W. Spakman - Academia.edu (original) (raw)

Papers by W. Spakman

Tectonophysics, 2018

Key in understanding the geodynamics governing subduction and orogeny is reconstructing the paleo... more Key in understanding the geodynamics governing subduction and orogeny is reconstructing the paleogeography of 'Greater India', the Indian plate lithosphere that subducted since Tethyan Himalayan continental collision with Asia. Here, we discuss this reconstruction from paleogeographic, kinematic, and geodynamic perspectives and isolate the evolution scenario that is consistent with all three. We follow recent constraints advocating ã 58 Ma initial collision and update a previous kinematic restoration of intra-Asian shortening with a recently proposed model that reconciles long-debated large and small estimates of Indochina extrusion. Our new reconstruction is tested against paleomagnetic data, and against seismic tomographic constraints on paleo-subduction zone locations. The resulting restoration shows~1000-1200 km of post-collisional intra-Asian shortening, leaving a 2600-3400 km wide Greater India. From a paleogeographic, sediment provenance perspective, Eocene sediments in the Lesser Himalaya and on undeformed India may be derived from Tibet, suggesting that all Greater Indian lithosphere was continental, but may alternatively be sourced from the contemporaneous western Indian orogen unrelated to India-Asia collision. A quantitative kinematic, paleomagnetic perspective prefers major Cretaceous extension and a 'Greater India Basin' opening within Greater India, but data uncertainty may speculatively allow for minimal extension. Finally, from a geodynamic perspective, assuming a fully continental Greater India would require that subduction rates close to 20 cm/yr was driven by a downgoing lithosphere-crust assemblage more buoyant than the mantle, which seems physically improbable. We conclude that the Greater India Basin scenario is the only sustainable one from all three perspectives. We infer that old pre-collisional lithosphere rapidly entered the lower mantle sustaining high subduction rates, whilst post-collisional continental and young Greater India basin lithosphere did not, inciting the rapid India-Asia convergence deceleration~8 Myr after collision. Subsequent absolute northward slab migration and overturning caused flat slab subduction, Tibetan shortening, arc migration and arc volume decrease.

During evolution of the South Sandwich subduction zone, which has consumed South American Plate o... more During evolution of the South Sandwich subduction zone, which has consumed South American Plate oceanic lithosphere, somehow continental crust of both the South American and Antarctic plates have become incorporated into its upper plate. Continental fragments of both plates are currently separated by small oceanic basins in the upper plate above the South Sandwich subduction zone, in the Scotia Sea region, but how fragments of both continents became incorporated in the same upper plate remains enigmatic. Here we present an updated kinematic reconstruction of the Scotia Sea region using the latest published marine magnetic anomaly constraints, and place this in a South America-Africa-Antarctica plate circuit in which we take intracontinental deformation into account. We show that a change in marine magnetic anomaly orientation in the Weddell Sea requires that previously inferred initiation of subduction of South American oceanic crust of the northern Weddell Sea below the eastern margin of South Orkney Islands continental crust, then still attached to the Antarctic Peninsula, already occurred around 80 Ma. Subsequently, between ~71-50 Ma, we propose that the trench propagated northwards into South America by delamination of South American lithosphere: this resulted in the transfer of delaminated South American continental crust to the overriding plate of the South Sandwich subduction zone. We show that continental delamination may have been facilitated by absolute southward motion of South America that was resisted by South Sandwich slab dragging. Pre-drift extension preceding the oceanic Scotia Sea basins led around 50 Ma to opening of the Drake Passage, preconditioning the southern ocean for the Antarctic Circumpolar Current. This 50 Ma extension was concurrent with a strong change in absolute plate motion of the South American Plate that changed from S to WNW, leading to upper plate retreat relative to the more or less mantle stationary South Sandwich Trench that did not partake in the absolute plate motion change. While subduction continued, this mantle-stationary trench setting lasted until ~30 Ma, after which rollback started to contribute to back-arc extension. We find that roll-back and upper plate retreat have contributed more or less equally to the total amount of ~2000 km of extension accommodated in the Scotia Sea basins. We highlight that viewing tectonic motions in a context of absolute plate motion is key for identifying slab motion (e.g., rollback, trench-parallel slab dragging) and consequently mantle-forcing of geological processes.

