Marco Bonini

Analogue modelling of continental extension: a review focused on the relations between the patterns of deformation and the presence of magma

Abstract Continental extension may occur in two main different modes, narrow and wide rifting, which mainly differ in the width of the deformed region. A third mechanism, the core complex, has been considered either a distinct mode of extension or a local anomaly within wide rifts. In terms of causative processes, continental rifting may be explained by both active or passive mechanisms, which also differ in the volume of magmatic products and in the rheological properties and stratification of the extending lithosphere. Both numerical and analogue models have investigated the main parameters controlling the extension of a rheologically layered lithosphere. In particular, analogue models have highlighted that the style of deformation is mainly controlled by the competition between the total resistance of the lithosphere and the gravitational forces; this competition, in turn, is mainly controlled by boundary conditions, such as the applied strain rate and the rheological characteristics of the extending lithosphere. Magmatic bodies eventually present within the continental lithosphere may significantly affect the process of extension. Both the thermal and mechanical effects related to the presence of magma strongly weaken the lithosphere and localise strain; this effect may have important implications for the mode of continental extension. At a crustal scale, magmatic intrusions may affect significantly the local fault pattern also favouring the development of core complex structures. Results of analogue models, performed taking into account the presence of an initially underplated magma and reproducing various continental extensional settings, suggest a close interaction between deformation and magma emplacement during extension. Particularly, magmatic underplating influences deformation localising strain in correspondence to the low-viscosity body, while on the other hand, rift kinematics and associated deformation has a major control on the pattern of magma emplacement. In particular: (1) During orthogonal rifting, magma is passively squeezed from an axial position towards the footwall of the major boundary faults; emplacement occurs in a lateral position in correspondence to lower crust domes. This process accounts for the close association between magmatism and the development of core complex structures, as well as for the occurrence of off-axis volcanoes in continental rifts. (2) During oblique rifting, deformation causes magma to emplace within the main rift depression, giving rise to intrusions with oblique and en echelon patterns. In nature, these patterns are found in continental rifts and also in some oceanic ridges. (3) Polyphase first orthogonal–second oblique rifting models suggest lateral squeezing and off-axis emplacement in the first phase and oblique en echelon intrusions in the successive oblique rifting phase. This evolution matches the magmatic and tectonic history of the Main Ethiopian Rift. (4) Development of transfer zones between offset rift segments has a great influence on both magma migration and deformation. Particularly, magma accumulates in correspondence to the transfer zone, with a main flow pattern that is perpendicular to the extension direction. This pattern may explain the concentration of magmatism at transfer zones in continental rifts. Overall, analysis of centrifuge models and their comparison with nature suggest that deformation and magma emplacement in the continental crust are intimately related, and their interactions constitute a key factor in deciphering the evolution of both continental and oceanic rifts.

Crustal section based on CROP seismic data across the North Tyrrhenian–Northern Apennines–Adriatic Sea

Abstract Using deep seismic reflection data from the Italian lithospheric exploration project CROP in the Central Mediterranean region, a 400-km-long section, composed of three different profiles crossing the Northern Tyrrhenian Sea (CROP M-12A profile), the Northern Apennines (CROP-03) and the Adriatic Sea (CROP M-16) is reconstructed and discussed. New data allow us to outline a seismically consistent tectono-stratigraphic setting for the crust and upper mantle of the Northern Apennines thrust–belt system and its Adriatic foreland. Time–space analysis of the deformation of the investigated chain and identification of existing macrostratigraphic crustal intervals and tectonic units allow a reasonably controlled interpretation of the geodynamic evolution and of the main orogenic stages. Careful seismic reprocessing and application of advanced techniques to key zones of the explored area (such as the Tuscan Archipelago) were determinants in obtaining fundamental information for understanding of the complex lithospheric structures and their evolution. Profile interpretation supports that the Northern Apennine chain is dominated by a compressive thrust system. Crustal extension, assumed by some authors as the dominating tectonic process for the whole Tuscan Apennine area, represents a subordinate geodynamic event of the last stage (Tyrrhenian). In the Early Cretaceous–Late Jurassic, the paleogeographic framework consisted of the Europe and Adria plates separated by the Alpine Tethys Ocean. During the Late Cretaceous–Early Eocene, Adria–Europe convergence (eo-Alpine stage) and subduction beneath the Adria plate closed the Alpine Tethys Sea, with the Tethyan slab being clearly seismically imaged. The first Apenninic geodynamic stage occurred in the Late Oligocene–Early Miocene with the opening of the Balearic Basin, which generated a first “lithospheric root” of the Apenninic chain in the Tuscan Archipelago area. This root is represented by Adria-verging thrust faults that progressively flatten eastward. Upper parts of the west-verging eo-Alpine thrust blocks were truncated by the east-verging thrust faults of the Balearic stage. A deeper seismic reflector, attributed to the top of the asthenosphere, forms a mantle high below the Elba Island. From the Late Miocene to Present, the Corsica basin and western hinterland area were affected by extensional tectonics related to the Tyrrhenian opening, whereas compressional tectonics continued in the eastern hinterland and mostly on the eastward migrating foreland, with development of a second “lithospheric root” constituted by high-angle thrust faults. These faults give rise to a huge basement culmination below the main Apennines watershed. Impressive E-directed gravity-sliding of sedimentary blocks over their sloping basement occur, generating the Umbria–Marche shallow seismicity. Crustal shortening of the Apennines system amounts to 170 km, 14 km of which are due to the eo-Alpine stage, 71 km to the Balearic and 85 km to the Tyrrhenian one. In the frame of Africa–Europe convergence, the Tyrrhenian–Apennines tectonodynamics were mainly conditioned by the Mesozoic paleogeography.

