Search

The role of lithospheric heterogeneities, presence or absence of melt, local and regional stresses, and gravitational potential energy in strain localization in continental rifts remains debated. We use new seismic and geodetic data to identify the location and orientation of the modern Nubia‐Somalia plate boundary in the 300‐km‐wide zone between the southern Main Ethiopian Rift (MER) and Eastern Rift (ER) across the Mesozoic Anza rift in the Turkana Depression. This region exhibits lithospheric heterogeneity, 45 Ma‐Recent magmatism, and more than 1,500 m of base‐level elevation change, enabling the assessment of strain localization mechanisms. We relocate 1716 earthquakes using a new 1‐D velocity model. Using a new local magnitude scaling with station corrections, we find 1 ≤ ML ≤ 4.5, and a b‐value of 1.22 ± 0.06. We present 59 first motion and 3 full moment tensor inversions, and invert for opening directions. We use complementary geodetic displacement vectors and strain rates to describe the geodetic strain field. Our seismic and geodetic strain zones demonstrate that only a small part of the 300 km‐wide region is currently active; low elevation and high‐elevation regions are active, as are areas with and without Holocene magmatism. Variations in the active plate boundary's location, orientation and strain rate appear to correspond to lithospheric heterogeneities. In the MER‐ER linkage zone, a belt of seismically fast mantle lithosphere generally lacking Recent magmatism is coincident with diffuse crustal deformation, whereas seismically slow mantle lithosphere and Recent magmatism are characterized by localized crustal strain; lithospheric heterogeneity drives strain localization. Plain Language Summary: Continental rifting involves the stretching, thinning, and heating of plates. Where present, magma adds heat, fluid, and new rock material into the plate, altering its physical and chemical properties. Pre‐existing lithospheric thickness variations and crustal shear zones may also influence when and where rifting occurs. Rifting processes can be monitored by tracking the vertical and horizontal movement of plates, and by analyzing earthquake patterns. To address debates concerning the role of plate thinning, magmatism, and pre‐rift lithospheric structures, we analyze new seismic and geodetic data from the Turkana Depression, a 300‐km‐wide region linking the Main Ethiopian Rift to the north and the Eastern Rift to the south. Additionally, the region has a failed Mesozoic rift with pre‐existing crustal shear zones. We identify regions that are actively deforming, defining the active plate boundary. The breadth of the actively rifting zone decreases above mantle low‐velocity zones, thinned mantle lithosphere, and/or hotter melt‐infiltrated zones. We also find little to no evidence of reactivation of faults within the thin crust of a failed Mesozoic rift that is circumnavigated by the active rifting in the region. Our findings suggest that differences in the plate‐scale physical and chemical properties dictate where and how faulting and magmatism occur. Key Points: New 1‐D velocity model relocates 1716 earthquakes with local magnitude ranging from 1 to 4.5 and b‐value of 1.22 ± 0.06 in Turkana DepressionBrittle strain localizes in narrow areas overlying uppermost mantle low velocity zones and avoids previously thinned crust of Mesozoic riftWe redefine the modern Nubia‐Somalia plate boundary across the Turkana Depression; only a part of the 300‐km‐wide region is active [ABSTRACT FROM AUTHOR]

[i] We give an explanation for the polarity, localization, shape, size, and initiation of subduction zones on Earth. By considering a soft, thin, curved lithospheric cap with either elastic or viscous rheology supported by a thick, nearly incompressible mantle, we find two different characteristic subduction geometries arise depending on boundary conditions: (1) plate boundaries where subduction results primarily from the gravitational body force (free subduction) have characteristic plate lengths and form arc-shaped dimpled segments resulting from the competition between bending and stretching in edge buckling modes of thin spherical shells, and (2) subduction zones due to localized applied loads that push one slab of thin, positively buoyant lithosphere beneath an overriding plate (forced subduction) form localized straight segments, consistent with the deformation of indented spherical shells. Both types of subduction are nonlinear subcritical instabilities, so small perturbations in the mechanical properties of the lithosphere have pronounced effects on subduction initiation and evolution. Yet in both cases, geometric relationships determined by the shape of the Earth itself play the most critical role in controlling the basic morphology and characteristic length scales of subduction zones. Citation: Mahadevan, L., R. Bendick, and H. Liang (2010), Why subduction zones are curved, Tectonics, 29, TC6002, doi: 10.1029/2010TC002720. Received 8 April 2010; revised 29 July 2010; accepted 10 August 2

