric greenhouse gas concentrations, and other environmental properties from sedimentary materials. The best-known proxy is δ18O of biogenic CaCO3 (in marine microfossils and animals), which has long been shown to reflect the combined effects of local temperature and global ice volume on seawater δ18O. More recently, oxygen isotope analysis of biogenic hydroxyapatite in marine and terrestrial fossils has been utilized as a proxy of seawater δ18O and of continental mean annual temperatures, respectively (e.g., Fricke and Wing, 2004; Buggisch et al., 2008; Trotter et al., 2008). During the past decade, a variety of new proxies have been developed that have led to a major improvement in our ability to reconstruct past climates (summarized in Understanding Earth’s Deep Past [NRC, 2011a]).
Despite the maturity of the stable isotope field, fundamentally new advances continue to be made—for example, by assessing the distribution or “clumping” of rare isotopes in minerals. Traditionally, the isotopic composition of a compound is determined by destroying the original structure of that compound and measuring the relative isotopic abundances of the bulk material. For example, δ13C and δ18O of calcite document the 13C/12C and 18O/16O ratios in the sample, retaining no record of how those isotopes were distributed. Recent advances that allow access to this distribution have ushered in a new and rich source of information contained in the stable isotopes. Most notably, Ghosh et al. (2006) showed that there is a temperature-dependent thermodynamic preference for heavy isotopes in calcite to share a bond—the lower the temperature, the stronger the preference for 13C-18O bonds compared to a completely random distribution. This discovery forms the basis of a completely new type of calcite paleothermometer. In particular, a measurement of the abundance of the13C δ18O16O variant of CO2 evolved from calcite relative to the random distribution of isotopes, referred to as Δ47, can provide formation temperatures to a precision of ±2°C. Importantly, unlike classical δ18O calcite thermometry, this “clumped isotope thermometer” is independent of assumptions about the composition of water from which the calcite precipitated.
Recent work demonstrates that clumped isotopes accurately record paleotemperatures in a wide variety of marine biogenic carbonates (Came et al., 2007; Tripati et al., 2010), cave and soil carbonates (Affek et al., 2008; Passey et al., 2010), and carbonate-fluorapatite in vertebrate bones (Eagle et al., 2010). Ongoing work (e.g., Passey et al., 2011) reveals an apparent sensitivity of clumped isotopes in low-temperature precipitates to diagenesis requiring further calibration and assessment studies.
FIGURE B2.8 Magnitude and duration of Late Ordovician–Early Silurian glaciation based on carbonate “clumped” isotope paleothermometry (modified from Finnegan et al., 2011). (A) Hypotheses regarding the duration of the icehouse interval: restricted largely or entirely to the Hirnantian stage lasting as few as 500,000 years, with a peak in the Hirnantian interval. Both the beginning and the end of the Hirnantian stage saw a decrease in marine invertebrate genus diversity. (B) Δ47-derived near-surface ocean temperature trend for the early Katian to late Aeronian interval. (C) δ18O (VPDB) trend over the same interval. (D) Relative contributions of temperature and δ18Owater to changes in d18O (Δδ18O) between successive time intervals. Bars are scaled to the magnitude of d18O, and color proportion is scaled to the relative contribution of temperature change (red) and change in the oxygen isotopic composition of seawater (blue) to Δδ18O. (E) δ18Owater (VSMOW) trend. Dotted lines indicate δ18Owater value during the Pleistocene LGM (10) and expected δ18Owater value for an ice-free world. Various symbols and colors indicate various fossil organisms and locations. SOURCE: Finnegan et al. (2011).
and proteomics (Gaucher et al., 2003), geochemical proxies (biomarkers) of various microbial groups, and analysis of the isotopic proxies of past environmental conditions has resulted not only in a better chronology of the major biotically mediated transformations of Earth but also provided a chronology of the evolutionary and physiological steps in the evolution of early life. Two surprising results from this work are (1) that our last universal common ancestor (LUCA) was plausibly a thermophile but not a hyperthermophile (Gaucher et al., 2008; Gouy and Chaussidon, 2008) in hydrothermal vents (Martin and Russell, 2007) and (2) while photosynthesis evolved very early, the early photosynthetic organisms did not produce oxygen (i.e., were anoxygenic).
Geochemists have made great progress in using the elemental and isotopic properties of ancient sediments to reconstruct the evolving redox state of the ocean and atmosphere. A decade ago Farquhar et al. (2000) established anomalous mass-independent fractionation of sulfur isotopes as the smoking gun for the near absence of O2 in the atmosphere before the Great Oxidation Event 2.4 billion years ago. Now, frontiers for sulfur isotope approaches lie with recognition of
specific microbial metabolisms in the very old record and their environmental implications (Johnston et al., 2008). Iron geochemistry calibrated in modern settings has become our most reliable inorganic fingerprint of local oxygen deficiency in the ancient ocean (Poulton and Canfield, 2005), while organic biomarkers further trace the co-evolution of life and the environment (Brocks et al., 2005). Other redox-sensitive elements, such as molybdenum, can provide a global picture of ocean oxygenation when viewed for their mass balance relationships (Scott et al., 2008) and even delineate times when biologically critical trace metals may have limited the evolutionary advance of life. At the same time, metal isotope systems, such as iron and molybdenum, are providing global perspectives on past ocean-atmosphere oxygen conditions as a backdrop to the early evolution of life (Johnson et al., 2008).
Complementary geochemical and genomic studies are informing our understanding of other major biogeochemically important paleobiological milestones. While there are too many to detail here, these milestones include the origin of animals in the late Proterozoic, the spread of grasslands and co-evolved grazers during the Neogene (Cerling, 1992; Bouchenak-
Khelladi et al., 2009), and our own evolution (Feakins et al., 2005; Steiper and Young, 2006; NRC, 2010b). These examples are associated with major shifts in climate mode and variability and elemental cycling. The major challenge and opportunity is linking the evolutionary events with the environmental causes or consequences via tests of mechanistic hypotheses, and the tools to do this now exist.
What Are the Patterns and Drivers in Extinction and Recovery?
The record of life on Earth is punctuated by a series of mass extinctions, including the so-called Big 5—Ordovician-Silurian, Late Devonian, end-Permian, end-Triassic, and Cretaceous-Paleogene (often called K-T)—as well as major biotic reorganization events such as the PETM. The committee’s view of these extinctions and recovery has in the past been strongly dominated by records of taxonomic change (e.g., Raup and Sepkoski, 1984). But this is changing, with greater emphasis on other kinds of diversity that may have at least as much impact on function in ecosystems and the biosphere as a whole. Important approaches include analysis of morphological and physiological disparity, biotic provinciality, and the role of biodiversity in functional (and ecological) redundancy and ecosystem stability.