Successive orthogonal and oblique extension episodes in a rift zone: Laboratory experiments with application to the Ethiopian Rift

Small-scale modeling was performed to examine the effects of the superposition of two successive extensional phases from orthogonal to oblique (type 1) and from oblique to orthogonal (type 2). In both the type 1 and type 2 models, faults produced during the first stage strongly control fault development during the second stage. In type 1 models, the oblique faults developed during the second oblique phase are confined within a first-phase graben, whereas in type 2 models the oblique faults, produced during the first phase, continue to develop during orthogonal extension and connect with each other to give sigmoidal fault blocks. Type 1 models are compared with the structural setting of the Ethiopian Rift; the evolution of the rift is related to a recent extensional event, whose principal direction of stretching trends at around 50° to preexisting major normal faults. Type 1 laboratory models are fairly comparable to the northern sector of the Ethiopian Rift, referred to here as MER. They account for both the development of the en echelon oblique faults of the Wonji Fault Belt and the sinistral shear gradient running parallel to the eastern border of the MER, which formed during an oblique rifting extension. The statistical analysis of the whole Ethiopian Rift fault pattern by reference to the experimental data allows the determination of a N100°–N110° mean direction of stretching.

Passive roof thrusting and forelandward fold propagation in scaled brittle‐ductile physical models of thrust wedges

A set of scaled analogue models was performed to investigate the role of a decollement layer in the structural style of fold-and-thrust belts. Silicone putty (SGM 36) has been employed to represent the ductile decollement, while frictional material (quartz sand) has been used to simulate the brittle behavior of roof and floor sequences, situated above and below the decollement, respectively. In order to explore a wide combination of strength profiles, models have been deformed under five distinct rates of compression (0.15, 0.3, 0.45, 0.6, and 1.5 cm h−1) employing four different thicknesses for the silicone layer (0.1, 0.2, 0.4, and 0.8 cm). Experimental results indicate that roof sequences exhibit two distinct styles of deformation: (1) passive roof duplex (PRD) geometry [e.g., Banks and Warburton, 1986] and (2) outward propagation of folding (OFP) along the decollement [e.g., Davis and Engelder, 1985]. In PRD geometry the roof sequence remains relatively “stationary” being underthrust by a wedge-shaped floor duplex, while in OFP, displacement along the floor thrusts is transferred horizontally along the decollement layer. Experiments suggest the occurrence of a genetic relationship between shear stresses (at the base of the roof sequence τb and within the ductile layer τd) and the development of PRD or OFP geometry. For a given strain rate the deformation pattern of roof sequences depends upon the τb/τd ratio as well as the absolute τb and τd values. The transition field between OFP and PRD styles has been approximated by curvilinear regression of transition-style experimental data points, with a minimum value of τd, varying between about 112 and 160 Pa, to enter the PRD field. Syntectonic sedimentation in the molasse basin is found to inhibit development of folding into the foreland but to promote PRD style. Experiments simulate many of the principal characteristics of fold-and-thrust belts developed above a basal evaporite decollement, as well as the development of passive roof duplexes at mountain fronts. Shear stress values inferred for many fold-and-thrust belts correlate well the deformation fields predicted by the present experimental study, allowing to propose a simple model describing the mechanical behavior of natural roof sequences underlain by a decollement layer.

