Shallow Lithosphere Research Articles

The dynamics of subduction and trench migration for viscosity stratification

SUMMARY Although subduction plays a dominant role in plate tectonics, related processes such as trench migration are not well understood. We investigate the influence of viscosity stratification and different flow boundary conditions on the trench migration dynamics of a subducting lithosphere in a 2-D model using a finite-difference code. Subduction and mantle flow are driven by a compositionally prescribed density contrast between the high-viscosity slab and the ambient mantle; no overriding plate is assumed. The model rheology is viscoplastic at shallow, and viscous at greater depth; the mantle is stratified with a higher viscosity in the lower part of the computational domain. The plastic rheology allows the plate to decouple from the top of the box and is described by a simplified friction law to mimic the brittle deformation of the shallow lithosphere. The flow patterns and slab morphologies differ as a function of the velocity boundary conditions (BCs). For reflective BCs, we model either a plate which is fixed on the box side or a plate that can freely move. A slab with periodic BCs is always laterally free. We find that, when the slabs interact with the highly viscous lower mantle, the slab with reflective BCs is more folded than the periodic one, due to the confinement of mantle flow around the slab. In general, the trench is observed to roll back toward the oceanic plate. For free slabs with reflective BCs the trench retreat velocity deceases after the interaction with the lower mantle. In contrast, for the free slab with periodic BCs it is nearly constant or even increases with time if described in a reference frame fixed to the lower mantle. The free slab with periodic BCs exerts a force on the lower mantle, which causes a net differential flow between the upper and lower mantle. We also carried out additional experiments with different mechanical slab properties and thicknesses. Stiffer, free slabs subduct more steeply and are able to penetrate straight into the lower mantle. After penetration into the lower mantle the trench moves toward the continent. Thinner slabs deform more easily and are folded for all boundary conditions. We can generalize the rollback systematics for free slabs by normalizing trench velocities by the plate speed. Free slabs, being more representative of the West Pacific subduction zones, show large temporal variations of normalized trench velocities ranging between a retreat value of 1.0 to an advancing value of −0.3.

Subducting oceanic crust: The source of deep diamonds

Inclusions of majoritic garnet in diamonds from the Jagersfontein kimberlite formed at unusually great depths of ∼250 to >500 km in the asthenosphere and transition zone. The original host rocks were derived from a much shallower, basaltic (eclogitic) source. The presence of negative Eu anomalies in all majoritic garnets requires a crustal origin, thereby linking these very deep diamond sources to subducting oceanic crust. The carbon isotope values (δ13C) of the host diamonds fall within a narrow range at ∼−20‰, which is fundamentally different from the broad range (−24‰ to −2‰) and bimodal distribution of carbon isotopes of Jagersfontein diamonds that formed in the shallower lithosphere. This indicates that majoritic garnet-bearing diamonds at Jagersfontein inherited their light carbon isotopic composition directly from organic matter contained in a subducting slab. These diamonds were likely formed by direct conversion from graphite, well within the diamond stability field.

Stratification of anisotropy in the Pacific upper mantle

On the basis of the use of broadband (25–150 s) Rayleigh wave group speeds to estimate the 2ψ component of azimuthal anisotropy, we present evidence for a stratification of anisotropy in the uppermost mantle at large scales across the Pacific basin. We confirm previous surface wave studies that established that the fast axis directions of azimuthal anisotropy for intermediate‐ and long‐period Rayleigh waves approximately align with present‐day plate motions. At shorter periods (25–50 s), however, fast axes align nearer to the paleospreading or fossil spreading direction which differs from present‐day plate motion in the old Pacific. These observations, as well as observations of the age dependence of the amplitude of azimuthal anisotropy, imply that azimuthal anisotropy in the Pacific upper lithosphere (<100 km depth) is fixed or “fossilized,” on average, reflecting the strain conditions extant during the early evolution of the lithosphere rather than the current ambient flow direction. In the deeper lithosphere and asthenosphere, anisotropic fast axis directions align nearer to present‐day plate motions, apparently having reoriented to conform to the current conditions of mantle flow. The mechanism of anisotropy stratification remains unclear, but observations are consistent with the anisotropy of the shallow lithosphere being fixed because the shear flows that can produce dislocation creep and a change in anisotropy will occur at increasing depths as the plate ages.