The deep-time record of past biotic turnovers and mass extinction events associated with warm periods (many associated with massive outgassing of carbon dioxide or methane), transient warmings, and major transitions between climate states offer an under-tapped repository from which unique insight can be obtained regarding patterns of ecosystem stress, the potential for ecological collapse, and mechanisms of ecosystem recovery (NRC, 2011a). Such periods of crisis naturally resonate with today’s global warming and biotic crises. For example, the warm, low-pH, and low-oxygen ocean that will come with global warming was first experienced in the Phanerozoic and Proterozoic. It was linked to profound global climatic and biological instability. At least some Phanerozoic mass extinctions appear to be associated with a doubling to tripling of carbon dioxide concentrations that occurred over human timescales. Examples include those at the end-Triassic (McElwain et al., 1999; Schaller et al., 2011) or Cretaceous-Paleogene (Beerling et al., 2003) that were caused by massive volcanic eruptions or bolide impacts. Hot, rapidly weathering soils in the coming century have loose analogs in deglacial Permian paleosols (throughout the Pangaean paleotropics) and in the postglacial phase of Proterozoic glaciation (i.e., in the wake of the hypothesized “snowball Earth”).
The recovery from mass extinction is more than just recovery of taxonomic diversity. The dynamics of recovery include coupled biological and geochemical feedbacks. They also include evolutionary responses such as rapid bursts of speciation in surviving clades, followed by increasing morphological disparity and biotic provinciality. The patterns of these different diversity changes are not well documented. However, they clearly have relevance to the present-day human-caused biodiversity crisis that appears to be causing a mass extinction that may be comparable in magnitude to the Big 5 of the Phaneozoic and perhaps larger in effect than the PETM.
How Has the Global Climate System Operated under States Different from Today?
Studies of our current glacial state provide an important baseline against which future climate change can be assessed. This understanding of a world characterized by ice sheets at both poles and atmospheric pCO2 minimally 25 percent less than present-day levels, however, captures only a fraction of the known range of climate phenomena. Under the current rate of carbon emissions to the atmosphere, greenhouse gas contents and associated radiative forcing will, by the end of this century, reach levels that fall within the probable range of the last greenhouse period of the Paleogene and Cretaceous (see Box 2.6). Critical insights into how Earth’s systems have functioned in such a high carbon dioxide environment are archived in the records of past warm periods and major climate transitions. For example, deep-time studies reveal past periods of anomalous tropical and polar warmth that were associated with major changes in ocean and atmospheric circulation, including at times marine anoxia and acidification, intensified hydrological cycling and regional drought, and consequent substantial impact on marine and terrestrial ecosystems. For many of these periods, the lack of thermostatic regu-
lation reflects the absence of those negative feedbacks that have stabilized surface temperatures during the current icehouse climate system. These reconstructions further reveal how certain processes and positive feedbacks that typically operate on longer timescales—or not at all in glacial climates—can be accelerated under warmer conditions. Furthermore, intervals of abrupt climate change documented by the deep-time geological record—most notably, past hyperthermals of the early Cenozoic and the last greenhouse-icehouse transition of the Late Paleozoic—reveal the nonlinear dynamics associated with pushing the climate system through critical thresholds.
How Does the Study of Interaction and Co-evolution of Life, Environment, and Climate Benefit Society in General?
According to a 2009 Gallup Poll, only 39 percent of the American public view evolution as the most reasonable explanation for the pattern of life on Earth, and there is a strong positive correspondence between acceptance of evolution and level of education (Newport, 2009). The United States ranks 33rd out of 34 developed countries in acceptance that species, including humans, evolved. According to the Pew Research Center (Kohut et al., 2009), only 57 percent of Americans accept the scientific evidence for atmospheric warming, down from 77 percent only 2 years earlier, and only 36 percent attribute global warming to the actions of humans. Many attribute contemporary change to natural cycles, such as sunspot activity, without any knowledge of the natural drivers, rates, patterns, possibilities, or consequences illuminated robustly by the short- and long-term records of Earth history.
The increasingly robust record of the co-evolution of life and the environment can be used to educate scientific and general populations about where Earth has been and where it might be heading. It is fair to conclude that NSF-EAR shoulders the responsibility of being the custodian of Earth history studies and the bridge to its future relevance. From a philosophical perspective, our understanding of geosphere-biosphere interactions in the past shapes our basic curiosity of where humans come from and our perception of human’s role in the world.
The ways in which ecosystems and landscapes have co-evolved through time and the nature of their coupled responses to human activity and climate change present tremendous new opportunities for advancing our understanding of Earth surface processes as well as providing critical scientific input to managers tasked with finding solutions to problems associated with environmental change. This research opportunity differs from the later section on biogeochemical cycles in that its roots are more in geomorphology and materials cycling than geochemistry.
Coupled Landscape and Ecosystem Dynamics
Recognition of the magnitude of influence that hydro-geomorphological processes exert on ecological systems and their influence on landscape processes and dynamics has opened up exciting new areas in the emerging fields of ecohydrology, ecogeomorphology, and geobiology. It is now widely documented that living systems influence the style and pace of surface processes and biogeochemical cycling and that disturbance regimes influence ecosystem trajectories and dynamics. The full scope and breadth of these linkages, however, are only beginning to be understood, in part because of the bi-directional nature of such feedbacks.
Over relatively short timescales, understanding the response of landscapes and ecosystems to disturbance requires explicit consideration of their interactions. Landslides, overgrazing, and flooding are just a few examples of disturbances in which geomorphic, hydrological, and ecological processes are inextricably coupled. Consider, for example, flooding. Vegetation on hill slopes and stream banks plays an important role in regulating the delivery of water and sediment to stream channels at the same time that overbank transport of water and sediment regulates the soil and nutrient conditions for vegetation in riparian zones and floodplains.
While natural disturbances have always been an important driver of landscape and ecosystem co-evolution, humans have, in many cases, altered the frequency, intensity, and impact of disturbances. Returning to the example of flooding, through activities such as defores-
tation, agriculture, installation of dams and levees, and increasing nutrient and contaminant loads in runoff and streamflow, humans have modified stream and floodplain morphology, hydrology, and ecology, often in ways never anticipated and often with the effect of exacerbating the magnitude, frequency, and damage associated with floods. In a time when humans are rapidly becoming the dominant change agent, human-environmental interactions can no longer be ignored in the quest for a unified model of the Earth surface system.
A similar coupling of landscapes and ecosystems is evident on the longer timescales of climate change, particularly in rapidly changing, marginal environments like wetlands, permafrost, and desert margins. Salt marshes, for example, can become unstable when they are flooded too frequently, a potential consequence of sea-level rise. The existence of salt marshes is dependent on an adequate sediment supply and the presence of intertidal vegetation, such as Spartina alterniflora on western Atlantic coasts. Vegetation slows water flow, promotes sediment deposition, and inhibits erosion. Sediment deposition, along with organic matter accumulation, supplies nutrients and maintains the marsh platform at elevations beneficial for primary biological production. These feedbacks result in rates of vertical marsh accretion close to rates of contemporary sea-level rise, provided a sufficient supply of sediment and undisturbed vegetation. The likely response of marshes to accelerated sea-level rise is a complex eco-hydro-geomorphological question currently receiving considerable attention.