Transition from continental break‐up to punctiform seafloor spreading: How fast, symmetric and magmatic

[1] We present a comparison between numerical and analogue models focusing on the role of inherited lithospheric structures in influencing the process of continental break-up. Our results highlight that the presence of pre-existing anisotropies localizes strain and favors continental break-up and formation of a new ocean. For a fixed strain rate, the pre-rift lithosphere configuration influences rift duration, melt production and width and symmetry of the continental margin pair. Model results show a mainly two-phase tectonic history from continental extension to oceanization. In the first phase extension affects contemporaneously the whole rift structure, while in the second phase asthenosphere upwelling occurs into punctiform regularly-spaced spots sequentially propagating in an extension-orthogonal direction. INDEX TERMS: 8109 Tectonophysics: Continental tectonics—extensional (0905); 8120 Tectonophysics: Dynamics of lithosphere and mantle—general; 8150 Tectonophysics: Plate boundary—general (3040). Citation: Corti, G., J. Van Wijk, M. Bonini, D. Sokoutis, S. Cloetingh, F. Innocenti, and P. Manetti, Transition from continental break-up to punctiform seafloor spreading: How fast, symmetric and magmatic, Geophys. Res. Lett., 30(12), 1604, doi:10.1029/2003GL017374, 2003.

Modelling hanging wall accommodation above rigid thrust ramps

Abstract Experimental models are used to study the role of material rheology in hanging wall accommodation above rigid flat–ramp–flat thrust footwalls. The deformation in the hanging wall was accomplished by forwards sliding along a rigid basal staircase trajectory with a variable ramp angle, α, ranging from 15° to 60°. We model different ramp angles to examine hanging wall accommodation styles above thrust ramps of overthrust faults (α ranging from 15° to 30°), as well as above pre-existing normal faults (α ranging from 45° to 60°). For the hanging walls we used stratified frictional (sand) and viscous (silicone putty) materials. In this paper we study three types of models. Type 1 models represent purely frictional hanging walls where accommodation above thrust ramps was by layer-parallel thickening and by generating a series of back thrusts. Type 2 and 3 models represent stratified frictional/viscous hanging walls. In these models, accommodation was by a complex association of reverse and normal faults, mainly controlled by the rheological anisotropy as well as by the ramp inclination angle α. In Type 2 models the silicone covered only the lower flat, while in Type 3 models it also covered the rigid ramp. For α≤30° in Type 2 models and α≤45° in Type 3 models, the viscous layer inhibited the development of back thrusts in the frictional hanging wall, instead the silicone thickened to develop a ‘ductile ramp’. For α-values higher than 30° in Type 2 models and α=45° in Type 3 models, back thrusts develop in response to the bulk compression. The experiments simulate many structures observed above natural thrust ramps with α≤30° and pre-existing normal faults with α≥45°. The models emphasise the importance of a basal ductile layer, which allows the hanging wall to step-up over the rigid ramp by building up its own ductile ramp. The models also emphasise that foreland-directed normal faulting can develop at a thrust front in the case that the vertical stress due to gravity exceeds the horizontal stress due to end-loading within a thrust wedge.

Plio-Quaternary volcanotectonic activity in the northern sector of the Main Ethiopian Rift: relationships with oblique rifting

Abstract Deformation and magmatism within the ∼90 km wide northern Ethiopian Rift system is concentrated along a narrow zone - the Wonji Fault Belt. Two key areas (the Nazret-Dera and Asela-Ziway areas), located along the eastern margin of the north-northeast to northeast trending Main Ethiopian Rift, have been investigated in order to reconstruct the recent tectonomagmatic evolution of the northern branch of the Main Ethiopian Rift. In these areas, Early Pleistocene volcanic products (Wonji Group) overlie Pliocene volcanic rocks (Eastern Margin Unit). Detailed stratigraphical reconstructions have revealed the presence of several tectonomagmatic units which can be correlated between the two study areas. The stratigraphical and petrological study of these units outlined (1) the bimodal composition (basalts-pantellerites) of the oldest and youngest units and the unimodal character (pantellerites) of the products erupted during the intervening period; (2) the mainly fissural origin of the ignimbrites and oldest basalts; and (3) a mafic/felsic volumetric ratio of 1:5. The geological data suggest that, around the Pliocene-Quaternary boundary, a change in the stress field occurred in this Main Ethiopian Rift sector, passing from a direction of extension roughly orthogonal to the rift shoulders, to oblique rifting related to an east-west trending extension. In this framework the change in the style of volcanism observed in the Nazret-Dera and Asela-Ziway areas can be related to the change of the stress field. A new geodynamic model is presented for the Late Pliocene to Recent evolution of this sector of the Main Ethiopian Rift. According to this model, a large volume of rhyolitic products was erupted during an oblique rifting phase, following a previous period of pure extension. The change in the tectonic regime favoured partial melting of the underplated basalts as a decrease in the pressure and an elevation of isotherms occurred.