Genesis and evolution of the lithospheric mantle beneath the Buffalo Head Terrane, Alberta (Canada)

Mantle xenoliths and xenocrysts were retrieved from three of the 88–86 Ma Buffalo Hills kimberlites (K6, K11, K14) for a reconnaissance study of the subcontinental lithospheric mantle (SCLM) beneath the Buffalo Head Terrane (Alberta, Canada). The xenoliths include spinel lherzolites, one garnet spinel lherzolite, garnet harzburgites, one sheared garnet lherzolite and pyroxenites. Pyroxenitic and wehrlitic garnet xenocrysts are derived primarily from the shallow mantle and lherzolitic garnet xenocrysts from the deep mantle. Harzburgite with Ca-saturated garnets is concentrated in a layer between ∼135–165 km depth. Garnet xenocrysts define a model conductive paleogeotherm corresponding to a heat flow of 38–39 mW/m 2. The sheared garnet lherzolite lies on an inflection of this geotherm and may constrain the depth of the lithosphere–asthenosphere boundary (LAB) beneath this region to ca 180 km depth. A loss of >20% partial melt is recorded by spinel lherzolites and up to 60% by the garnet harzburgites, which may be related to lithosphere formation. The mantle was subsequently modified during at least two metasomatic events. An older metasomatic event is evident in incompatible-element enrichments in homogeneous equilibrated garnet and clinopyroxene. Silicate melt metasomatism predominated in the deep lithosphere and led to enrichments in the HFSE with minor enrichments in LREE. Metasomatism by small-volume volatile-rich melts, such as carbonatite, appears to have been more important in the shallow lithosphere and led to enrichments in LREE with minor enrichments in HFSE. An intermediate metasomatic style, possibly a signature of volatile-rich silicate melts, is also recognised. These metasomatic styles may be related through modification of a single melt during progressive interaction with the mantle. This metasomatism is suggested to have occurred during Paleoproterozoic rifting of the Buffalo Head Terrane from the neighbouring Rae Province and may be responsible for the evolution of some samples toward unradiogenic Nd and Hf isotopic compositions. Disturbed Re–Os isotope systematics, evident in implausible model ages, were obtained in situ for sulfides in several spinel lherzolites and suggest that many sulfides are secondary (metasomatic) or mixtures of primary and secondary sulfides. Sulfide in one peridotite has unradiogenic 187Os/ 188Os and gives a model age of 1.89±0.38 Ga. This age coincides with the inferred emplacement of mafic sheets in the crust and suggests that the melts parental to the intrusions interacted with the lithospheric mantle. A younger metasomatic event is indicated by the occurrence of sulfide-rich melt patches, unequilibrated mineral compositions and overgrowths on spinel that are Ti-, Cr- and Fe-rich but Zn-poor. Subsequent cooling is recorded by fine exsolution lamellae in the pyroxenes and by arrested mineral reactions. If the lithosphere beneath the Buffalo Head Terrane was formed in the Archaean, any unambiguous signatures of this ancient origin may have been obliterated during these multiple events.