Realistic appraisal of paleoclimatic information obtained from a particular location requires acc... more Realistic appraisal of paleoclimatic information obtained from a particular location requires accurate knowledge of its paleolatitude defined relative to the Earth’s spin-axis. This is cru-cial to, among others, correctly assess the amount of solar energy received at a location at the moment of sediment deposition. The paleolatitude of an arbitrary location can in princi-ple be reconstructed from tectonic plate reconstructions that (1) restore the relative motions between plates based on (marine) magnetic anomalies, and (2) reconstruct all plates rela-tive to the spin axis using a paleomagnetic reference frame based on a global apparent polar wander path. Whereas many studies do employ high-quality relative plate reconstruc-tions, the necessity of using a paleomagnetic reference frame for climate studies rather than a mantle reference frame appears under-appreciated. In this paper, we briefly summa-rize the theory of plate tectonic reconstructions and their reference frames tailored...

Digital Seismology and Fine Modeling of the Lithosphere, 1989

For several years geophysicists have applied tomographic techniques to image velocity structures ... more For several years geophysicists have applied tomographic techniques to image velocity structures on a global or regional scale or very detailed structures in exploration seismics. In studies of large areas they often make use of the delay time data provided by the International Seismological Centre. Many results have been obtained using P delays. For instance, Spakman (1986) has used 500,000 ISC P delays to derive a tomographic model of the Upper Mantle structures beneath Central Europe, the Mediterranean and the Middle East where the African and Eurasian plates converge. Less effort has been paid to tomographic imaging of shear velocity structures using ISC S delays. A tomographic S velocity model in addition to P model could be of great importance for the interpretation of P anomalies in terms of mineral or temperature causes. With the use of S delays instead of P delays some extra problems are to be expected. Fewer S data are available from the ISC and S delays are subject to larger errors due to identification problems and inaccuracies in the readings of the exact onsets. The choice of the starting or reference model to be used for tomographic inversions should be considered carefully.

Tectonics, 2014

The western Mediterranean recorded subduction rollback, slab segmentation and separation. Here we... more The western Mediterranean recorded subduction rollback, slab segmentation and separation. Here we address the questions of what caused Oligocene rollback initiation, and how its subsequent evolution split up an originally coherent fore arc into circum-southwest Mediterranean segments. We kinematically reconstruct western Mediterranean geology from subduction initiation to present, using Atlantic plate reconstructions as boundary condition. We test possible reconstructions against remnants of subducted lithosphere imaged by seismic tomography. Transform motion between Africa and Iberia (including the Baleares) between~120 and 85 Ma was followed by up to 150 km convergence until 30 Ma. Subduction likely initiated along the transform fault that accommodated pre-85 Ma translation. By thẽ 30 Ma inception of rollback, up to 150 km of convergence had formed a small slab below the Baleares. Iberia was disconnected from Sardinia/Calabria through the North Balearic Transform Zone (NBTZ). Subduction below Sardinia/Calabria was slightly faster than below the Baleares, the difference being accommodated in the Pyrenees. A moving triple junction at the trench-NBTZ intersection formed a subduction transform edge propagator fault between the Baleares and Calabria slab segments. Calabria rolled back eastward, whereas the Baleares slab underwent radial (SW-S-SE) rollback. After Kabylides-Africa collision, the western slab segment retreated toward Gibraltar, here reconstructed as the maximum rollback end-member model, and a Kabylides slab detached from Africa. Opening of a slab window below the NBTZ allowed asthenospheric rise to the base of the fore arc creating high-temperature metamorphism. Western Mediterranean rollback commenced only after sufficient slab-pull was created from 100 to 150 km of slow, forced subduction before~30 Ma. VAN HINSBERGEN ET AL.

Solid Earth, 2014

The level set method allows for tracking material surfaces in 2-D and 3-D flow modeling and is we... more The level set method allows for tracking material surfaces in 2-D and 3-D flow modeling and is well suited for applications of multi-material flow modeling. The level set method utilizes smooth level set functions to define material interfaces, which makes the method stable and free of oscillations that are typically observed in case step-like functions parameterize interfaces. By design the level set function is a signed distance function and gives for each point in the domain the exact distance to the interface as well as on which side it is located. In this paper we present four benchmarks which show the validity, accuracy and simplicity of using the level set method for multi-material flow modeling. The benchmarks are simplified setups of dynamical geophysical processes such as the Rayleigh-Taylor instability, post-glacial rebound, subduction and slab detachment. We also demonstrate the benefit of using the level set method for modeling a free surface with the sticky air approach. Our results show that the level set method allows for accurate material flow modeling and that the combination with the sticky air approach works well in mimicking Earth's free surface. Since the level set method tracks material interfaces instead of materials themselves, it has the advantage that the location of these interfaces is accurately known and that it represents a viable alternative to the more commonly used tracer method.

