ALBERT

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1996

Description: Two-thirds of the surface of the Earth is created at mid-ocean ridges where magmas rise from the mantle and cool to form the oceanic crust. The objective of this Thesis is to examine the influence of magma supply and eruptive processes on axial morphology, crustal construction, and the properties of crustal magma chambers at intermediate and fast spreading ridges. Variations in magma supply on time scales of ~100 Kyr generate along-axis changes in crustal thickness and temperature. Magma sill properties and hydrothermal activity are closely linked to spreading events which occur on much shorter time scales (ca. 10-100 yr) than the longer-term variations in magma supply reflected in along-axis changes in ridge morphology. The seismically constrained depths of ridge crest magma sills (〉1-2 km) are considerably deeper than the level of neutral buoyancy (100-400 m). The apparent inverse relationship between magma sill depth and spreading rate suggests that a thermally controlled permeability boundary, such as the solidus horizon, controls the depth at which magma ponds beneath mid-ocean ridges. Recent thermo-mechanical models predict that, at intermediate spreading rates, rift valley and magma sill formation are sensitive to small changes in crustal thickness and mantle temperature. Analysis of gravity at an intermediate spreading ridge shows that small differences in crustal thickness (300-700 m) and mantle temperature (10-15°C) are indeed sufficient to produce major changes in lithospheric strength and axial morphology. A stochastic model for the emplacement of dikes and lava flows with a bimodal distribution of lava flows is required to satisfy geological and geophysical constraints on the construction of the extrusive section. Most dikes are intruded within a narrow zone at the ridge axis. Short flows build up approximately half the extrusive volume. Occasional flows that pond at a considerable distance off-axis build up the remainder of the extrusive section. This Thesis underlines the importance of eruption dynamics in the emplacement of the uppermost volcanic layer of the crust and of the crustal thermal structure in controlling local variations in magma sill depth and ridge morphology.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1996