Indentation of a continent with a built-in thickness change: experiment and nature

Orogens oblique to the direction of plate convergence are currently attributed to obliquity between the margins of one or both of the sutured continents to their direction convergence. We use a single analogue experiment and natural examples to illustrate a potential additional factor: variations in strength of the indented continent at a high angle to the convergence direction. The wavelengths of structures in laterally shortened lithosphere depend on the strength of the most competent layers. Lateral variations in crustal thickness must therefore lead to structures oblique to any applied lateral compression.An analogue experiment was performed to explore this phenomenon. A two-layer ‘indented continent’ was modelled by a brittle upper crust of sand above a lower crust of high-viscosity polymer floating on a single layer of low-viscosity syrup representing the mantle. The well-known strike-slip structures allowing lateral escape to distant weak boundaries were hindered by lateral boundaries in front of the indenter. This allowed us to focus on the effects of a thickness change built into the ‘indented continent’ along a zone parallel to the direction in which a vertical rigid wall advancing at a steady rate represented the indenter. Vertical escape led to an ‘orogenic belt’ oblique to the advancing wall; this obliquity influences subsequent lateral escape. Model scaling and interpretations are based on Extended Thin Sheet Approximation (ETSA) and standard theories of faulting.Four sectors of the Alpine–Himalayan orogen (Iran, Tunisia, the Eastern Alps and the Himalaya) are oblique to the continental convergence direction, and we point to thickness changes at high angles to the suture that may account for this geometry. As crustal thicknesses north of oblique sectors of the Himalayas are not yet known, we speculate on them.We infer from the main difference between our experiment and all our examples chosen from nature that vertical orogenic escape was oblique to our model suture but can be parallel to natural sutures.

Dynamics of magma emplacement in centrifuge models of continental extension with implications for flank volcanism

A set of experimental small-scale models of continental extension investigated the emplacement of orthogonal and oblique magma chambers, initially underplated at the base of the crust. The models are driven by a centrifugal body force which simulates the role of gravity in nature. The models represent crustal conditions analogous to relatively mature continental rifts and consider both symmetric and asymmetric extension. The experimental results suggest that magma emplacement is controlled by the interactions between tectonics and rheology of the crustal layers. In particular, extension is mainly accommodated by lateral flow and ductile doming in the viscous layer simulating the lower crust and by listric normal faults, in the overlying brittle layer, associated with the ductile domal uplift. Stretching of the continental crust induces a “reactive” migration of viscous layers into the footwall of major normal faults, where magma accumulation at the core of the domes takes place. This behavior suggests that widespread magmatism is expected to localize in the footwall of major normal faults, a situation that is often observed in core complex structures. Although the models presented here are intended to simulate magma emplacement during extension of a two-layer brittle-ductile system, lateral flow and migration of magma initially underplated at the base of the crust beneath narrow rifts (e.g., the Ethiopian rift) may also provide a similar mechanism and a possible explanation for the occurrence of important volcanoes on the plateaus flanking the rift zones. In the suggested model, such volcanoes are related to large basaltic magma reservoirs, located at the base of the crust, to accommodate the space vacated by lateral flow and thinning of the ductile crust. The similarity in structure of the models with natural examples of continental extension modes (such as rift systems and core complexes) may suggest a close similarity of dynamic processes.

Indenter growth in analogue models of Alpine‐type deformation

A series of analogue experiments were carried out to simulate continental convergence, as seen in a profile through the Central Alps. A rigid indenter, representing the Adriatic plate, was driven laterally into a sand pack representing the brittle upper crust of Europe, detached and thickening above its subducting ductile lower crust. The rigid indenter advanced at the same steady rate in each experiment, but the dip of its front face was steepened in 15° increments from 15° to 90°. Where the rigid indenter face dipped at 45° or less, a sand wedge rose and was bound by a series of forekinks that nucleated at the toe of the indenter. Where the face of the rigid indenter dipped 60° or more, the wedge was defined by a single forekink and one or more backkinks that nucleated from a point advancing in front of the indenter toe. We interpret these results as indicating that slices of the sand pack and rising wedge are transferred across kink bands to build an “effective” indenter with a frontal dip closer to that dictated by the changing shear strength of the sand pile, which thickens vertically as it shortens laterally. One of our models (with a rigid indenter dipping 75°) simulates most of the major structures shown in recent syntheses of surface geology and deep seismic data in the Central Alps, without the isostatic lithospheric depression. This model accounts for the late collisional stage (Oligocene to Present) complex strain and metamorphic histories in the core of the orogenic wedge, the rapid rise and extrusion of small pips of Alpine eclogites, and the current passivity of the Insubric Line. It also emphasizes that lateral extension along gently dipping “thrusts” (orogen-normal horizontal escape) is confined to the extruded portion of the rising wedge.