Journal of Geophysical Research: Solid Earth, 2014

No consensus exists on the tectonic evolution of the western Mediterranean since~35 Ma. Three dis... more No consensus exists on the tectonic evolution of the western Mediterranean since~35 Ma. Three disparate tectonic evolution scenarios are identified, each portraying slab rollback as the driving mechanism but with rollback starting from strongly different subduction geometries. As a critical test for the validity of each tectonic scenario we employ thermomechanical modeling of the 3-D subduction evolution. From each tectonic scenario we configure an initial condition for numerical modeling that mimics the perceived subduction geometry at~35 Ma. We seek to optimize the fit between observed and predicted slab morphology by varying the nonlinear viscoplastic rheology for mantle, slab, and continental margins. From a wide range of experiments we conclude that a tectonic scenario that starts from NW dipping subduction confined to the Balearic margin at 35 Ma is successful in predicting present-day slab morphology. The other two scenarios (initial subduction from Gibraltar to the Baleares and initial subduction under the African margin) lead to mantle structure much different from what is tomographically imaged. The preferred model predicts slab rotation by more than 180°, east-west lithosphere tearing along the north African margin and a resulting steep east dipping slab under the Gibraltar Strait. The preferred subduction model also meets the first-order temporal constraints corresponding to Mid-Miocene (~16 Ma) thrusting of the Kabylides onto the African margin and nearly stalled subduction under the Rif-Gibraltar-Betic arc since the Tortonian (~8 Ma). Our modeling also provides constraints on the rheological properties of the mantle and slab, and of continental margins in the region.

The TRANSMED Atlas. The Mediterranean Region from Crust to Mantle, 2004

ABSTRACT During the Cenozoic, the Western Mediterranean region has experienced a complex subducti... more ABSTRACT During the Cenozoic, the Western Mediterranean region has experienced a complex subduction history which involved the destruction of the Late Triassic/Jurassic Ligurian ocean and the West Alpine-Tethys. Lithosphere remnants of this evolution have been detected in the upper mantle by seismic tomography imaging. However, no general consensus exists on the interpretation of these remnants/slabs in the context of Ligurian ocean and West Alpine-Tethys subduction. In this paper we search for subduction remnants of the entire Cenozoic evolution in the recent global tomography model of Bijwaard and Spakman (2000) and compare these tomography results and our interpretations with those obtained in previous studies. Next, we present an analysis of imaged mantle structure in the context of the tectonic evolution of the Western Mediterranean during the Cenozoic. Our analysis leads to the following main results:

Frontiers in Earth Sciences, 2009

ABSTRACT Particularly interesting stages in the evolution of subduction zones are the two main tr... more ABSTRACT Particularly interesting stages in the evolution of subduction zones are the two main transient stages: initiation and termination. In this contribution the focus is on the second of these: terminal stage subduction, often triggered by continental collision or arc-continent collision. The landlocked basin setting of the Mediterranean region, in particular the western-central Mediterranean, provides unique opportunities to study terminal stage subduction and its consequences We use seismic tomography results on lithosphere and upper mantle structure as a source of information on plate boundary structure, and concentrate on the lithospheric scale aspects. Combining this structural information with process-oriented numerical modelling studies and regional observations, we present a 3D model for convergent plate boundary evolution after collision, in which slab detachment and the formation of tear or STEP (Subduction-Transform-Edge-Propagator) faults are key elements. A STEP fault laterally decouples subducting lithosphere from non-subducting litho-sphere in a scissor type of fashion. It enhances the ability of a slab to retreat through the mantle fl ow around the edge of the subducted slab. In this way collision and back-arc extension may occur in close proximity. In our study area this specifically pertains to collision along the north African margin, STEP formation in easterly direction, CCW rotation of the southern Apennines slab and the opening of the Tyrrhenian Sea. Vertical tearing of subducted lithosphere may play an important role as well, but is probably not crucial. On the basis of the good agreement between the Mediterranean-based model and the evolution of the Tonga-Fiji region we expect that the model may shed light on other complex convergent plate boundary regions, as well. In summary: Upon continental (or arc-continent) collision, along-trench variations in lithospheric properties of the subducting lithosphere may lead to disruption and segmentation of the subduction system. Following slab detachment along limited segments of a convergent plate boundary, the development of STEP faults is expected. These faults contribute to an increase in arc curvature within plate boundary segments. This contributes to the sinuous geometry of long subduction systems such as in the western and southwest Pacifi c.