Description: We analyze bathymetric and gravity anomalies at five plume-ridge systems to constrain crustal and mantle density structure at these prominent oceanic features. Numerical models are then used to explore the physical mechanisms controlling plume-ridge interaction and to place theoretical constraints on the temperature anomalies, dimensions, and fluxes of the Icelandic and Galapagos plumes. In Chapter 1 we analyze bathymetric and gravity anomalies along the hotspot-influenced Galapagos Spreading Center. We find that the Galapagos plume generates along-axis bathymetric and mantle-Bouguer gravity anomalies (MBA) that extend 〉500 km east and west of the Galapagos Islands. The along-axis MBA becomes increasingly negative towards the plume center, reaching a minimum of ~-90 mGal near 91°W, and axial topography shallows by ~1.1 km toward the plume. These variations in MBA and bathymetry are attributed to the combined effects of crustal thickening and anomalously low mantle densities, both of which are due to a mantle temperature anomaly imposed beneath the ridge by the Galapagos plume. Passive mantle flow models predict a temperature anomaly of 50±25°C is sufficient to produce the 2-4 km excess crust required to explain the along-axis anomalies. 70-75% of the along-axis bathymetric and MBA variations are estimated to arise from the crust with the remaining 25-30% generated by the anomalously hot, thus low-density mantle. Along Cocos-plate isochrons, bathymetric and MBA variations increase with increasing isochron age, suggesting the subaxial mantle temperature anomaly was greater in the past when the plume was closer, to the ridge axis. In addition to the Galapagos plume-ridge system, in Chapter 2 we examine alongisochron bathymetric and MBA variations at four other plume-ridge systems associated with the Iceland, Azores, Easter and Tristan hotspots. We show that residual bathymetry (up to 4.7 km) and mantle-Bouguer gravity anomalies (up to -340 mGal) are greatest at on-axis plumes and decreases with increasing ridge-hotspot separation distance, until becoming insignificant at a plume-ridge separation of ~500 km. Along-isochron widths of bathymetric anomalies (up to 2700 km) decrease with increasing paleo-spreading rate, reflecting the extent to which plume material flows along-axis before being swept away by the spreading lithosphere. Scaling arguments suggest an average ridgeward plume flux of -2.2x106 km/my. Assuming that the amplitudes of the MBA and bathymetric anomalies reflect crustal thickness and mantle density variations, passive mantle flow models predict maximum subaxial mantle temperature anomalies to be 150-225°C for ridge-center plumes, which decrease as the ridges migrate away from the plumes. The dynamics of mantle flow and melting at ridge-centered plumes are investigated in Chapters 3 using three-dimensional, variable-viscosity, numerical models. Three buoyancy sources are examined: temperature, melt depletion, and melt retention. The width W to which a plume spreads along a ridge axis depends on plume volume flux Q, full spreading rate U, buoyancy number B = (QΔρg)/(48η0U2), and ambient/plume viscosity contrast ϒ according to W=2.37(Q/U)l/2(Bϒ)0.04. Thermal buoyancy is first order in controlling along-axis plume spreading while latent heat loss due to melting, and depletion and retention buoyancy forces contribute second order effects. Two end-member models of the Iceland-Mid-Atlantic Ridge (MAR) system are examined. The first endmember model has a broad plume source of radius 300 km, temperature anomaly of 75°C, and volume flux of 1.2xl07 km3/my. The second model has a narrower plume source of radius 60 km, temperature anomaly of l70°C, and flux of 2.1 x106 km3/my. The first model predicts successfully the observed crustal thickness, topographic, and MBA variations along the MAR, but the second model requires substantial along-axis melt transport in order to explain the observed along-axis variations in crustal thickness, bathymetry, and gravity. We favor this second model because it predicts a mantle P-wave velocity reduction in the plume of ~2% as consistent with recent seismic observations beneath Iceland. Finally in Chapter 4 we use three-dimensional numerical models to investigate the interaction of plumes and migrating midocean ridges. Scaling laws of axial plume spreading width Ware derived first for stationary ridges and off-axis plumes, which yield results consistent with those obtained from independent studies of Ribe [1996]. Wand the maximum plume-ridge interaction distance Xmax again scale with (Q/U)l/2 as in the case of ridge-centered plumes and increase with ϒ and buoyancy number. In the case of a migrating ridge, Xmax is reduced when a ridge migrates toward the plume due to excess drag of the faster-moving leading plate, and enhanced when a ridge migrates away from the plume due to reduced drag of the slower-moving trailing plate. Thermal erosion of the lithospheric boundary layer by the previously ridge-centered plume further enhances Wand Xmax but to a degree that is secondary to the differential migration rates of the two plates. Model predictions are compared with observed along-isochron bathymetric and MBA variations at the Galapagos plume-ridge system. The anomaly amplitudes and widths, as well as the increase in anomaly amplitude with age are predicted with a plume source temperature anomaly of 80-120°C, radius of 80-100 km, and volume flux of 4.5x106 km3/m.y. Our numerical models also predict crustal production rates of the Galapagos Islands consistent with those estimated independently using the observed island topography. Predictions of the geochemical signature of the plume along the present-day ridge suggest that mixing between the plume and ambient mantle sources is unlikely to occur in the asthenosphere or shallow crust, but most likely deeper in the mantle possibly by entrainment of ambient mantle as the plume ascends through the depleted portion of the mantle from its deep source reservoir.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution March 1997

Description: The formation of new oceanic crust is the result of a complex geodynamic system in which mantle rises beneath spreading centers and undergoes decompression melting. The melt segregates from the matrix and is focused to the rise axis, where it is eventually intruded and/or erupted to form the oceanic crust. This thesis combines surface observations with laboratory studies and geodynamic modeling to study this crustal-production system. Quantitative modeling of the crustal and mantle contributions to the axial gravity and topography observed at the East Pacific Rise shows that the retained melt fraction in the mantle is small (〈3%) and is focused into a narrow column extending up to 70 km beneath the ridge axis. Consistent with geochemical constraints, the extraction of melt from the mantle therefore appears to be efficiently focus melt toward the ridge axis. A combination of laboratory and numerical studies are used to constrain the pattern of mantle flow beneath highly-segmented ridges. Even when the buoyant component of mantle flow is constrained to be two-dimensional, laboratory studies show that a segmented ridge will drive three-dimensional mantle upwelling. However, using reasonable mantle parameters in numerical models, it is difficult to induce large-amplitude three-dimensional mantle upwelling at the relatively short wavelengths of individual segments (~50 km). Instead, a simple model of three-dimensional melt migration shows that the observed segment-scale variations in crustal thickness can be explained by focusing of melt as it upwells through a more two-dimensional mantle flow field. At the Reykjanes Ridge, the melt appears to accumulate in small crustal magma chambers, before erupting in small batches to form numerous overlapping hummocky lava flows and small volcanoes. This suggests that crustal accretion, particularly at slow-spreading centers, may be a highly discontinuous process. Long-wavelength variations in crustal accretion may be dominated by variations in mantle upwelling while short-wavelength, segment-scale variations are more likely controlled by a complex three-dimensional processes of melt extraction and magma eruption.