High-resolution body wave tomography beneath the SVEKALAPKO array: I.A priorithree-dimensional crustal model and associated traveltime effects on teleseismic wave fronts

Assessment of contributions from shallow lithosphere to teleseismic wave front distortion is a prerequisite for high-resolution regional teleseismic tomography. Several methods have been proposed in the past for the correction of these effects, e.g. application of station correction terms. We propose an approach that is independent of the subsequent inversion and uses the available a priori knowledge of the crustal structure to calculate crustal traveltime effects of teleseismic wave fronts. Our approach involves the construction of a 3-D crustal model based on controlled source seismology data and calculation of the associated traveltime anomalies for incoming teleseismic wave fronts. The model for central Fennoscandia shows a maximum crustal thickness of 64 km and includes a high-velocity lower crust as derived for parts of the study area by previous authors. Traveltimes calculated using finite differences for teleseismic waves travelling through this crustal model are compared with those from the standard reference model IASP91 and the residuals are used to correct observed teleseismic arrival times at the SVEKALAPKO array. To test the performance of this approach, in a second part of the study a synthetic traveltime data set is obtained by tracing wave fronts through a mantle structure with known velocity anomalies and the 3-D crustal model. This data set is inverted with and without correction for crustal effects. The 3-D crustal effects alone with a homogeneous mantle are also inverted and the results show that the crustal effects propagate down to 450 km. The comparison of the inversion results demonstrates the need to apply appropriate 3-D crustal corrections in high-resolution regional tomography for upper-mantle structure beneath the Baltic Shield.

Constraints on the Trace Element Composition of the Archean Mantle Root beneath Somerset Island, Arctic Canada

Peridotites that sample Archean mantle roots are frequently in- INTRODUCTION compatible trace element enriched despite their refractory major The mantle underlying many Archean cratons has anelement compositions. To constrain the trace element budget of the omalously high seismic velocities to depths of 350–400 km lithosphere beneath the Canadian craton, trace element and rare ( Jordan, 1988; Grand, 1994), indicating the presence of earth element (REE) abundances were determined for a suite of cold refractory roots, depleted in the fusible major elegarnet peridotites and garnet pyroxenites from the Nikos kimberlite ments compared with fertile mantle (Boyd & Mertzman, pipe on Somerset Island, Canadian Arctic, their constituent garnet 1987; McDonough, 1990). These deep residual peridotite and clinopyroxene, and the host kimberlite. These refractory mantle roots probably contribute to the stability of Archean xenoliths are depleted in fusible major elements, but enriched in continental lithosphere because of their lower density and incompatible trace elements, such as large ion lithophile elements higher viscosity compared with that of the surrounding (LILE), Th, U and light rare earth elements (LREE). Mass asthenospheric mantle (Boyd & McCallister, 1976; Jorbalance calculations based on modal abundances of clinopyroxene dan, 1979; Pollack, 1986). Mantle xenoliths that are and garnet and their respective REE contents yield discrepancies hosted by kimberlites and alkaline basalts are our only between calculated and analyzed REE contents for the Nikos bulk window into the subcontinental lithosphere. They provide rocks that amount to LREE deficiencies of 70‐99%, suggesting essential evidence on the chemical composition and evoluthe presence of small amounts of interstitial kimberlite liquid tion of the upper mantle to depths of >200 km. Studies (0·4‐2 wt %) to account for the excess LREE abundances. These of these mantle xenoliths enable us to characterize the results indicate that the peridotites had in fact depleted or flat abundance and distribution of major, minor and trace LREE patterns before contamination by their host kimberlite. LREE elements in peridotites and between their constituent and Sr enrichment in clinopyroxene and low Zr and Sr abundances minerals. Most subcratonic peridotite samples have in garnet in low-temperature peridotites (800‐1100°C) compared undergone a complex history of melt extraction that has with high-temperature peridotites (1200‐1400°C) suggest that the changed their chemical composition and resulted in shallow lithosphere is geochemically distinct from the deep lithosphere depletion of the residual mantle in fusible major elements beneath the northern margin of the Canadian craton. The Somerset such as Fe, Al and Ca (e.g. Nixon, 1987; Herzberg, 1993; mantle root appears to be characterized by a depth zonation that Boyd et al., 1997). In contrast to their depletion in may date from the time of its stabilization in the Archean. incompatible major elements, however, many peridotite

Timing of volcanism in the Sierra Nevada of California: Evidence for Pliocene delamination of the batholithic root?