In the context of a changing climate, it is particularly important to understand why some regions of Earth’s surface are relatively resilient to change, whereas others are not. It is reasonable to assume that long-term trends of warming temperatures will result in fundamental alterations to polar, glacial, and periglacial landscapes and ecosystems, but at what point are these changes irreversible? More frequent climate extremes are also among the expected manifestations of climate change. Drought, for example, poses severe challenges with regard to food and water resources as well as soil erosion. Yet there are regions of Earth that are able to support annual and perennial plant growth despite low water availability. In these and other landscapes, understanding the factors and processes governing landscape resilience, and in particular the nature of feedbacks and thresholds in system response that may fundamentally alter landscape and ecosystem characteristics, processes, and dynamics are essential for forecasting and interpreting landscape change. Research opportunities for such issues are found in the records of past environmental and landscape change, in studies of contemporary processes, and in model simulations of future scenarios.
Research at the intersections of geomorphology, hydrology, and ecology is providing new insight into the mechanisms of landscape-ecosystem interactions and co-evolution. For example, Roering et al. (2010) have brought an ecogeomorphic perspective to questions related to rates of soil formation in forested landscapes. Soil covers can only be maintained if rates of soil production equal or exceed rates of soil erosion. Roering et al. found that large volumes of bedrock were incorporated into the roots of large coniferous trees (>0.5 m diameter) overturned during storms in the Oregon Coast Range. They suggest that the penetration of deep root systems into bedrock is important in initiating soil formation processes (see Figure 2.19), which in turn helps maintain the mineral-rich soils that support coniferous forest ecosystems in temperate, active tectonic settings like the Pacific Northwest. In drier climates with sparse vegetation, Owen et al. (2011) have shown that bedrock erosion becomes more sensitive to precipitation.
The rapid growth in the field of ecohydrology is providing a theoretical framework and new, testable hypotheses to explain complex ecosystem dynamics and patterns (D’Odorico et al., 2010b). The dominant landscape control on most terrestrial vegetation is soil moisture through its effects on transpiration and photosynthesis. Soil moisture variations are regulated by external factors like topography and soil composition, as well as feedbacks with vegetation, microbial communities, and animal activities, including burrowing and grazing. Landforms and their associated surface-water and groundwater flows also play essential roles in structuring biotic communities. Stream networks, for example, enhance connectivity across the landscape and provide preferential pathways for transport of water, nutrients, sediment, and propagules.
FIGURE 2.19 Profile of (A) filtered and transformed and (B) unprocessed ground-penetrating radar data for a hilltop in the Hadsall Creek catchment in the Oregon Coast Range. The locations of Douglas fir stumps within 1 m of the profile, and their diameters, are shown in (A). (C) Soil depth estimated from the radar data. SOURCE: Roering et al. (2010).
Changes in land use and climate can modify precipitation, runoff, and soil moisture, favoring some species over others, leading to shifts in plant, animal, and microbial composition. Examples include shifts from vegetated to bare soil during periods of extended drought and establishment of water-intolerant species following the drainage of wetland soils. These changes can, in turn, affect water and biogeochemical cycling. For example, draining and drying of wetlands can increase soil respiration and convert wetlands into a source of carbon, fueling further increases in greenhouse gas emissions (Strack and Waddington, 2007). Feedbacks among hydrological and geomorphological processes and biotic communities can allow some species to live in otherwise unfavorable conditions (e.g., water-intolerant plants in wetland environments) or the existence of alternative stable states (e.g., desert and savanna; see Figure 2.20). Improved observations and models of soil moisture variability and its feedbacks with landforms and ecosystems are needed to understand the role of landscape and hydrological change in biodiversity, species invasions, and shifts in plant functional types.
Role of Humans in Landscape Change
Recognition that people are now one of the dominant forces shaping Earth’s surface has opened new areas in the study of recent (i.e., historical) environmental records and in forecasting the effects of future population growth and development on environmental systems and landscapes. There is growing societal recognition that the geomorphological impacts of human land use have shaped ancient societies and continue to do so today, from the role of marsh destruction in exacerbating hurricane impacts on coastal cities to the erosion of the soil in which food is grown.
In many parts of the world, society’s reaction to landscape disturbance is an engineered response: dams and levees to mitigate floods; groins and breakwaters to slow coastal erosion; various forms of hill slope stabilization to limit landslides; and more recently, restoration of rivers and wetlands that have been impaired by human activities. The frequent failure of these interventions to accomplish their goals and/or the unintended consequences of these engineered solutions highlight the critical need for better scientific understanding of the underlying processes and ability to predict the suc-
FIGURE 2.20 Illustration of the effects on soil moisture–vegetation feedback on vegetation patterns in a dryland ecosystem. (a) and (b) Alternate stable states (solid lines) and unstable states (dashed lines) of vegetation biomass; R is annual rainfall in millimeters, and arrows indicate convergence toward a stable state. The thin lines in (b) show stable states under randomly varying rainfall conditions characterized by the indicated coefficients of variation (CV). (c) Noise-induced patterns of vegetation cover (f = fraction covered) for varying precipitation conditions (P is the probability of no water stress). Vegetation patterning occurs at intermediate precipitation conditions when stressed and unstressed states alternate but not for lower or higher values of P. SOURCE: D’Odorico et al. (2010b).
cess and impacts of proposed solutions. The National Center for Earth-Surface Dynamics (NCED), an NSF Science and Technology Center, hosted at the University of Minnesota, is developing leading advances in the science and practice of stream restoration by conducting and coordinating research directed toward multidisciplinary quantitative prediction and development of improved tools to transfer this knowledge into practice (see Figure 2.21). Similarly, the USGS Grand Canyon Monitoring and Research Center and the Glen Canyon Adaptive Management Program have been spearheading high-resolution data collection and stateof-the-art model development to determine if planned water releases from Glen Canyon Dam designed to mimic natural seasonal flooding can be used to improve downstream resources in Grand Canyon National Park.
A key challenge in designing sustainable land uses, from forestry to urban drainage systems, is how to develop regional understanding of landscape history, processes, and change due to both human activity and climate change (past and future). The geography, geomorphology, and ecology of specific landscapes hold the key to understanding human influences on landscapes and therefore are central to correctly diagnosing ecosystem condition and designing effective mitigation, restoration, or adaptation techniques. In this sense the history and effects of land use in different regions could be considered as individual experiments to be probed in the search for deeper, more general, understanding. Similarly, carefully monitored restoration efforts offer case studies that can be used to test and improve quantitative models of landscape evolution. These models, in turn, suggest gaps in our understanding of the underlying processes and critical observations needed to move forward. When observations and modeling go hand in hand, rapid progress can be made in our ability to understand past change and predict future landscape response to restoration activities and other change.
FIGURE 2.21 The Outdoor StreamLab (OSL) facility at the National Center for Earth Surface Dynamics (NCED), located on the banks of the Mississippi River at the University of Minnesota, is dedicated to stream restoration research. OSL uses an abandoned flood bypass channel near St. Anthony Falls to study interactions among river channels, floodplains, and vegetation. Dams and bridge piers can be added to the OSL channel to investigate human-river interactions. SOURCE: Available at http://www.nced.umn.edu/content/outdoor-streamlab-osl. Courtesy of the University of Minnesota.
Such studies are important because the restoration of rivers, wetlands, and deltas is already a major enterprise, and there is a compelling need for Earth scientists to contribute to developing and evaluating methods, strategies, and insights into how to efficiently proceed in many environments.