Description: During my first three years in the Joint Program, I was supported by an National Science Foundation Graduate Student Fellowship. Other support has been derived from National Science Foundation grants OCE-9296017, OCE-9224738, OCE-9215544, and EAR grant 93-07400.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution December 1996

Description: The objective of this Thesis was to interpret the structural development of slowspreading ridge segments by: 1) delineating the nature, magnitude, and relative importance of primary tectonic and volcanic processes that control crustal morphology, 2) investigating the spatial and temporal variability of these processes, and 3) examining how rheological variations in the lithosphere control its structural configuration. To that end, this Thesis provides detailed documentation of faults and volcanoes (seamounts) at the Mid-Atlantic Ridge from 25°25'N to 27°10'N and extending from zero-age crust at the ridge axis to -29 Ma crust on the ridge flank. This information was used to analyze the evolution of ocean crust from initial formation in the rift valley to degradation by aging processes on the ridge flank. Accumulation of sediments affects the seafloor morphological expression of ocean crustal structure, and sediment thicknesses were also mapped to facilitate study of the morphological record of crustal accretion and tectonism. In addition, deformation conditions in the lithosphere were analyzed by study of microstructure and geothermometry of abyssal peridotite mylonites recovered from fault zones at slow-spreading ridges.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1998

Description: Planktonic protozoan grazers have the potential to significantly affect the chemistry of particle-associated trace metals. This is due both to the importance of protists as consumers of bacterial-sized particles, and to the unique low-pH, enzyme-rich microenvironment of the grazer food vacuole. This thesis examines the role of protozoan grazers in the marine geochemistry of strongly hydrolyzed, particle-reactive trace metals, in particular Th and Fe. A series of tracer experiments was carried out in model systems in order to determine the effect of grazer-mediated transformations on the chemical speciation and partitioning of radioisotopes C9Fe, 234Th, 51Cr) associated with prey cells. Results indicate that protozoan grazers are equally able to mobilize intracellular and extracellular trace metals. In some cases, protozoan regeneration of trace metals appears to lead to the formation of metal-organic complexes. Protozoan grazing may generate colloidal material that can scavenge trace metals and, via aggregation, lead to an increase in the metal/organic carbon ratio of aggregated particles. Model system experiments were also conducted in order to determine the effect of grazers on mineral phases, specifically colloidal iron oxide (ferrihydrite). Several independent techniques were employed, including size fractionation ors9Fe-labeled colloids, competitive ligand exchange, and iron-limited diatoms as "probes" for bioavailable Fe. Experimental evidence strongly suggests that protozoan grazing can affect the surface chemistry and increase the dissolution rate of iron oxide phases through phagotrophic ingestion. In further work on protozoan-mediated dissolution of colloidal Fe oxides, a novel tracer technique was developed based on the synthesis of colloidal ferrihydrite impregnated with 133Ba as an inert tracer. This technique was shown to be a sensitive, quantitative indicator for the extent of ferrihydrite dissolution/alteration by a variety of mechanisms, including photochemical reduction and ligand-mediated dissolution. In field experiments using this technique, grazing by naturally occuring protistan assemblages was shown to significantly enhance the dissolution rate of colloidal ferrihydrite over that in non-grazing controls. Laboratory and field results indicate that, when integrated temporally over the entire euphotic zone, protozoan grazing may equal or exceed photoreduction as a pathway for the dissolution of iron oxides.

Description: This work was financially supported by a Department of Defense ONR-NDSEG Graduate Fellowship, Office ofNaval Research AASERT Award (N00014-94-1-0711), and the National Science Foundation EGB Program (OCE-9523910).