Observations of relative motion in a geodetic network in Ladakh, India, and across southern Tibet indicate slow shear on the Karakorum fault, rapid east-west extension across the whole of southern Tibet, and constant arc-normal convergence between India and southern Tibet along the Himalayan arc. Measurements of ten campaign-style and six permanent sites with global positioning system (GPS) precise geodesy provide these bounds on the style and rates of the large-scale deformation in the Tibet-Himalaya region. Divergence between sites at Leb, Ladakh, India, and Shiquanhe, western Tibet, as well as slow relative motion among sites within the Ladakh network, limit right-lateral slip parallel to the Karakorum fault to only 3.4 [+ or -] 5 mm/yr. This low rate concurs with a recent estimate of 3-4 mm/yr for Late Holocene time, but disagrees with the much higher rate of 30-35 mm/yr that has been used to argue for plate-like behavior of the Tibetan Plateau. Convergence between Ladakh and the Indian subcontinent at 18.8 [+ or -] 3 mm/yr at 224[degrees] [+ or -] 17[degrees] (16) differs little from estimates of convergence across the central segment of the Himalaya. Finally, lengthening of the baseline between Leh, Ladakh, and Lhasa (in southeastern Tibet) at 17.8 [+ or -] 1 mm/yr or between Leh and Bayi (farther to the southeast) at 18 [+ or -] 3 mm/yr, is consistent with an extrapolation of rates of east-west extension of the Tibetan Plateau based both on shorter GPS baselines (e.g., Lhasa-Simikot) and on diverging slip vectors of earthquakes in the Himalaya. We interpret these results to indicate that Tibet behaves more like a fluid than like a plate. Keywords: geodesy, geodynamics, Tibetan Plateau, Himalaya, Karakorum, continuum mechanics.

The Turkana Depression, a topographically subdued, broadly rifted zone between the elevated East African and Ethiopian plateaus, disrupts the N-S, fault-bounded rift basin morphology that characterizes most of the East African Rift. The unusual breadth of the Turkana Depression leaves unanswered questions about the initiation and evolution of rifting between the Main Ethiopian Rift (MER) and Eastern Rift. Hypotheses explaining the unusually broad, low-lying area include superposed Mesozoic and Cenozoic rifting and a lack of mantle lithospheric thinning and dynamic support. To address these issues, we have carried out the first body-wave tomographic study of the Depression's upper mantle. Seismically derived temperatures at 100 km depth exceed petrological estimates, suggesting the presence of mantle melt, although not as voluminous as the MER, contributes to velocity anomalies. A NW-SE-trending high wavespeed band in southern Ethiopia at <200 km depth is interpreted as refractory Proterozoic lithosphere which has likely influenced the localization of both Mesozoic and Cenozoic rifting. At <100 km depth below the central Depression, a single localized low wavespeed zone is lacking. Only in the northernmost Eastern Rift and southern Lake Turkana is there evidence for focused low wavespeeds resembling the MER, that bifurcate below the Depression and broaden approaching southern Ethiopia further north. These low wavespeeds may be attributed to melt-intruded mantle lithosphere or ponded asthenospheric material below lithospheric thin-spots induced by the region's multiple rifting phases. Low wavespeeds persist to the mantle transition zone suggesting the Depression may not lack mantle dynamic support in comparison to the two plateaus. [ABSTRACT FROM AUTHOR]