Recent seismic experiments across the southern Sierra Nevada, California, show that the range lacks a thick crustal root. Xenolith studies indicate that delamination and loss of much of the lower crust may have occurred between 10 and 3 Ma. We estimate that delamination occurred ca. 3.5 Ma on the basis of a sudden pulse of mafic potassic magmatism within and just east of the Sierra Nevada from 4 to 3 Ma. This pulse interrupted the low level of volcanism that had prevailed regionally since 13 Ma. Delamination could have promoted magmatism via decompression melting, heating of the shallow lithosphere by upwelling asthenosphere, and foundering of crustal rocks into the mantle, and could have provided dynamic support to the range. Foundering of the lower crust may have occurred spontaneously, owing to negative buoyancy, or may have been triggered by extension between the Sierra Nevada and Death Valley.

Isotope geochemistry of the Darwin Rise seamounts and the nature of long‐term mantle dynamics beneath the south central Pacific

The Darwin Rise region in the west central Pacific Ocean contains an extraordinarily large number of Cretaceous seamounts with isotopic compositions spanning a range nearly equal to that of the global ocean island basalt data set. These edifices formed in what is now the south central Pacific, which is characterized by anomalously shallow lithosphere and concentrated intraplate volcanism. We present Sr, Nd, and Pb isotopic data for Darwin Rise seamount basalts recovered by Ocean Drilling Program Leg 143 and Deep Sea Drilling Project Leg 7. Comparison of chemical and isotopic data for Darwin Rise and French Polynesian volcanoes confirms that the same range of distinct sources has been tapped by intraplate volcanism in the south central Pacific since the Early Cretaceous. Moreover, the increased data coverage reveals that most of the isotopic variation within Darwin Rise volcanic chains can be explained by shallow mixing between plumes and asthenosphere or oceanic lithosphere. On the basis of geophysical and geochemical evidence we propose an alternative model for volcanism in the south central Pacific from the Cretaceous to the present. Instead of the conventional hypothesis involving a few long‐lived mantle plumes, we propose that numerous hotspots formed and died out in the region over the past 130 Myr. We envision that these hotspots are the surface expression of “plumelets” continually separating from a shallow thermochemical boundary layer at the top of a large heterogeneous body of material which rose from the lower mantle. These plumelets rise and melt beneath the south central Pacific lithosphere, producing chains of volcanic edifices which terminate when they become exhausted of melt. The surfacing of multiple plumelets in proximity can explain the complex temporal, spatial, and geochemical patterns of volcanism observed in most volcanic chains in the south central Pacific and Darwin Rise.

The Darwin Rise: A Cretaceous superswell?

The Japanese Guyots, Wake Guyots, and Mid‐Pacific Mountains are part of a broad area of Cretaceous volcanism in the western Pacific termed the “Darwin Rise.” Based on Seabeam bathymetric data we classify these drowned volcanic islands as: type “A,” those that advanced to the atoll stage before final submergence; type “B,” those that drowned at the barrier reef stage; and type “V,” those with little or no reef material on their volcanic summits. Widespread evidence for karst topography extending to depths of 200 m on the summits of A and B guyots sheds new light on events leading to the synchronous extinction of reefs on the Darwin Rise in the mid‐Cretaceous. We propose that after the formation of the reefs on the A and B guyots, the entire region was elevated at approximately the Aptian‐Albian boundary (113 Ma) to form a superswell similar to that existing now in French Polynesia. The type V guyots formed on this anomalously shallow lithosphere. The demise of the reefs was the direct result of the rise of this superswell, although climate factors may have prevented reef recolonization following its later subsidence.