Eos, Transactions American Geophysical Union, 2004

The aim of this program is to monitor, analyze, and interpret the surface motions occurring in re... more The aim of this program is to monitor, analyze, and interpret the surface motions occurring in response to active crust-lithosphere dynamics of the southeast Carpathians in Romania (Figure 1), using the Global Positioning System (GPS). For this region, observations of surface kinematics constitute a new and independent data source. In combination with other infor mation, mainly obtained by geologic and geo physical studies, surface motions may help to unravel the intriguing tectonics of the region. Particularly SUBDUCT focuses on the dynamic processes of the Vrancea high-seismicity region. The present-day tectonic activity in the Vrancea region is best characterized by a small zone of intense shallow-to intermediate-depth seismicity that is often interpreted as reflecting the late stage of intra-continental collision. Other regions with a similar intra-continental setting are Bucaramanga (Colombia-Bolivia) and Hindu-Kush (Afghanistan).Although few in number, these peculiar regions may hold important clues to the late-stage development of active ocean-continent or continent-continent inter action in regions of relatively small spatial extent, when compared to the scale of global plate tectonics. Plate interactions on this scale may have occurred many times in the geologic past and elsewhere on Earth, but are perhaps not recognized as such in the geologic record. These tectonically complicated areas are not yet well understood. To shed light on the active geodynamic processes of the southeast Carpathians, the region is the subject of many complementary geological and geophysical studies.These range from classical potential-field investigations

The rather narrow Valais Ocean began to open near the Jurassic-Cretaceous boundary. Sea floor spr... more The rather narrow Valais Ocean began to open near the Jurassic-Cretaceous boundary. Sea floor spreading in this northerly branch of the Alpine Tethys was kinematically linked to the northward propagation of the Mid-Atlantic spreading axis and the separation of the Iberia-Briançonnais microcontinent from Europe. This entailed opening of an oceanic basin, that extended from the Bay of Biscay via the area of the future Pyrenees into the domain of the Valais Ocean to the north of the Briançonnais Terrane (Frisch 1979; Stampfli 1993). In the Eastern Alps, however, this opening must have taken place within an already existing oceanic realm, representing the eastern continuation of the Piemont-Liguria basin. Tectonic units attributed to the Valais Ocean in the Eastern Alps are derived from areas where sedimentation persisted into the Tertiary, as documented for units in the core of the Engadine window and the Rhenodanubian flysch (Oberhauser 1995), but only suspected for the "Obere Schieferhülle" of the Tauern window. Remnants of this northern branch of the Alpine Tethys define a northern Alpine suture between the European margin and the continental Briançonnais Terrane (in case of the Western Alps), or an orogenic lid consisting of previously stacked Piemont-Liguria and Apulian (Austroalpine) units (in case of the Eastern Alps), respectively. This Valais Ocean closed during the Middle to Late Eocene. In respect to high-pressure units derived from the internal Briançonnais and Piemont-Liguria units, the Valais suture, together with the most distal parts of the European margin, defines a second and more external high-pressure belt, which extends from the Western Alps all the way into the Tauern window (Bousquet et al. 2002). Eclogitic mafic rocks are found in the Versoyen of the Western Alps and in parts of the Tauern window, while blueschists and other low temperature-high pressure rocks are preserved in the Engadine window (Bousquet et al. 1998). Note that the units derived from the Valais Ocean are separated from those derived from the Piemont-Liguria Ocean by nappes derived from the Briançonnais Terrane (see below) in case of transects IV and V (see also Figs. 2a-c) across the Western Alps. This is, however, not the case in the Eastern Alps (transect VI and profiles depicted in Fig. 2d,e) where remnants of the intervening Briançonnais terrane are not present (see also Fig. 1).

