Structural evidence for lithospheric foundering in the Puna Plateau, NW Argentina
Geophysical and magmatic data have been used to argue for loss of the lithospheric root of the Puna Plateau of NW Argentina at around 7 Ma. Models for lithospheric loss through delamination or the formation of Rayleigh-Taylor (RT) instabilities generally predict crustal thickening and shortening followed by extension, but with the orientation and timing of deformation dependent on the geometry of the delaminating or dripping region. Surficial structural observations are therefore suited to testing whether, how and when lithosphere was removed in the Puna. As deformation generally propagated west to each across the region in Eocene to Oligicoene time, younger, out-of-sequence shortening may reflect a phase of crustal thickening related to lithospheric detachment. Mapping and U-Pb geochronology from the southern Puna (Pasto Ventura) demonstrate syn-depositional shortening until at late as 8 Ma, followed by a rapid transition to extension, consistent with models for formation of an RT instability. Out-of-sequence deformation in other regions (such as the Antofalla basin) could reflect this as well. Miocene to Recent normal faulting has been documented across the southern Puna plateau, and basaltic volcanism indicates horizontal extension by ~7 Ma. Our mapping and a regional compilation indicate that (1) extension occurs throughout the Puna-Altiplano plateau, but is more extensive and the extension direction is more variable in the southern Puna; (2) extension appears to have initiated in the center of the southern Puna plateau, propagating outward with time, reaching the margins of the plateau in some cases by no earlier than 3.5 Ma. These observations support the formation of a radially symmetric RT instability. However, new igneous geochemistry, stable isotope data on pedogenic carbonates and an offset between the region of extension and the region of geophysically imaged thinned-lithosphere point to a complex scenario. Therefore, additional structural observations, geochronology and modeling will be necessary to fully explore the timing and magnitude of potential lithospheric loss and its effects on surficial deformation in the Puna Plateau.
Role of Lithospheric Delamination in the Uplift of the Central Andean Puna-Altiplano Plateau
Delamination of dense lower crust and mantle lithosphere has become a common mechanism to explain late Neogene surface uplift of the Central Andean Puna-Altiplano plateau as well as the mafic to shoshonitic lava flows and giant ignimbrite eruptions in the region. Seismic evidence for delamination has been suggested from tomographic images showing large low velocity anomalies in the mantle wedge under the northern Puna (e.g., Schurr et al., 2006, Tectonophysics). A currently operating 72 station passive seismic array will soon provide images of the upper mantle and lower crust under the southern Puna. A scenario to explain the spatial and temporal evolution of magmatic and structural events from the central Altiplano to the southern Puna is for delamination to occur as the subduction zone steepens after the Juan Fernandez Ridge on the subducting Nazca plate has passed southward. Evidence for shallowing of the subduction zone comes from patterns of magmatism and contractional deformation. Decompression melting of the mantle below a hydrated lower lithosphere and basal crust can produce mafic lavas that ascend into the lower crust where melting occurs producing hybrid magmas. Upward migration of these magmas leaves a gravitationally unstable dense garnet-bearing residue that can delaminate, which in turn generates more mantle melting leading to further eruptions until the lower crust becomes infertile. From chemical mass balance, the large Puna ignimbrites are considered to be near 50:50 mixtures of melted crust and new magma addition from the mantle. Seismic evidence for an irregular Moho boundary supports recent delamination. Crustal shortening and a hot mid-crust contribute to crustal flow which is an additional factor in plateau uplift. The relative contribution of the different uplift mechanisms depends on relative steepening and shallowing of the subducting Nazca plate and the pre- existing state of the continental lithosphere.