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1995

Description: The global mid-ocean ridge system is one of the most striking geological features on the surface of the Earth. In this system, the East Pacific Rise (EPR) is the fastest spreading ridge and is thus considered as the most active magmatically among the plate boundaries. In January and February of 1988, an extensive survey by the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution was conducted along the EPR between 9°05' and 9°55'N to study the crustal structure of the axial region. This thesis, the result of that cruise, comprises four main topics: (1) characterization of normal faulting from Sea Beam bathymetric data, (2) application of mechanical models to explore the hypothesis that buoyancy arising from crustal magma chambers and gravitational spreading of the upper crust are the principal processes leading to the initiation and development of normal faults, (3) investigation of seafloor magnetization anomalies to constrain upper crustal structure, and (4) analysis of gravity anomalies to examine possible correlations between observed variations in seafloor manifestations of volcanism and deformation and underlying structure. Thus, each topic focuses on different levels of the mid-ocean ridge. Together with the results of seismic and other observations, the findings are woven into a better understanding of the tectonic processes and structure of fastspreading mid-ocean ridges. First, to understand the characteristics of normal faults at fast-spreading ridges, we utilized swaths of Sea Beam bathymetry and estimated the distribution and geometry of normal fault zones using the slope of the seafloor as the criterion for a faulted surface. In our survey area, nonnal fault activity begins 2-8 km off-axis and continues at least to 30-40 km from the axis, as indicated by an increase in the total and average throws of normal fault zones versus distance from the axis. There appears to be no significant difference in the plan-view area of inward- and outward-facing nonnal fault zones. The distance from the rise axis to the nearest large-offset fault zone (throw 〉 20 m) on either side of the axis is approximately symmetric to the north of 9°23'N, but the midpoint between nearest largeoffset fault zones is offset 2-3 km to the west of the bathymetric axis to the south of 9°23'N. The continued growth of nonnal fault zones suggests that significant extensional stress persists to greater distances from the axis than previously thought and that the rise axis possesses a finite strength. The argument that the rise axis has finite strength is consistent with recent evidence for solidified axial dikes along magmatically active portions of the EPR from near-bottom seismic refraction experiments, which suggests that, while eruption of magma at the rise axis weakens the axis, the persistence of such weak zones is short-lived and the emplacement zones at any given time are localized along the axis. We examined how the presence of a low-density, low-strength magma chamber within the crust and gravitational spreading of a mechanically strong upper crust over an underlying substrate contribute to the fonnation of faults at a fast spreading mid-ocean ridge by comparing the predicted stress field with the observed pattern of normal fault zones. We employed boundary element methods to incorporate buoyancy and gravitational spreading as body forces in an elastic medium, and we detennined stress and strain fields for a variety of rise axis conditions and a range of possible sets of material properties for different parts of the mid-ocean ridge. Our results show that the strength of the rise axis is one of the most crucial factors governing the near-axis stress field. If the rise axis is mechanically weak, the maximum extensional stress from buoyancy occurs at shallow depth off the rise axis. A weak rise axis may result from recent magmatism such as the intrusion of dikes into the upper crust. On the other hand, if the rise axis is mechanically strong, which may result after solidification and cooling of the dike zone, the maximum surface extensional stress occurs on the rise axis. However, the reduction in size of a magma chamber that would accompany cessation of dike injection would lead to less buoyancy and thus a lower likelihood of stress levels sufficient for faulting. For a given set of material strengths and a given magnitude of buoyancy force, the flexural rigidity of the upper crust plays an important role in detennining if a zone of extension will develop off axis and, if so, the position and horizontal extent of that zone. A thin or mechanically weak upper crust is more likely to develop a zone of extension than one that is thick or mechanically strong. The stress field resulting from gravitational spreading is similarly affected by the strength of the rise axis. While buoyancy can explain a consistent distance at which normal faults initiate off-axis, gravitational spreading can account for continued activity on normal faults to a greater distance from the axis than can buoyancy. The existence of a magma lens can play an important role in reducing the magnitude of the stress field for a weak rise axis, as the crust above the magma lens can slide and thus relieve the thickness-averaged extensional stress. Next, we inverted surface ship measurements of the scalar magnetic field along the EPR between 9°10' and 9°50'N. We