Hydrologic processes have been shown to influence seismic productivity in many regions around the world, especially on active plate boundaries. To examine the influence of hydrologic loading cycles on seismicity in intraplate regions, we investigate temporal patterns of seismic productivity in the northern Rocky Mountains of Montana and Idaho. Seasonal variations in seismicity are present, with enhanced productivity in December and January, and reduced productivity in June and July. Using Snowpack Telemetry and GPS data, we find that seismicity is temporally correlated with the highest hydrologic loading rates rather than peak load, consistent with rate and state models of fault behavior for faults in critically stressed domains. However, we cannot distinguish between high hydrologic stress rates and pore pressure increases at seismogenic depths lagging ∼6 months after peak snowmelt. Plain Language Summary: Earthquake catalogs from the Rocky Mountains of Montana and Idaho show increased seismicity occurring during December and January. Increased seismic activity is strongly correlated with loading rate from snowfall, though it could alternatively be linked to seasonal changes in regional groundwater hydrology. Key Points: Seismicity rates in Montana and Idaho, excluding Yellowstone, exhibit seasonal signals with most events occurring in winterSeismicity rate variations are most strongly correlated with maximum hydrologic loading rateIntraplate crust in the northern Rockies is critically stressed [ABSTRACT FROM AUTHOR]

Phase alignment (synchronization) is a generalized property of interacting oscillators. If such interactions apply to earthquakes, they should manifest as time‐dependent variations in earthquake productivity organized according to a characteristic elastic loading period. Defining this period as renewal interval, the time required to accumulate the elastic potential energy released in a rupture, gives a consistent scaling property that can be used to search for temporal organization. We test for the expected structure in earthquake productivity using three different statistical tools optimized for different temporal sensitivities: Schuster spectra for events with short renewal intervals (0–25 years), Fourier power spectra for events with short and intermediate renewal intervals (0–100 years), and topological data analysis (TDA) for events with long renewal intervals (>100 years). All three indicate that earthquakes are organized in time according to renewal interval. Accounting for such unsteady temporal organization may improve forecasting skill by providing time‐dependent event probabilities. Plain Language Summary: Earthquakes can be thought of as elastic objects that store potential energy as they are loaded over time by tectonics and then release that stored energy during rupture. This loading and unloading cycle defines a characteristic timescale for an earthquake, its renewal interval. Generalized systems of many objects of this type are known to have nonlinear interactions over many cycles of loading and unloading that lead to patterns of events in time, such that events with similar natural periods become synchronized. We test for these interactions in earthquakes by sorting the global earthquake catalog by renewal interval and then testing whether events with similar renewal intervals tend to occur closer together in time than would be expected for independent event occurrence. Key Points: Earthquakes cluster in time according to an event characteristic timescale, renewal intervalEmergent organization of earthquake timing without direct triggering is consistent with nonlinear interactions among earthquakesTime‐dependent probability related to such clustering could improve forecasting skill [ABSTRACT FROM AUTHOR]

Geodetic observations in the Turkana Depression of southern Ethiopia and northern Kenya constrain the kinematic relay of extension from a single rift in Ethiopia to parallel rifts in Kenya and Uganda. Global Position System stations in the region record approximately 4.7 mm/year of total eastward extension, consistent with the ITRF14 Euler pole for Nubia‐Somalia angular velocity. Extension is partitioned into high strain rates on localized structures and lower strain rates in areas of elevated topography, as across the Ethiopian Plateau. Where high topography is absent, extension is relayed between the Main Ethiopian Rift and the Eastern Rift across the Turkana Depression exclusively through localized extension on and immediately east of Lake Turkana (up to 0.2 microstrain/year across Lake Turkana). The observed scaling and location of active extension in the Turkana Depression are inconsistent with mechanical models predicting distributed stretching due to either inherited lithospheric weakness or reactivated structures oblique to the present‐day extension direction.Plain Language Summary: The continent of Africa is breaking up into multiple pieces. This divergence is accommodated through extension along the East African Rift System. The Turkana Depression, which lies on the border of Ethiopia and Kenya, is of interest due to its previous rifting episodes, low elevation compared to the surrounding rift system, and location linking more prominent rift valleys to the north and south. Global Position System observations of surface velocities show that extension in the Turkana Depression is confined to a narrow region, not distributed across a broad area. These results suggest the East African Rift System is accommodating the breakup of the African continent through a combination of distributed deformation in areas with high topography and localized extension across low elevation rift basins.Key Points: The Turkana Depression accommodates divergence through localized extensionAfrican extension is accommodated by a combination of high and low strain ratesTopographic gradients appear to influence the style of extension throughout the African rift system [ABSTRACT FROM AUTHOR]