For these problems and many more, it is crucial to develop mechanistic models of the influence(s) of human actions on landscapes and ecosystems. The NSF-funded Community Surface Dynamics Modeling System (CSDMS) was launched in 2004 to provide the cyber-infrastructure and protocols for coupling and running a suite of numerical models representing diverse processes and scales across Earth’s surface, with the goal of facilitating exploration of surface response to environmental change. CSDMS is moving toward its goal of providing a user-contributed, modular, open-source modeling environment capable of significantly advancing fundamental Earth system science. Investigators are utilizing the CSDMS modeling framework to address proof-of-concept challenges, such as dynamic coupling of fluvial and coastal processes and their evolution over time.
A critical gap in most surface process models is explicit consideration of the role of humans. Some hydrological models have accounted for such influences as the impact of humans on the flux of terrestrial sediment to the global coastal ocean (e.g., Syvitski et al., 2005); however, few have attempted to account for the active role of humans in landscape change. McNamara and Werner (2008) found that interactions of humans and surface processes might best be exhibited at intermediate timescales (years to decades). They constructed a coupled barrier island-resort model to explore emergent instabilities in the landscape induced by human behavior. Resorts and barrier islands are linked through potential resort damage by storm over wash and flooding and the resulting efforts to limit physical and economic damage through site location and size and to maximize revenue by renting many rooms at a relatively high price. Using an agent-based model of human activity coupled with a physically based model for barrier island elevation and evolution, McNamara and Werner concluded that developed barrier islands are lower lying and farther offshore than undeveloped islands, that island vulnerability increases when property is insured, and that protection measures at best postpone widespread damage. This research demonstrates the high social value of coupled mechanistic agent-based models.
Coastal Landscape Response to Sea-Level Rise and Narutal and Anthropogenic Disturbance
Located at the interface between land and sea, coastal systems are particularly sensitive to changes in climate and land use because they are subject to forcings from both ocean and land processes. Climate change effects are pronounced in all coastal regions from the tropics to the poles and include accelerated sea-level rise; ocean acidification; and changes in temperature, precipitation, and storm frequency. Both urbanization and agricultural intensification in coastal watersheds lead to landscape change, including loss of habitat, nutrient buffers, and protective barriers (islands, dunes, wetlands) as well as eutrophication effects, including low-oxygen dead zones, harmful algal blooms, and fisheries’ losses. With most of the world’s major cities and more than 60 percent of the world’s population living near the coast, these changes can be expected to have profound societal and economic consequences globally. Yet our understanding of the impacts of climate and human-induced change on coastal systems is not well developed due in part to the lack of a large, integrated coastal research program.
Coastal environments are strongly influenced by the landscape-ecosystem-human interactions discussed earlier. Close coupling of geomorphic, hydrological, ecological, climatic, and biogeochemical processes shape modern coastal landscapes and dictate their sensitivity and resilience to short-term disturbance events and longer-term trends in climate, land use, and sea level. Changing climate and land use affect coastal systems at multiple spatial and temporal scales. Understanding the effects of these external drivers as well as the interactions and feedbacks among landscape units and processes demands a unifying ecomorphodynamic framework for investigating these complex systems. Studies of specific coastal environments (e.g., barrier islands, marshes, coral reefs, mangroves, seagrasses, estuaries) and their linkages are necessary to understand impacts at regional and global scales.
Coastal systems face accelerated change associated with climate and land use change. At high latitudes, coastal erosion is increasing in response to warming temperatures, sea-level rise, increasing storminess, and decreasing sea-ice extent (e.g., Jones et al., 2009). The thawing of coastal permafrost, with associated decomposition, is likely to result in the release of large amounts of stored carbon to the atmosphere and major ecosystem changes (Schuur et al., 2008). At mid-latitudes there is growing concern about wetlands loss and flood risk with rising sea level, changes in storm magnitude and frequency, and increased temperatures and population pressure. Nicholls et al. (1999) estimated that sea-level rise alone could lead to a loss of almost a quarter of the world’s coastal wetlands by 2080; accounting for added human impacts could increase the losses to 70 percent. However, accounting for feedbacks among inundation, primary production, and accretion of organic and inorganic material on marshes suggests that marsh surface elevations may be able to keep pace with rates of sea-level rise on the low end of future projections if sufficient sediment is available. Marsh erosion rates on the high end of projections are likely to eliminate most existing marshes in this century (Kirwan et al., 2010). At lower latitudes, mangroves and coral reefs offer critical protection from storm-produced erosion to the coastal areas they fringe, and mangroves face many of the same threats as salt marshes, and coral reef systems are even more endangered. Coral reefs are part of the coastal marine ecosystem and are adversely impacted by nutrients, pollution, and sediment from terrestrial runoff (Hoegh-Guldberg et al., 2007). Globally, trapping of sediment in reservoirs and channeling of river flows by levees and other structures has significantly reduced the natural supply of terrestrial sediment to the coastal zone, resulting in sinking deltas and eroding coastlines (Syvitski et al., 2009). Barrier island systems, which make up close to 10 percent of the continental coastlines, are also highly vulnerable to the impacts of climate change and human disturbance (see Figure 2.22).
The history of human modification of coastal environments extends back at least several thousand years (e.g., Stanley and Warne, 1993; Weinstein et al., 2007), including drainage of wetlands, dredging of channels, damming of rivers, mining of sand, and coastal constructions designed to reduce wave energy and shoreline erosion. History has shown that these kinds of modifications tend to increase the vulnerability of coastal environments to catastrophic flooding and storm damage, such as was witnessed during Hurricane Katrina on the Gulf Coast of the United States. Despite
FIGURE 2.22 Aerial photo comparison of developed (right images) and undeveloped (left images) sections of a barrier island response to Hurricane Katrina. While areas on Dauphin Island, Alabama, covered by native vegetation (left) appear to have been less impacted by overwash than developed areas (right) during Hurricane Katrina, Feagin et al. question whether the decrease in erosion and overwash was due to the direct effects of vegetation cover or to the presence of higher coastal dunes that was indirectly built through vegetation interactions with wind-blown sediment transport processes. As noted by Feagin et al., the answer has important management implications. SOURCE: Feagin et al. (2010).
the susceptibility of coastal systems to climate change, human activities are likely to be the dominant impact on coastal systems for the foreseeable future (Weinstein et al., 2007; McNamara and Werner, 2008; Kirwan et al., 2010).
Given the high value of coastal systems, both economic and environmental, it is imperative that more effective strategies be found for coastal restoration, stabilization, and adaptation. This requires an investment in fundamental science to develop a far greater understanding of the interactions and feedbacks among hydrodynamics, morphodynamics, ecosystem response, mitigation strategies, human agency, and economic valuation than is presently available. For example, beach stabilization by sand addition (beach nourishment) may have significant negative impacts on beach ecosystems, but neither the monitoring nor the understanding of the underlying physical and biological processes is adequate to evaluate the long-term risks associated with this practice (Peterson and Bishop, 2005). This lack of understanding extends to the full range of coastal environments and includes such fundamental questions as the degree and nature of coastal protection offered by mangroves, wetlands, reefs, and dunes (Barbier et al., 2008; Valiela and Fox, 2008; Feagin et al., 2009, 2010; see Figure 2.22). Several recent studies have attempted to couple models of ecogeomorphological processes with economic models (e.g., McNamara and Werner, 2008); to identify strategies for moving toward a more rational assessment of, for example, the minimum level of landscape stability needed for human occupation of coastal environments (Feagin et al., 2010); and to consider the role of human adaptation in scenarios of future coastal change (Nicholls and Cazanave, 2010).