ABSTRACT Iron bearing minerals in the Earths lower mantle show a transition from high-spin to low... more ABSTRACT Iron bearing minerals in the Earths lower mantle show a transition from high-spin to low-spin in the iron constituent. This has been observed in particular for ferropericlase both experimentally (Fei et al, 2007, Lin et al. 2005) and in first principles calculations (Wu et al, 2009). The situation is less unambiguous for perovskite. Umemoto et al (2010) showed that the effect on volume is small compared to experimental uncertainty. Therefore we only considered the spin effects in ferropericlase in our models. The spin transition is characterized by a high valued positive Clapeyron slope ? = 19MPa-K while the smoothness of the transition increases with temperature. Fei et al. (2007) showed that at room temperature the spin transition pressure for iron richer composition occurs at higher values, e.g 40 GPa at 20 mol% FeO, 60 GPa at 40 mol% FeO. In order to get a full thermodynamic description of mantle material that includes the effects of spin transitions in ferropericlase we developed a model based on the multi-Einstein vibrational model approach of Jacobs et al. (2013). This model represents volume-pressure data of Lin et al. (2005), spin fraction data predicted by Wu et al. (2009) and it also includes the observed composition dependence of the spin transition pressure. Our new model further includes the thermodynamic description of Jacobs and de Jong (2007) that has been extended to describe thermodynamic properties of iron bearing (Mg,Fe)SiO3 perovskite. Because the spin transition pressure is composition dependent, the spin transition results in the formation of miscibility gap regions separating compositions enriched in high spin and compositions enriched in low-spin state. The spin transition affects thermodynamic properties, density, thermal expansivity, bulk modulus and heat capacity which in turn impact the convection dynamics of the Earth mantle. For instance, due to the high positive Clapeyron-slope of the transition convective mixing becomes more vigorous as observed in Boussinesq type modelling results of Bower et al, 2009, Shanas et al, 2011. Negative buoyancy of lithospheric slabs in the deep mantle is enhanced by the increase of thermal expansivity induced by the spin transition. Therefore the sinking rate of slabs are affected by the presence of the spin transition. Therefore the effects of the transition must be included in mantle convection modelling, done in order to bracket mantle viscosity values (Cizkova et al., 2012). Here we investigate the impact of the iron spin transition on the convective dynamics of the mantle and the distribution of material properties. As the spin transition related variations of material properties (e.g. thermal expansivity) are significant especially at lower temperatures, we concentrate mainly on the consequences for slab dynamics. To this end we use a compressible convection model based on a self consistent formulation of the thermo-physical material properties density, thermal expansivity and specific heat at constant pressure as described in (Jacobs and van den Berg, 2011). Finally, we evaluate the consequences of spin induced density contrasts in cold downwellings for the interpretation of the geoid. Bower et al. (2009) Geophys Res Lett, 36, L10306 Cizkova et al. (2012) Phys Earth Planet Inter 200, 56-62 Fei et al. (2007) Geophy res Lett, 34, L17307, 1-5 Jacobs and de Jong (2007) Geochim Cosmochim Acta, 71, 3630-3655 Jacobs and van den Berg (2011) Phys Earth Planet Inter, 186, 36-48 Jacobs et al. (2013) Phys Chem Minerals, in press Lin et al. (2005) Nature 436, 377-380 Shahnas et al (2011) J Geophys Res 116, B08205, 1-16 Umemoto et al (2010) Phys Earth Planet Int, 180, 209-214 Wu et al (2009) Phys Rev B 80, 014409, 1-8

ABSTRACT Being able to accurately model the free surface of the Earth is important for nearly all... more ABSTRACT Being able to accurately model the free surface of the Earth is important for nearly all geodynamical problems. Importantly, free surface deformation is coupled to vertical motions and therefore allows for selfconsistent modelling of topography build up. Such modelling allows for additional prediction from numerical models which can be compared to real Earth observations. When using an Eulerian framework, modelling the free surface is not straightforward and the so called "sticky air" approach is often used in which the "air" is modelled as a zero density fluid which has sufficient thickness (~100km) and a low viscosity compared to the underlying crust-mantle system (~5 orders of magnitude less, e.g. Crameri et al 2012). Tracers are commonly used to account for the tracking of all materials and interfaces. We here propose to use the level set method to track the interface between the crust and the air as a simulation of the free surface. The level set method represents a n-dimensional interface by a (n+1)-dimensional function chosen to be zero at the interface and it is mathematically described as a smooth (signed-distance) function. The target interface coincides with the zero-level set and its location can be traced through time by solving the advection equation for the level set function and subsequently locating the zero-level set every time-step. The level set method has several advantages over tracers, namely 1) with the level set method the exact location of the interface is known, 2) the exact distance of every point in the domain to the interface is known and 3) it is computationally less expensive compared to using tracers (particularly in 3-D applications). We will show results of benchmarks and 3-D numerical subduction models which contain multiple level sets representing the free surface and selected internal surfaces. This research is funded by The Netherlands Research Centre for Integrated Solid Earth Science (ISES).