Geodetically derived velocities from Central Asia show that Northern Afghanistan, the Tajik Pamir, and northwestern Pakistan all move northward with comparable large velocities toward Eurasia. Steep velocity gradients, hence high strain rates, occur only across the Main Pamir Fault zone and with lesser magnitude between the northernmost Hindu Kush and the south and southeast margins of the Tajik Depression. Localized shortening is not apparent on any active India‐Hindu Kush crustal boundary; hence, crustal convergence between India and Eurasia in Central Asia is absorbed primarily on the northern and western margins of the Pamir. This concentrated strain on the Pamir margins is consistent with one, geometrically complex, interface between subducting Asian lithosphere and the Pamir. That interface might curve westward such that the Hindu Kush seismic zone is a continuation of the Pamir seismic zone, or alternatively, Hindu Kush earthquakes might occur in convectively unstable mantle lithosphere mechanically detached from surface faults. Plain Language Summary: Using Global Positioning System (GPS) measurements of surface velocities, we find that much of the relative motion between India and Eurasia in Central Asia is accommodated on a single crustal boundary on the north side of the Pamir, wrapping around the eastern and southern margins of the Tajik Depression. Key Points: Crustal shortening is accommodated across the northern margins of the Pamir and Hindu Kush between NW Afghanistan and the Tajik DepressionThere is little geodetic evidence for localized crustal shortening on the south side of the Hindu Kush between Indian and Pamir crustIntermediate depth seismicity in the Hindu Kush cannot easily be linked to ongoing subduction of Indian lithosphere [ABSTRACT FROM AUTHOR]

GPS time series of vertical displacement include the elastic response of the Earth to a combination of regional and local loading signals arising from hydrologic mass transfer. The regional loading, controlled by seasonal, synoptic precipitation patterns, dominates the displacement of individual stations and is highly correlated among stations with separation distances from 10 to 300 km. The local loading, controlled by small‐scale precipitation and storage variability, has much shorter correlation lengths of <30 km. We develop a new method to separate the regional and local contributions using common mode analysis and show that GPS is capable of measuring the local hydrologic load changes at watershed scales of tens of kilometers. Using this methodology, GPS‐measured displacement provides an integrated measurement of hydrologic load at a spatial scale between the existing long‐wavelength resolution of the Gravity Recovery and Climate Experiment and point measurement resolution of a precipitation station. Thus, GPS time series record critical observations for monitoring integrated hydrologic budgets at scales useful for water management and assessment of the hydro‐ecological response to climate change. Plain Language Summary: Even though estimating water quantity is critical for human communities, there are few good methods for doing so, especially in areas with lots of topography. Snow measurements and weather stations are only point observations that do not always represent the surrounding area, and satellite observations do not resolve small mountain watersheds well. We leverage the fact that the surface of the Earth is displaced by the weight of water, and that this deflection can be measured with GPS, to more accurately estimate water loading and unloading. We do this in snowpack‐dominated areas of the northern Rockies and show that these methods complement the available tools. Key Points: Observations of vertical displacement with GPS can be separated into regional and local surface mass loading contributionsThe local loading signal is dominated by watershed scale hydrologic loadingGPS observations are capable of providing an independent measure of total water storage at the watershed scale [ABSTRACT FROM AUTHOR]

The Himalayan plate boundary, because it is entirely subaerial, is both the most dramatic and the most accessible to direct observation of all active convergent boundaries on Earth. The shape of this boundary can be described as a small circle of radius 1696 [+ or -] 55 km, centered at long 91.6 [degrees] [+ or -] 1.6 [degrees] E and lat 42.4 [degrees] [+ or -] 2.1 [degrees] N for the extent of the arc between long 77.2 [degrees] and 92.1 [degrees] E. The pole of this small circle is consistent whether seismicity, topography, or stress state is used to define the position of the tectonic boundary. The defined small circle also coincides with a peak in microseismicity, the maximum horizontal strain rate, and a peak in the vertical velocity field. This quantitative definition of a stable, curved tectonic boundary is a prerequisite to modeling the dynamics of curvature in convergent arcs and applying appropriate boundary conditions to other regional models. Keywords: convergence, arcs, tectonics, geomorphology, Himalaya.