Technical and methodological advances are also shedding new light on coastal processes. Methods that
were in their infancy a decade ago, such as Acoustic Doppler Current Profilers (ADCPs), have matured to the point where they are now available as off-the-shelf technology. Near-shore currents were previously measured at discreet points that were interpolated and modeled to infer the flow field. The advent of ADCPs allows true three-dimensional flow fields to be measured for the first time and is leading to significant advances in coastal science. These technical advances and coupled ecogeomorphological-economic models represent first steps in what must be a transdisciplinary effort among scientists studying coastal processes and ecosystems, engineers, economists, and other social scientists to address the pressing problems facing coastal environments. Advances in coastal sciences would be accelerated by a dedicated NSF initiative that integrates physical, chemical, and biological processes with human activities and their interconnections across coastal watersheds, into the coastal zone, and beyond to the near-shore zone. This effort will necessarily involve several GEO divisions but is most naturally led by EAR because the majority of processes in question are solid-earth processes.
Human land use, climate change, and energy demand are transforming geochemical and geobiological systems and, in particular, the cycling of water, carbon, and nitrogen in these settings. Humans are now managing and altering 50 percent of Earth’s land surface—dubbed the “critical zone” in the Basic Research Opportunities in Earth Science report (NRC, 2001)—and, in so doing, are transforming the physical, chemical, and biological states and feedbacks among essential components of the Earth surface system. Over the past century, soil erosion rates have accelerated; metals and toxins have enriched and mobilized far beyond natural rates; agriculture has industrialized the nitrogen cycle; freshwater usage has grown to exceed recharge in major population centers; and natural ecosystems have been heavily overprinted by fragmentation, extinction, global-scale biogeographic shifts, and invasive species. At the same time, atmospheric temperature and carbon dioxide levels have increased, impacting carbon storage in the terrestrial environment, the water cycle, and a range of intertwined biogeochemical cycles and atmospheric properties that feed back on climate and ecosystems (terrestrial and marine). This research opportunity differs from the earlier section on hydrogeomorphic-ecosystem response in that its roots are more in geochemistry than geomorphology.
EAR is poised to play a leadership role in comprehensive, uniquely integrated studies of the terrestrial environment in the face of human activity and climate change. This work spans diverse programs within EAR and more broadly across diverse divisions and directorates within NSF and other governmental agencies, such as the USGS and DOE. An existing suite of observatories provide insight into Earth’s ecosystems and related dynamics. These natural laboratories include the NSF Critical Zone Observatory (CZO) and Long Term Ecological Research (LTER) programs and those within the National Ecological Observatory Network (NEON) and the Free-Air Carbon Dioxide Enrichment (FACE) program of the DOE. The EarthScope facility also shows potential for providing data needed for ecosystem and water cycle studies through indirect measurements of soil moisture and snow cover from the EarthScope GPS network. These programs are alike in their prioritization of integrated science, and now, increasingly, these complementary programs are philosophically and collaboratively bound together by common goals focused on common questions about terrestrial ecosystems impacted under human influence and climate change.
Integrated Soil, Water, and Biogeochemical Dynamics in the Critical Zone
The dynamics of the critical zone—the dynamic interface between the solid Earth and its fluid envelopes (NRC, 2001)—are governed by the interplay between hydrological, geomorphic, biogeochemical, and biotic processes that transform and rearrange materials in the Earth surface environment. Plant growth, for example, affects surficial weathering and hill slope form through bioturbation, fracture formation, alteration of hydrological fluxes, soil carbon dioxide generation, and profusion of organic weathering reagents. We are not yet able to weave these and other individual processes into a predictive conceptual model of critical zone evolution.
This limitation is primarily due to incomplete knowledge of couplings between the physical, chemical, and biological processes in the critical zone, including both positive and negative feedbacks and their distribution in time and space.
An example of processes not adequately understood at present literally lies beneath our feet. We lack observation and theory of the weathering front (the interface between regolith and bedrock) that strongly influences processes in the critical zone. Coupled with the rapid development of soil ecology as a distinct discipline over the past several decades, this sets the stage for significant advances in our understanding of how life above ground and life below ground are adapted to each other and to spatially variable, hydrogeomorphic processes. The thin layer of weathered rock and soil that mantles Earth’s surface offers exceptional opportunities for research on both fundamental processes shaping landscapes and applied issues related to the geobiological basis for soil fertility. Chemical weathering and erosion of bedrock and soil influence climate, river and groundwater chemistry, bedrock erodibility, and ecosystem properties. Despite the fundamental importance of soil formation and fertility for life on Earth’s surface, soils and the breakdown of rock to form soil remain among the least understood areas of the Earth sciences. Quantifying the controls on rates of rock breakdown to form soils is needed to understand the processes of soil formation and how they vary in different landscapes, climates, and tectonic regimes.
Interdisciplinary studies of the critical zone are yielding new ideas about the interactions of weathering, erosion, and biology in the critical zone. These include hypotheses concerning the evolution of the critical zone, such as that in relatively stable landscapes where biology drives weathering in the initial stages of plant establishment while weathering drives biology over the long term (Brantley et al., 2011). This work also suggests that future land use change may impact critical zone processes more than climate change and that restoration efforts are likely to restore hydrological functions on shorter timescales (decades or less) than biogeochemical functions and biodiversity (Brantley et al., 2011).
A substantial investment in in situ environmental sensors, field instruments, geochemical tools, remote sensing, surface and subsurface imaging, and development of new technologies will be required to test these hypotheses. For example, geochemists now possess powerful tools that permit the characterization of fundamental processes and elemental, molecular, and isotopic properties at scales from submicroscopic to planetary, fueled in part by tremendous advances at the nanoscale and in computational and instrumental toolkits. Among these advances are abilities to date processes in the critical zone at increasingly fine resolution using cosmogenic and uranium series isotope systems.
Two interdisciplinary techniques currently supported by EAR also show significant promise. First, geodetic techniques are increasingly being used to measure changes in the components of the water cycle. Long-term and seasonal subsidence can be observed via GPS and InSAR, providing important constraints on groundwater depletion due to withdrawals for irrigation and municipal use. Gravity data measured using satellites are being used to monitor changes in water storage at the basin scale that cannot be observed using any other technique (Famiglietti et al., 2011). Second, GPS receivers in the EarthScope Plate Boundary Observatory (PBO) are being used to measure critical environmental parameters such as soil moisture, snow depth, biomass changes, and glacier retreat. These data are valuable to both climate scientists and water managers for drought and flood prediction. These PBO studies demonstrate how infrastructure developed for geophysical studies can simultaneously be used for water cycle studies funded through the hydrological sciences within EAR, the Division of Atmospheric and Geospace Sciences (AGS), and non-GEO directorates such as the Directorate for Biological Sciences (BIO) and the Office of Polar Programs (OPP).