examined whether the axial magnetization high, which increases in amplitude to the south in our area, can best be explained by variations in the thickness or in the magnetization intensity of the source layer. The variation in axial magnetization is too large to be explained solely by the variations in the depth to the top of the axial magma chamber indicated by reflection seismology. For a magnetic source layer that is 500 or 750 m thick, the observed along-axis variations in FeO and Ti02 explain only 36 and 60%, respectively, of the total variance of axial magnetization anomalies. Therefore, a combination of variations in magnetic layer thickness and in intensity of magnetization (by variations in the FeO and Ti02 contents of the source rock or by other mechanisms) is needed to explain the along-axis variation of axial magnetization. In addition to the increase in amplitude to the south, the axial magnetization high exhibits at least three marked changes in magnitude and offsets in its along-axis linearity ('magnetic devals') (at 9°25', 9°37', and 9°45'N) which appear to be related to boundaries or offsets between the segments of the axial summit caldera (ASC). Because the amplitudes of the axial magnetization anomalies are highest at the midpoints of the ASC segments, we speculate that midpoints of the ASC segments are the loci of more frequent lava eruptions, and the seafloor basalts at the midpoints are thus younger and more magnetic, than at the segment ends. The magnetization shows distinct short-wavelength (~ 5 km) banding to the north of 9°25'N over a region that does not appear to have been affected by an overlapping spreading center. Among the possible explanations for these off-axis magnetization anomalies are short geomagnetic reversal events within the Brunhes epoch, variations in the paleointensity of the Earth's field, variations in the magnetization intensity of the source rock due to variability in the magmatic supply, and variations in the degree of hydrothermal alteration at the rise axis. On the basis of comparisons of forward models and observations, short geomagnetic reversal events appear to be the most likely explanation of these anomalies. The analysis of sea-surface gravity field measurements shows an axial residual mantle Bouguer gravity anomaly too large to be explained by the anomalous temperature of the mantle or by changes in the thickness of the crust. The broad axial residual gravity low is interpreted as a signal arising largely from the upper mantle, presumably by presence of partial melt along the rise axis. A northward increase in the width of the low implies a greater melt fraction in the region to the north than to the south, especially on the Pacific plate side. The residual gravity anomaly also shows several short-wavelength local lows along the axis (e.g., 9°21', 9°32', and 9°42'N) which correlate with along-axis variations in axial magnetization and tomographic images of mid-crustal seismic velocities. Along axis the local lows have an amplitude of 1.5-3 mGal and appear at a nearly regular spacing (10- 15 km). Across the axis, however, the local lows show a greater variation (3-5 mGal), suggesting that there is an additional gravity anomaly signal arising from a low-density structure that is approximately continuous along the axis. The anomalous masses producing the local lows are interpreted as zones of relatively high melt concentration, formed within the crust by recent replenishment of magma from the upper mantle, that are surrounded by a region of lesser melt concentration corresponding to the low-velocity volume imaged by seismic tomography. If the zone of high melt concentration are modeled as circular rods of radius 1 km, along-axis length 10 km, and center of mass 2.25 km below the seafloor, density contrasts of 200-350 kg/m3 are needed to match the observed anomalies. For larger anomalous mass volumes, the density contrasts would be lower. The findings of this study support the hypothesis that the axis of the EPR can be divided into segments 10-15 km in length, with each segment defmed by the locus and timing of most recent emplacement of magma in the axial crust. The segments in the study area appear to be in different phases of a magmatic cycle, but the period of such a magmatic cycle is not known. By this view, the discrete emplacement of magma bodies gives rise to along-axis variations in crustal structure manifested as short-wavelength residual gravity anomalies and magnetic devals. Another consequence of a rise axis at which magma is emplaced at discrete locations is that the mechanical strength of the axial upper crust varies with position along the axis and over time. During active magmatism, the rise axis acts as a weak zone and the buoyancy of the axial magma chamber and surrounding low-velocity volume can lead to initiation of off-axis normal faulting. However, for a long segment of the rise bounded by transform faults, the axis will have sections with a solidified rucial injection zone as well as sections undergoing active magmatism, and thus the rise overall may appear to have finite strength. If such a finitestrength ridge axis is subject to significant extensional stress as a result of gravitational spreading, mantle convective tractions, or differential cooling, then continued normal fault activity would extend over a broad region to distances of at least several tens of kilometers from the spreading axis.