Abstract: Spatial spectral analyses of topography and strain rate magnitude across the western United States reveal the presence of at least two separable spectral peaks, representative of characteristic length scales, one short (~10–50 km) and one long (~150–250 km). Less spectral power at intermediate wavelengths aside from that expected for a red, or power law, spectrum occurs in either data set. These results quantify previous qualitative observations that the topography of western North America contains both short and long characteristic wavelength features. Comparing the spectral results to simplified bounding solutions for elastic, viscous, and layered models under tension shows that multiple spectral peaks of comparable power can be reproduced only by mechanical models with more than one layer. Therefore, the simplest model of lithospheric architecture capable of generating the observed dominant characteristic length scales is an elastoplastic lid over a viscous layer. [ABSTRACT FROM AUTHOR]

Abstract: Surface velocities derived from Global Positioning System observations and Quaternary fault slip rates measured throughout an extended region of high topography in South Asia vary smoothly over thousands of kilometers and are broadly symmetrical, with components of both north‐south shortening and east‐west extension relative to stable Eurasia. The observed velocity field does not contain discontinuities or steep gradients attributable to along‐strike differences in collision architecture, despite the well‐documented presence of a lithospheric slab beneath the Pamir but not the Tibetan Plateau. We use a modified Akaike information criterion (AICc) to show that surface velocities do not efficiently constrain 3D rheology, geometry, or force balance. Therefore, although other geophysical and geological observations may indicate the presence of mechanical or dynamic heterogeneities within the Indian‐Asian collision, the surface Global Positioning System velocities contain little or no usable information about them. [ABSTRACT FROM AUTHOR]

Insofar as slip in an earthquake is related to the strain accumulated near a fault since a previous earthquake, and this process repeats many times, the earthquake cycle approximates an autonomous oscillator. Its asymmetric slow accumulation of strain and rapid release is quite unlike the harmonic motion of a pendulum and need not be time predictable, but still resembles a class of repeating systems known as integrate-and-fire oscillators, whose behavior has been shown to demonstrate a remarkable ability to synchronize to either external or self-organized forcing. Given sufficient time and even very weak physical coupling, the phases of sets of such oscillators, with similar though not necessarily identical period, approach each other. Topological and time series analyses presented here demonstrate that earthquakes worldwide show evidence of such synchronization. Though numerous studies demonstrate that the composite temporal distribution of major earthquakes in the instrumental record is indistinguishable from random, the additional consideration of event renewal interval serves to identify earthquake groupings suggestive of synchronization that are absent in synthetic catalogs. We envisage the weak forces responsible for clustering originate from lithospheric strain induced by seismicity itself, by finite strains over teleseismic distances, or by other sources of lithospheric loading such as Earth's variable rotation. For example, quasi-periodic maxima in rotational deceleration are accompanied by increased global seismicity at multidecadal intervals. [ABSTRACT FROM AUTHOR]

Earthquakes represent the highest risk in terms of potential loss of lives and economic damage for Central Asian countries. Knowledge of fault location and behavior is essential in calculating and mapping seismic hazard. Previous efforts in compiling fault information for Central Asia have generated a large amount of data that are published in limited-access journals with no digital maps publicly available, or are limited in their description of important fault parameters such as slip rates. This study builds on previous work by improving access to fault information through a webbased interactive map and an online database with search capabilities that allow users to organize data by different fields. The data presented in this compilation include fault location, its geographic, seismic and structural characteristics, short descriptions, narrative comments and references to peer-reviewed publications. The interactive map displays 1196 fault segments and 34 000 earthquake locations on a shaded-relief map. The online database contains attributes for 122 faults mentioned in the literature, with Quaternary and geodetic slip rates reported for 38 and 26 faults respectively, and earthquake history reported for 39 faults. This work has implications for seismic hazard studies in Central Asia as it summarizes important fault parameters, and can reduce earthquake risk by enhancing public access to information. It also allows scientists and hazard assessment teams to identify structures and regions where data gaps exist and future investigations are needed. [ABSTRACT FROM AUTHOR]