The payoffs of such investments in data acquisition are potentially enormous if the fluxes of energy, water, and materials within and through the critical zone can be resolved and if fundamental insight can be provided into ecosystem and landscape evolution and resilience. The data sets and understanding developed through such measurements will form the basis for coupled systems models that allow study of interactions and feedbacks between biological and physical processes in the critical zone through assimilation of hydrological, meteorological, biogeochemical, and geomicrobiological measurements.
Quantitative estimation of watershed carbon bal-
ance provides a compelling example. Findings from the late 1980s to mid-1990s indicating that only ~30 percent of the carbon dioxide released by fossil fuel burning stayed in the atmosphere, with ocean uptake accounting for an additional ~30 percent, launched a stampede of terrestrial ecosystem and surface Earth scientists to every biome on Earth to look for the missing sink for the remaining 40 percent. However, after 15 years of effort, a consensus has yet to emerge regarding the spatial distribution of, or the processes responsible for, the 2 to 4 Pg C y–1 continental sink of the 1990s (Solomon et al., 2007)—or the observation that continents were likely a net carbon source in the 2000s. One roadblock is that net ecosystem production (NEP) measured at local scales does not often extrapolate well to larger scales (Ometto et al., 2005; Stephens et al., 2007), very possibly due to lack of consideration of lateral export (Chapin et al., 2006; Lovett et al., 2006) and the details of spatial and temporal variability. The importance of full watershed-scale carbon balances is illustrated by the one published study that accounted for both vertical carbon fluxes (via eddy covariance tower) and lateral carbon exports via streams, demonstrating that Net Ecosystem Exchange (NEE) went from a net sink of 0.278 Mg C ha–1 yr–1 to a net source of 0.083 Mg C ha–1 yr–1 when lateral stream fluxes were accounted for (Aufdenkampe et al., 2011).
The integrated watershed studies needed to advance our understanding of the critical zone is a distinctive feature of the CZO framework and their multidisciplinary science teams. CZOs provide essential data sets and a coordinated community of researchers who integrate hydrological, ecological, geochemical, and geomorphic processes from mineral grain to watershed scales to illuminate the rich complexity of interactions between the lithosphere, the pedosphere, the hydrosphere, the biosphere, and the atmosphere. CZO sites are establishing infrastructure for the intensive data-gathering effort required to support their science teams and the conceptual and mathematical models they develop (see Box 2.9). The development of more diverse observatory sites could facilitate comparison and sensitivity studies that might then serve with reasonable confidence in a broader predictive mode across non-observatory sites.
Critical Zone Observatories
The Critical Zone concept, introduced in the 2001 NRC report Basic Research Opportunities in Earth Science, provides a research framework for the portion of Earth most closely linked to society and terrestrial life. A network of Critical Zone Observatories (CZOs) is being established to capitalize on this new research framework by providing locations and funding mechanisms for integrated, multidisciplinary research. Five observatories are located in the continental United States and a sixth is in Puerto Rico, and each is in a different representative landscape. This CZO network is connected to an international network through collaboration with a parallel effort in the European Union, and data and infrastructure are open to all researchers. Past studies of the Critical Zone rarely were able to conduct long-term monitoring efforts. Establishing semi-permanent observatories is allowing long-term studies to be conducted and has the potential to fill large gaps in our knowledge of Critical Zone systems. Because human agency plays such a large role in nearly every system of the Critical Zone, the traditional Earth science objective of constructing a universal model cannot be accomplished without including the influence of human activities. This is an evolution in thinking for conventional Earth sciences, but it holds promise of transformative discoveries that will be both useful to society and add value to the larger corpus of Earth science understanding. The CZO network is designed to be the mechanism for making those discoveries.
Responses and Feedbacks of Carbon, Nitrogen, and Water Cycles to Climate Change
Each year about 120 Pg of carbon is exchanged between the atmosphere and terrestrial ecosystems through photosynthesis and respiration. This is more than an order of magnitude larger than estimates of exchange directly due to human activities (8.7 Pg C/year from fossil fuel combustion and 1.2 Pg C/year associated with land use change in 2008; Le Quéré et al., 2009). As a result, global changes in sources and sinks of carbon due to climate change could be at least as important to global carbon cycles as the total of all direct anthropogenic fluxes. Indirectly, humans have and continue to be an important agent of past and future climate change, primarily through fossil fuel burning. Identification of carbon sources and sinks requires studies at landscape and regional scales, whereas most research to date on carbon cycling has
been at global (e.g., Global Climate Models simulations) or local (e.g., flux tower) scales.
Environments in which climate change could trigger relatively rapid vegetation and landscape change, such as permafrost areas and wetlands, are of particular concern to regional and global carbon exchange. For example, there are 23 × 106 km2 of ice-rich permafrost in the Northern Hemisphere, more than a third of which could be actively thawing by 2100, according to model projections (Grosse et al., 2011). An estimated 1,600 Pg C is stored in the top 3 m of ground in Northern Hemisphere permafrost regions (see Figure 2.23). Thawing of permafrost and associated microbial decomposition of organic carbon have the potential to transfer large quantities of carbon to the atmosphere, with estimates in the range of 50 to 100 Pg by 2100 (Schuur et al., 2008). However, assessments of the vulnerability and resilience of permafrost to warming and thawing, as well as potential carbon losses, are complicated by positive and negative feedbacks among snow cover, vegetation, soil, active layer properties, and surface water and groundwater (Grosse et al., 2011). Thawing of permafrost is also likely to produce rapid landscape degradation, including development of thermokarst, accelerated coastal erosion, channel network expansion, and mass wasting (Rowland et al., 2010). Improved understanding of the impacts of climate change on carbon, soil, ecosystem, and landscape dynamics in permafrost regions will require coordinated observation and modeling efforts by multidisciplinary teams of scientists.
The elemental stoichiometry wired into living organisms guarantees that the carbon cycle is coupled with those of nitrogen and phosphorus, while processes such as biological nitrogen fixation link nitrogen cycles to those of other elements, such as iron (e.g., Finzi et al., 2011). Carbon, nitrogen, and other elemental cycles respond variously to changes in temperature and precipitation, and their coupling creates a complex system of interactions and feedbacks among elemental cycles, ecosystems, and climate. The coupling between carbon and nitrogen cycles and climate change is one aspect of this system currently receiving considerable attention owing to uncertainties as to whether feedbacks between nitrogen and carbon cycles will act to buffer or amplify the response of Earth’s climate to continued anthropogenic carbon dioxide flux to the atmosphere.
The terrestrial nitrogen cycle has been dramatically accelerated by industrial production of reactive nitrogen for use as fertilizer, as well as by combustion of fossil fuels and cultivation of legumes. These three anthropogenic sources of nitrogen are estimated to have added more nitrogen (187 Tg N/year in 2005; Galloway et al., 2008) into the terrestrial environment during the past few decades than natural sources (110 Tg N/year; see Figure 2.24; Gruber and Galloway, 2008). In addition, anthropogenic emissions of nitrous oxide (N2O, a greenhouse gas) directly contribute to stratospheric ozone depletion and tropospheric N2O accumulation (Ravishankara et al., 2009), while emissions of nitrogen oxides (NOx) indirectly contribute to tropospheric ozone and aerosol formation (Arneth et al., 2010).