Description: The work in this thesis was supported by the National Science Foundation under grants OCE-8615797, OCE-8615892, and OCE-9000177.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1999

Description: Changes in morphology of the Marquesas Fracture Zone are correlated with small changes in Pacific-Farallon relative motion. The simple flexural signal of a locked fracture zone may be obscured by tectonic effects, and there is no evidence for the release of shear stress on the fracture zone by vertical slip after leaving the active transform. One such small change in plate motion is documented in the South ern Austral Island region of the South Pacific. A twelve degree clock wise change in Pacific-Farallon relative motion occurred around fifty million years ago. This Eocene change in spreading direction and rate is locally constrained with observations of magnetic anomalies and spreading fabric orientation. At the southeastern end of the Cook-Austral Island chain, multiple episodes of volcanism have left a diverse population of seamounts. Volume estimates from geophysical data and modeling show that one-half to two-thirds of the volcanic material is over thirty million years old, while the remainder is less than five million years old. Seismic and bathymetric data imply the presence of abyssal basalt flows in the flexural moat of the Austral Islands, probably associated with Austral Islands volcanism, which may contribute a significant amount of material to the archipelagic apron.

Description: The research presented in Chapter 2 was supported by National Science Founda tion grants OCE-9012949 and OCE-9012529. Chapters 3, 4 and 5 were supported by National Science Foundation grant OCE-9415930. A National Science Foundation graduate fellowship supported my first three years of graduate study.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1988

Description: Heat flow, tectonic subsidence and crustal thickness distributions in the Ligurian Basin are best explained by asymmetric lithospheric thinning mechanisms. Over 150 heat flow measurements are made on several transects between Nice, France and Calvi, Corsica on continental slope and rise settings. Thermal gradient determinations are improved using an optimization technique. Piston core data and surface sediment 3.5 kHz reflectivity patterns help constrain thermal conductivity obtained from over 100 in situ stations. Plio-Quaternary stratigraphy is revised using new seismic reflection profiles: a boundary fault system associated with postrift margin uplift, a Pleistocene-age Var Fan construction, and recent diapirism of Messinian salt are indicated. After assessing local thermal disturbances (mass-wasting, microtopography, and salt refraction), positive heat flow corrections are made for multi-lithologic sedimentation histories and glacial paleotemperatures. Using boundary-layer cooling models, equilibrium heat flow estimates support geologic evidence for Oligocene and early Miocene rifting. Heat flow maxima correlate well with two "oceanic" sub-basins, suggesting that the southeastern trough near Corsica is ~5 Myr younger, consistent with the southeastern progression of volcanism and back arc rifting in the Western Mediterranean. Tectonic subsidence-crustal thickness trends indicate lithospheric stretching, with heat flow supporting asymmetric sub-crustal lithospheric thinning during the conjugate margin formation.

Description: Submitted in partial fulfillment of the requirements for the degree of Master of Science at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution May 1988

Description: The age of the Nazca plate where it enters the Peru and northern Chile trenches varies from 30 Ma in the north to 45 Ma in the south as its dip beneath the South American continent steepens from 13° to 30°. If the elastic thickness Te of oceanic lithosphere depends only on its age, and therefore thermal state, we would expect that Te determined from fitting the flexure of the lithosphere over the outer rise as revealed in the depth and geoid anomalies would increase from the Peru Trench in the north to the northern Chile Trench further south. We find that just the opposite is true: the lithosphere appears stiffer outboard of the Peru Trench than it does further south, and the isotherm controlling the elastic/ductile transition must be 600°C or greater if the thermal structure of the plate is that predicted by the standard thermal plate model. Because the decrease in plate stiffness to the south is correlated with a decrease in the minimum radius of curvature of the flexed plate over the outer rise and outer trench wall, we interpret our result in terms of inelastic yielding of the oceanic lithosphere when bent to high strains. The fact that the more highly bent segment of subducting lithosphere also dips at a steeper angle at greater depth beneath the continent might suggest that the amount of inelastic weakening of the lithosphere could be predicted from seismic images of the down going slab, but we find little support for this correlation worldwide. The forces and moments controlling the shallow deformation of the plate seaward of the trench do not appear to be linked to upper mantle processes which impose the dip at greater depth. Finally, we consider the possibility that the elastic thickness of the lithosphere would be reduced for trenches that are highly arcuate in map view, again due to inelastic yielding. If such a relationship exists, the effect for oceanic lithosphere is much smaller than what is documented for continental plates where they underthrust highly arcuate fold belts.