The variation of mechanical properties as a function of depth in the lithosphere controls the relationship between surface deformation and whole-lithosphere deformation. Where mechanical competence (elastic strength or viscous stiffness) is a continuous function with depth, surface deformation can be used to constrain either force balance or rheological parameters for the lithosphere. However, where competence is vertically discontinuous, the surface deformation and deformation at depth may have a range of relationships, including complete independence. Therefore, both inversions and forward models designed to test either rheology or stress state are highly sensitive to the a priori choice of a mechanical model. [ABSTRACT FROM AUTHOR]

The variation of mechanical properties with depth in the lithosphere determines the relationship between surface deformation and whole-lithosphere deformation, hence between surface deformation and whole-lithosphere dynamics. Where viscosity (or elastic strength) is a continuous function with depth, surface deformation can be used to constrain both force balance and rheological parameters. Where viscosity is discontinuous, but the upper crust and mantle lithosphere have comparable maximum values, surface deformation can be used to approximate force balance and rheological parameters, but tradeoffs mean that estimates of stress and viscosity are effective equivalent values rather than actual values. Where viscosity is both discontinuous and differs by much more than an order of magnitude between the upper crust and mantle lithosphere, information about both force balance and rheology are absent from the surface deformation, so surface observations alone are insufficient to estimate either the dynamic or mechanical state of the lithosphere. [ABSTRACT FROM AUTHOR]

Collision between an undeformable indenter and a viscous region generates isostatically compensated topography by solid-state flow. We model this process numerically, using a finite element scheme. The slope, amplitude and symmetry of the topographic signal depend on the indenter size and the Argand number of the viscous region, a dimensionless ratio of gravitational body forces to viscous forces. When applied to convergent continental settings, these scaling rules provide estimates of the position of an indenter at depth and the mechanical properties of the viscous region, especially effective viscosity. In Tibet, forward modelling suggests that some elevated, low relief topography within the northern plateau may be attributed to lower crustal flow, stimulated by a crustal indenter, possibly Indian lithosphere. The best-fit model constrains the northernmost limit of this indenter to 33.7°N and the maximum effective viscosity of Eurasian middle and lower crust to Pa s. [ABSTRACT FROM AUTHOR]

A viscous region with deformable boundaries is used to model simultaneous crustal flow and lithospheric coupling in northern Tibet. This model suggests that (1) the deformation of northern Tibet is different from that of southern Tibet because of structural differences between the regions; (2) the viscosity contrast between the crust and the mantle lithosphere is relatively small beneath northern Tibet; and (3) crustal flow is compatible with crust-mantle coupling under these conditions. [ABSTRACT FROM AUTHOR]

The Main Ethiopian Rift (MER) in central Ethiopia extended in the rift-normal direction at a mean rate of 4.0 ± 0.9 mm yr−1 (1σ) during the period 1992–2003, nearly a factor of two slower than the opening rate estimated from global plate motion inversions. Rift opening near a geodetic array during this period was accommodated by a single dike injection event in 1993, spatially coincident with active magmatic segments, probably triggered by observed seismicity. Following dike injection, the crust in the rift relaxed as a layered medium, with a ∼15-km-thick elastic lid over a viscous half space of 1018 Pa s. Diking, rather than normal faulting on rift-bounding faults, appears to be the predominant mechanism of extension in the MER, explaining the very low regional rates of moment release. The length scale and temporal behaviour of surface displacements require viscoelastic rheology in the rift. [ABSTRACT FROM AUTHOR]