FIGURE 2.23 Idealized cross section through northern permafrost regions indicating significant known and assumed carbon pools, including estimated carbon storage in petagrams (Pg C) for the terrestrial and marine portions of the permafrost system. SOURCE: Grosse et al. (2011).
FIGURE 2.24 Natural (blue) and anthropogenic (orange) nitrogen fluxes for the terrestrial (left) and marine (right) nitrogen cycles. Illustrates major sources, sinks, and processes associated with production of reactive nitrogen and the coupling of the nitrogen cycles with those of carbon and phosphorus. Values are for the 1990s in Tg N/year. SOURCE: Gruber and Galloway (2008). Reprinted by permission from Macmillan Publishers Ltd.
Understanding how all of this additional nitrogen will affect climate, terrestrial ecosystems, and carbon cycling is essential as we attempt to anticipate future environmental change and possible mitigation strategies. For example, recent modeling studies indicate that nitrogen feedbacks represent an important control on changes in terrestrial carbon storage driven by increases in atmospheric carbon dioxide, though the nature of this control varies between tropical, temperate, and high-latitude ecosystems and the magnitude of the effect remains uncertain (e.g., Zaehle et al., 2010). Nitrogen-related changes in carbon storage feed back into climate by regulating atmospheric carbon dioxide levels. In addition, any changes in the C:N ratio of terrestrial plants and/or changes in rates or the geographic distribution of biological nitrogen fixation and denitrification would alter regional and global carbon cycles (Gruber and Galloway, 2008). Redistribution of nitrogen, carbon, and other elements in terrestrial systems by runoff, land-atmosphere exchange, and other surface processes connects the biogeochemical cycles operating in soil-based, freshwater, and marine systems. Quantifying changes in the water cycle associated with climate change is therefore a critical element of building an understanding of future changes in biogeochemical cycles.
Reconstruction of the monthly discharge of the largest rivers by Labat et al. (2004) indicates that global continental runoff increased during the 20th century. Changes in runoff have been linked to changes in precipitation, evapotranspiration, and land use. Modeling of the relative contributions of precipitation, temperature, carbon dioxide concentration, land cover, and land use to increases in river discharge in the 20th century indicates that increases in precipitation are the dominant driver of global increases in discharge (Gerten et al., 2008). Precipitation is expected to increase with increasing temperature, though the rate of increase may be moderated by the influence of tropospheric greenhouse gas forcing and black carbon aerosols on precipitation (e.g., Frieler et al., 2011). Land use practices also contribute to increases in discharge, particularly
in watersheds characterized by extensive agriculture or deforestation. For example, there is a strong correlation between agricultural land cover in the Mississippi River basin and increased discharge under average precipitation conditions, with agricultural land use accounting for more of the increase in Mississippi River discharge in the past 50 years than do increases in precipitation (Raymond et al., 2008). This agriculturally enhanced runoff can carry high concentrations of nitrogen, phosphorus, and carbon (in the form of bicarbonate) that impact the biogeochemistry of the receiving rivers and downstream marine systems.
The role of climate-related changes in evapotranspiration in the intensification of the water cycle is more challenging to sort out, in part because of feedbacks between evapotranspiration and soil moisture. Elevated atmospheric carbon dioxide has been tied to decreases in stomatal conductance (e.g., Leakey et al., 2009), which could lead to decreased evapotranspiration and increased soil moisture (e.g., Gedney et al., 2006). However, several lines of hydrological evidence (water balance estimates, lysimeter and pan evaporation measurements, length of growing season) point to an increase in evapotranspiration in temperate regions over the past 50 years (Huntington, 2008). These results suggest that, at present, the effects of higher temperatures are generally able to offset the effects of increased carbon dioxide on evapotranspiration, though their relative effects are likely to vary geographically and may change with future changes in climate and land cover.
While there are relatively long and spatially distributed records of runoff and precipitation, fundamental hydrological parameters like soil moisture and evapotranspiration are difficult to measure and, for the most part, existing data are temporally and spatially sparse. To advance the science, measurements at points on the landscape (e.g., from networks of flux towers) will have to be integrated smoothly with areally distributed estimates derived from remote sensing (e.g., satellite measurements of soil moisture). All these measurements will have to be coordinated through new data assimilation methods with new theory appropriate for landscape and regional scales. These and other new approaches to quantifying essential hydrological parameters are necessary to resolve spatial and temporal trends in the water cycle and related biogeochemical cycles caused by climate change as well as by land use change and other human impacts.
Human Impacts on Water, Carbon, and Nitrogen Cycles
Humans have altered the terrestrial water cycle through activities like reservoir construction, agriculture, groundwater extraction, and urbanization. More than half (52 percent) of the world’s largest rivers are regulated by dams, including 85 percent of the most biogeographically diverse large river systems (systems that span five or more biomes; Nilsson et al., 2005). Regulation and fragmentation of rivers by dams also strongly impact sediment storage and the discharge of terrestrial sediment to the coastal ocean. While surface freshwater resources exceed global water demand at present, variations in water availability and demand in time and place result in regions of high water stress. In these water-stressed regions, groundwater withdrawal often exceeds recharge, with recent estimates suggesting that groundwater depletion (withdrawal in excess of recharge) has more than doubled since the 1960s (Wada et al., 2010). Virtual trade of water used in the production of goods or services is likely to become increasingly important in supporting human populations in water-stressed regions, especially during drought, but may also facilitate unsupportable population growth in regions of water scarcity (D’Odorico et al., 2010a). Accurate assessments of water availability, water demand, and sustainable water use require more complete global hydrological data sets, compilations of operational data regarding water use, and advances in modeling coupled with hydrological and socioeconomic systems.
Because of the centrality of the carbon cycle to climate, it is critical that the effects of human activities on the carbon cycle be quantified, that the response of the carbon cycle to disturbance be determined, that potential future impacts on carbon cycling and carbon pools (e.g., ocean acidification and methane dynamics) be evaluated, and that possible mitigation strategies be considered (Canadell et al., 2010). The potential for rising atmospheric carbon dioxide levels to significantly impact climate, ecosystems, and human populations has given rise to a variety of ideas for slowing rates of future increases in atmospheric carbon dioxide, rang-
ing from energy-saving measures and use of renewable energy sources to schemes for increasing terrestrial and marine carbon storage (Gussow et al., 2010). Proposed engineered approaches to reducing atmospheric carbon dioxide include ocean iron fertilization, large-scale forestation using nonnative species, and injection of carbon dioxide in deep-sea sediments and aquifers. Geoengineering proposals for carbon storage can involve substantial risks, possible unintended consequences, and potentially limited benefit (Bala, 2009; Finzi, 2011). Both the American Meteorological Society (AMS) and the American Geophysical Union (AGU) have adopted position statements on geoengineering that recommend further research on the intended and unintended Earth system response to geoengineering proposals and coordinated, interdisciplinary study of the relevant scientific, social, legal, and ethical issues (AGU, 2009; AMS, 2009).