Chemical constraints on lithosphere composition and evolution beneath the Colorado Plateau

We have measured Sr and Nd isotopic composition and rare earth element (REE) abundances in a variety of inclusions and host minettes from the Navajo Volcanic Field of the Colorado Plateau in order to develop geochemical constraints on the composition and evolution of the mantle beneath the relatively undisturbed Proterozoic crust of the Colorado Plateau. Spinel and garnet peridotites and eclogites derived from less than 80 km in depth have geochemical signatures (e.g., light REE (LREE) depletion) similar to oceanic lithosphere; we interpret these rocks to be fragments of lithosphere, possibly part of an island arc which was accreted to the continent in the Proterozoic. Isotopic evidence suggests that some of the peridotites were hydrated by an incompatible element‐poor, H2O‐rich fluid, whereas the eclogites were metasomatized by a Na2O‐rich phase with high 87Sr/86Sr. Some of the peridotites have highly radiogenic Nd isotopic ratios which reflect long‐term isolation from mantle convection and show that they probably are not samples of recently subducted lithosphere. Garnet peridotites derived from at or near the base of the lithosphere are isotopically distinct from the shallow xenoliths and similar to ocean island basalts. Many of these same peridotites are enriched in the LREE. An inverse correlation between temperature of equilibration and 87Sr/86Sr in these garnet peridotites may reflect infiltration of relatively low 87Sr/86Sr melts into refractory, relatively high 87Sr/86Sr wall rock (lithosphere?). These high‐temperature, low 87Sr/86Sr melts were possibly parental to the megacrystalline inclusions associated with the garnet peridotites. Overall, our data for the Colorado Plateau suggest that highly depleted, shallow lithosphere is underlain by material enriched in incompatible elements. This latter material was located near the inferred Oligocene base of the lithosphere and its composition may reflect mixing between asthenosphere and lithosphere. Melting of this material as a consequence of warming of the Colorado Plateau lithosphere during the late Tertiary may explain the origin of the Hopi Buttes alkaline magmas. The host minettes are isotopically similar to the bulk earth and their major element chemical variation can be explained by fractional crystallization. However, isotopic heterogeneity in the same eruptive center and decreasing Nb and Ta abundances with magmatic evolution indicate a complex petrogenesis involving exotic phases. The variation of isotopic composition with trace element ratios is inconsistent with bulk assimilation of crust, and we suggest that it reflects melting of a veined source containing phlogopite, apatite, and possibly a Nb‐Ta‐rich phase in addition to normal peridotite minerals.

Constraints on thermal models of the basin and range province

Several mechanisms that have been proposed to explain the high heat flow in regions of tectonic extension are examined. The transient thermal regime is computed and the results are discussed. Conductive heating of the lithosphere from below requires a time of at least 75 Ma to affect the surface heat flow. Advective heating by the injection of magmas into the lithosphere does not produce steady-state conditions before 75 to 300 Ma (depending on boundary conditions). Only mechanical stretching of at least the shallow lithosphere can explain the increase in surface heat flow in a time of the order of 30 to 50 Ma. The heat flow data are compatible with several mechanisms for the extension of the lower lithosphere (underplating or dike injection) and other geologic and geophysical data are needed to constrain the models.