The Turkana Depression separates the uplifted Ethiopian and East African Plateaus. It was the site, in Mesozoic times, of a failed episode of NE–SW-oriented rifting (the Anza Rift), but now hosts E–W-oriented Nubia–Somalia separation at the junction between the Main Ethiopian Rift in the north and the Eastern Rift to the south. However, the time-integrated effect of these rifting phases on crustal and lithospheric mantle architecture and thermal structure is poorly understood. Utilising data from new seismograph networks in the Turkana Depression and northern Tanzania Craton, we produce a detailed anisotropic crustal and uppermost mantle shear-wave velocity model of the region. Within the Tanzania Craton, slightly lower uppermost mantle wavespeeds (4.4–4.5 km/s) compared to neighbouring regions, and coincident rift-parallel crustal anisotropy, imply the Nyanza Rift developed in relatively weak mobile belt lithosphere between two refractory Archean blocks. At upper-crustal (≲10 km) depths in the Turkana Depression, the slowest velocities (≲3.2 km/s) are attributed to thick Mesozoic-age sedimentary basins. Nowhere within the Depression is the mid-to-lower crust or lithospheric mantle associated with wavespeeds as slow, or seismic anisotropy as strong, as that observed below the melt-rich central and northern Main Ethiopian Rift (MER) and Ethiopian Plateau further north. High upper mantle wavespeeds (≳4.5 km/s), coinciding with the broadening of MER-rifting into southern Ethiopia, confirm the presence of refractory Proterozoic lithosphere acting as a rheological boundary to rift development. Thinned crustal zones associated with failed Mesozoic Anza rifting are also underlain by fast wavespeed (>4.5 km/s) mantle lithosphere, implying this area has resisted significant thermomechanical modification from Miocene-Recent extension and magmatism. Pre-existing crustal thin zones do not, therefore, necessarily represent zones of plate-weakness where subsequent phases of rifting will develop. • Lithospheric shear wave model for the Turkana Depression and northern Tanzania Craton. • Lithospheric seismic signature of magmatism lacking below the Depression. • Magmatic extension below the Depression must be young and localised. • Failed Mesozoic rift zones not markedly modified by Cenozoic rifting and magmatism. [ABSTRACT FROM AUTHOR]

The Turkana Depression in Eastern Africa separates the elevated plateaus of East Africa to the south and Ethiopia-Yemen to the north. It remains unclear whether the Depression lacks dynamic mantle support, or if the entire East Africa region is dynamically supported and the Depression compensated isostatically by thinned crust. Also poorly understood is how Miocene-Recent extension has developed across the Depression, connecting spatially separated magmatic rift zones in Ethiopia and Kenya. Receiver function analysis is used to constrain Moho depth and bulk-crustal V P / V S ratio below new seismograph networks in the Depression, and on the northern Tanzania craton. Crustal thickness is ∼40 km below northern Uganda and 30–35 km below southern Ethiopia, but 20–30 km below most of the Depression, where mass-balance calculations reveal low elevations can be explained adequately by crustal thinning alone. Despite the fact that magmatism has occurred for 45 Ma across the Depression, more than 15 Ma before East African Rift (EAR) extension initiated, bulk crustal V P / V S across southern Ethiopia and the Turkana Depression (∼1.74) is similar to that observed in areas unaffected by Cenozoic rifting and magmatism. Evidence for voluminous lower crustal intrusions and/or melt, widespread below the Ethiopian rift and Ethiopian plateau to the north, is therefore lacking. These observations, when reviewed in light of high stretching factors (β ≤ 2.11), suggest Cenozoic extension has been dominated until recently by faulting and plate stretching, rather than magma intrusion, which is likely an incipient process, operating directly below seismically-active Lake Turkana. Early-stage EAR basins to the west of Lake Turkana, with associated stretching factors of β ≈ 2 , formed in crust only moderately thinned during earlier rifting episodes. Conversely, ∼23 km-thick crust beneath the Kino Sogo Fault Belt (KSFB) has small offset faults and thin sedimentary strata, suggesting almost all of the observed stretching occurred in Mesozoic times. Despite the KSFB marking the shortest path between focused extensional zones to the north and south, seismicity and GPS data show that modern extension is localized below Lake Turkana to the west. Failed Mesozoic rift zones, now characterized by thinned crust and relatively refractory mantle lithosphere, are being circumnavigated, not exploited by EAR rifting. • Receiver function constraints on Moho depth and V P / V S from the Turkana Depression. • Depression crust 20–30 km thick; bulk crustal V P / V S ratios generally low (≤1.75). • Thin crust explains Depression's low elevation, not a lack of dynamic mantle support. • Past and present magma-assisted rifting not well established below the Depression. • Failed Anza rift is circumnavigated, not exploited, by subsequent East African Rifting. [ABSTRACT FROM AUTHOR]