Humans have also significantly impacted other biogeochemical cycles. As noted above, industrial production of fertilizer, fossil fuel combustion, and cultivation of legumes are currently adding more new reactive nitrogen to the environment than natural terrestrial processes. Impacts of reactive nitrogen on the environment are exacerbated by its cascading effect as it moves through the environment, such that each molecule of nitrogen can contribute to multiple environmental problems. Future population increases, and improvements in standards of living, will likely add to this anthropogenic nitrogen load through growing use of energy, additional demand for food production, and improvements in diet. Policies and practices for nitrogen use must balance the excesses and inefficiencies associated with nitrogen use in much of the developed world with the need for food in other parts of the world (Galloway et al., 2008).
Understanding these and other anthropogenic impacts on the environment requires integrated, interdisciplinary studies of climate, biogeochemical cycles, water, ecosystems, and humans. In particular, it is important that Earth scientists identify processes and thresholds that, when crossed, would lead to irreversible and unacceptable environmental change. Rockström et al. (2009) suggest that this threshold has already been crossed with respect to atmospheric carbon dioxide, the nitrogen cycle, and biodiversity loss (see Figure 2.25).
FIGURE 2.25 Earth-system processes and their proximity to crossing threshold conditions that lead to unacceptable environmental change. Processes are indicated outside each sector. Green colors denote safe operating conditions. The heights of the red-colored wedges represent the status of each process with respect to safe operating conditions. In this figure, climate change, the nitrogen cycle, and biodiversity loss have crossed the threshold of unacceptable environmental change. SOURCE: Rockström et al. (2009). Reprinted by permission from Macmillan Publishers Ltd.
A common theme running through previous sections of this chapter is the growing reliance on geochronology to provide quantitative estimates of the age, duration, and rate of events and processes over many different timescales. As a result of improvements in analytical methods and in the theoretical underpinnings and calibrations of a variety of dating methods, the past few years have seen transformative advances in many approaches to geochronology. Areas of notable growth include surface exposure dating using rare isotopes produced by cosmic rays, determination of cooling histories of rocks (thermochronometry), extremely high precision dating of volcanic ashes, and high-throughput dating of detrital minerals. These geochemical techniques provide quantitative estimates of time that are an essential complement to dates and rates established using magnetostratigraphy and increasingly reliable methods of cyclostratigraphy (counting of orbitally paced oscillations recorded in sedimentary rocks).
Recent work greatly improving the ability to extract extremely precise and accurate ages from both the U/Pb and 40Ar/39Ar methods underscore recent advances and illustrate likely future directions both in terms of method development and application.
High Precision-High Accuracy Radiometric Dating
Given the wide applicability of the U/Pb and 40Ar/39Ar methods, especially to dating ashfalls in sedimentary sequences, recent improvements have had and will continue to have a major impact on the Earth sciences. In the case of U/Pb dating, a remarkable series of discoveries culminating in the work of Mattinson (2005) has revealed an analytical approach by which the consequences of Pb loss on zircon U/Pb dates can be almost entirely removed. This new approach permits routine determination of U/Pb dates with a precision of better than 0.1 percent. Geochronologists are also continuing to reduce other sources of error, including spike calibration, instrumental mass fractionation, decay constants, and the magma chamber residence time of zircon crystals prior to eruption and deposition.
Profound new insights into the rates of geochemical and biological processes are possible with ages precise to a small fraction of a percent. For example, Maloof et al. (2010) recently investigated a portion of the early Cambrian period associated with the appearance of the first calcite biomineralizing organisms and an associated dramatic change in global carbon cycling, as indicated by a large δ13C shift of marine carbonate (see Figure 2.26). Dates of multiple ash fall zircons show that the event occurred at 525.34 ± 0.09 Ma, and the adjustment in global carbon cycling occurred in 506 ± 126 kyr. The rate of this event suggests that these changes arose from biological diversification occurring at that time.
The ability to obtain extremely accurate and reliably inter-calibrated ages allows previously impossible high-precision cross-correlation of events recorded in different localities. For example, Schoene et al. (2010) dated the end-Triassic mass extinction to 201.32 Ma in sedimentary sections in both Peru and Nevada and determined that the extinction was complete in <300 kyr. Additional dates from the Central Atlantic Magmatic Province yielded precisely the same age, providing compelling evidence of a linkage between the extinction and massive volcanic eruptions.
Similar advances have occurred in 40Ar/39Ar dating, which is important because not all samples of interest contain datable zircons. Furthermore, the ability to date coexisting minerals by two different high-precision methods allows the detection of possible age biases arising from such factors as daughter product loss, inheritance, and magma residence time. Much of the improvement in 40Ar/39Ar dating has arisen from refinements to the 40K decay constant (Renne et al., 2010) and to the ages of the standards that are essential to the method. As an important example of standard calibration, Kuiper et al. (2008) assigned extremely precise and accurate ages from the astronomical timescale (counting of Milankovitch cycles) to ashfall sanidines in Miocene sediments. These sanidines were analyzed for 40Ar/39Ar ratio and then used to back-calculate the true age of the widely used Fish Canyon sanidine standard (the new age of this standard of 28.201 ± 0.046 Ma is remarkably more precise than the previously adopted value of 28.02 ± 0.56 Ma).
Productive interplay between astronomical dating and the improved accuracy of the 40Ar/39Ar and U/Pb
FIGURE 2.26 An example of the new insights possible with ages precise to a small fraction of a percent. Chart shows carbon isotope variability in marine carbonate in the early Cambrian period. High-precision U/Pb zircon ages of intercalated tuffs shown in boxed numbers (in Ma). SOURCE: Maloof et al. (2010). Reproduced with permission of Geological Society of America.
chronometers is likely to continue in the coming years. Such interplay is nicely illustrated by work on the Cretaceous-Tertiary (K-T) boundary (Kuiper et al., 2008). While excellent cyclostratigraphy is apparent in some K-T boundary sections (see Figure 2.27), there is ambiguity in precisely how to map the sedimentary signals to the independently computed astronomical forcings. Improved dating accuracy has provided a new, high-accuracy-age tie point at the K-T boundary. This new tie point provides a new and more robust (but not yet definitive) age anchor on which to pin the astronomical timescale.
Geochronology has its roots in analytical geochemistry and has greatly benefited from improvements in instrumentation and in a refined understanding of the underlying geochemical principles. Geochronology
FIGURE 2.27 Milankaovitch cycles at the Zumaya K-T boundary section, Spain. High-precision radiometric dates permit improved assignment of the absolute ages of these cycles and hence a more accurate geological timescale. SOURCE: Kuiper et al. (2008). Reprinted with permission from AAAS.
is a vibrant research subdiscipline, and the next decade will likely see continued advances in this area. However, as the fidelity, availability, and diversity of dating methods expand, the need for close collaboration among those who develop techniques and make the measurements with those who select key samples and interpret results is becoming increasingly apparent. In many cases—for example, in surface exposure dating and thermochronometry—sophisticated models are essential to extract the full meaning from the data. Thus, continued and robust advances in geochronology will involve a broad cross section of the Earth science community.