Constraints on the origin of Hawaiian lavas

Geochemical and geophysical data can be used to constrain models of Hawaiian lava generation. Geochemical data, including radiogenic isotopes, major and trace elements, and transition metals, indicate that the source region of Hawaiian tholeiites must be within the stability field of garnet. The source material consists of at least three isotopically distinguishable components: depleted, enriched, and primitive mantle. Generation of parental Hawaiian magma by extensive fractional crystallization of high‐magnesium primary mantle melts is difficult to reconcile with high transition metal abundances of Hawaiian tholeiites. Rare earth element modeling indicates that the source of at least some Hawaiian tholeiites does not require a pre‐enrichment (metasomatic) event to generate melt. Where incompatible element abundances increase in lavas with no change in radiogenic isotopes (such as in the Waianae range or in the late shield‐building stage at Kohala), the degree of melting of the tholeiite source is decreasing. Where chemical changes are correlated with isotopic variations (as at Haleakala and Kauai), mixing between tholeiite magma and melt from depleted mantle is indicated. Geophysical evidence suggests that melting occurs at depths greater than 60 km. Thus extensive melting of shallow lithosphere to generate Hawaiian tholeiites is invalid. Melting of deep lithosphere is unlikely because the source of late shield‐building alkalic basalts would be severely depleted in incompatible elements. The origin of the Hawaiian Swell is still open to conjecture and cannot constrain whether melting of lithosphere or emplacement of low‐density material beneath the lithosphere occurs. The orientation of volcanic rift zones and the continuity of tholeiitic magmatism from one center to the next imply that the so‐called “periodicity” of melting in the source region is not critical but that the lithosphere itself exerts strong control over the spacing of volcanoes. Calculated eruption rates for the late shield‐building stage at Kohala indicate an inverse exponential dependence on melt fraction. A model consistent with the preceding constraints involves the partial melting of an upwelling deep‐mantle plume at about 60–100 km. The plume consists of primitive mantle, asthenosphere, and recycled oceanic crust, which become well mixed in the center of the plume. The degree of melting to generate tholeiites may be much higher in the plume than in “typical” mantle because of the high modal abundance of clinopyroxene (from recycled crust). A plume length of 60 km is required for an individual volcano if the degree of plume melting is 20%. Calculated plume ascent rates are very similar to rates of motion for the Pacific plate. Initially, tholeiites erupt almost simultaneously with alkalic lavas (Loihi). As shield building continues, only tholeiitic magmas reach the surface (Kilauea). As the volcanic center moves off the hot spot, either the degree of melting decreases leading to postcaldera or late shield‐building alkalic volcanism (Kohala), or increased mixing of depleted mantle melts with plume‐derived melts occurs (Haleakala). After a long hiatus, resulting from slow segregation rates for very small degrees of melting, nephelinitic volcanism may occur on the flanks of the shield (Haleakala, Kauai).

Time delays in geophysical consolidation problems

Abstract The tectonic response of the lithosphere to loads applied over a period of years is one of the few relatively direct ways of measuring lithospheric mechanical properties. We discuss here a method for estimating gross permeability of shallow lithosphere if such a lithosphere can be modeled as a Biot solid. Induced seismicity at artificial lakes sometimes lags the history of lake filling. Clearly this indicates the anomalous load takes some finite time to create a stress increment over the tectonic regime associated with the lake. Such delays may result from diffusion into inhomogeneous regimes, but intuitively it seems that the Biot consolidation theory ought to contain the physics required to produce delayed response in the simplest model, an isotropic half-space with arbitrary vertical layering. The response of such a half-space can be calculated most quickly from matrix solutions of first-order differential equations. We explore here a consistent formulation for the physics of the problem and examine the relation between the rate of diffusion of changes on the boundary, the geometry of the boundary change, and the physical properties of the material. The resulting formulas can be used to estimate probable delays in response of the physical system. Unfortunately, the values of physical properties required to make such estimates are hard to obtain.

Lithosphere evolution and changing tectonic regimes

A model is presented in which the Earth is characterized by the mantle deep convection, and the thickness of the continental and oceanic lithospheres change with time as the pressure-temperature-geochemical conditions alter in the upper mantle. This model predicts that no island arc differentiation took place before 4000 Ma. During the Archaean, the shallow lithosphere allowed local, tectonic coherence and a gradual increase in the depth of subduction differentiation processes. The change to large cratonic-scale tectonic stability during the Proterozoic occurred as a result of lower lithosphere conditions allowing granulites to be stable—the consequential dehydration of the continental lithosphere creating extensive lateral and vertical stability. The change to ‘modern’ plate tectonics finally took place when the strengths of the oceanic and continental lithospheres became sufficiently close that mantle convection could produce continental splitting. It is suggested that this model accounts for the overall changing tectonic and geochemical features of the continental crust, although such generalities must be applied cautiously to specific localities.