Lithospheric Delamination in an Orogenic Setting: An Example From the Eastern Anatolia Collision Zone
Paleoaltimetry Constraints on the Removal of Mantle Lithosphere from Beneath Tibet
Plate kinematic reconstructions of the convergence of India and Asia imply approximately 3500 km of convergence since the onset of collision at about 50 Ma at a longitude of 90°E. About 75% of this convergence is accounted for by excess crustal mass in a crustal transect north across the Himalayas and Tibetan Plateau. The remaining 25% has been exported by erosion (∼10%) + extrusion (primarily through lower crustal flow to the east) (∼10%) + lithospheric escape (∼5%). 3500 km of surface shortening must be matched by mantle lithospheric convergence, except that the export terms are reduced to only removal by lithospheric escape. Two end-member solutions to the lithospheric mantle history are (1) shortening and thickening and (2) continuous advective removal. Shortening and thickening of the mantle lithosphere synchronous with crustal shortening and thickening is a favored model with the added component that at about 8 Ma, this mantle lithospheric root became gravitationally unstable and was convectively removed. Mass balance would predict removal of an approximately 250 km thick mantle lithospheric root, were it of the dimension of the Plateau at 8 to 10 Ma, that would have resulted in 1 to 2.5 km of surface uplift subsequent to its removal. Continuous advective removal of the mantle lithosphere predicts that crustal isostasy dominates the process and resulting in a history of surface elevations that would have been raised to mean plateau elevations but then remained essentially constant or decrease with time. Stable isotope based paleoaltimetry estimates derived from a number of different archival records from the Himalayas and southern and central Tibet are remarkably consistent and imply that surfaces elevations in this region have been as high as they are today for as old as reliable data have been collected, at least 20 m.y. in the Himalayas, ∼15 to 20 m.y. in southern Tibet, and at least 25 to 35 Ma and possibly older in central Tibet. The existing paleoaltimetry data are thus most consistent with models invoking essentially continuous advective removal of the mantle lithosphere, with no discernable elevation signal from discontinuous convective removal of thickened mantle lithosphere from beneath this region. This paleoaltimetry-based conclusion is supported by the absence of a recognizable fast seismic velocity anomaly in seismic tomography models commensurate in dimensions with the India-Asia convergence history in the upper mantle transition zone beneath Tibet.
Removal of continental backarc lithosphere by flow-induced gravitational instability
Many continental backarcs have thin (∼60 km) lithosphere for 100's of km behind the volcanic arc, even where there has been no extension. One mechanism to produce thin backarc lithosphere is through thinning of normal thickness lithosphere by gravitational instability. Mantle lithosphere is cooler and therefore denser than sublithospheric mantle, making it prone to removal if it is perturbed in a manner in which the gravitationally-driven growth rate of the perturbation exceeds the rate at which thermal diffusion acts to suppress lateral density variations. To examine the stability of backarc mantle lithosphere, we use thermal-mechanical models of subduction of a ∼70 Myr old oceanic plate beneath continental lithosphere with an initial thickness of 120 km and a thermal structure similar to average Phanerozoic continental lithosphere. As the oceanic plate descends into the mantle, subduction-induced mantle flow shears the base of the backarc lithosphere, producing lateral density perturbations. Owing to the non- Newtonian lithosphere rheology, shearing also reduces the effective viscosity of the lowermost lithosphere, enabling the perturbations to become gravitationally unstable. The rapidly-growing downwellings result in removal of lower backarc lithosphere on timescales of 5-10 Ma. Conductive heating and shearing of the remaining lithosphere lead to a second, more muted, phase of gravitational instability and thinning. Lithosphere instability is enhanced by higher subduction rates, weaker intrinsic rheology, higher compositional density, and hotter initial thermal structure, in good agreement with predictions of buoyancy stability analysis. As both rheology and density depend on lithosphere composition, significant thinning may be restricted to continental mantle lithosphere that is fertile and contains a small amount of water. To produce a final backarc lithosphere thickness comparable to that observed but without lithospheric contraction or extension, it is necessary to have a weak lithosphere rheology, an initial thermal structure hotter than average Phanerozoic continental geotherms, or a combination of the two. The region of thin lithosphere at the northern Cascadia backarc coincides with terranes that were accreted to the North American craton. The known fertile composition of the backarc lithosphere may allow it to be thinned, while the drier, more refractory craton lithosphere is resistant to thinning.
Pervasive Thermal and Rheological Discontinuities in Mantle Lithosphere: Effects on Mantle Flow and Intraplate Motions
The geodynamic implications of strong lateral thermal and rheological discontinuities in the upper mantle likely have a profound influence on the deformation patterns, heat flow evolution, and surface response of intraplate lithosphere. Many stable continental platforms are bordered by hot and weak regions that connect active plate margins to continental interiors. The related first-order discontinuities in the thermal and rheological structure of the upper mantle can introduce flow instabilities and possibly intraplate tectonic activity. This study uses numerical geodynamic experiments to test the coupled response of mantle flow and overlying lithosphere to stark lateral discontinuities in upper mantle structure. A continental lithosphere is considered that is thin, hot and weak to one side and stepping to a thick, cool and strong cratonic lithosphere. We observe that mantle flow starts at the thermal and rheological discontinuity, but the flow subsequently affects the transition and can erode the lithosphere base. This leads to thermal and rheological re-adjustments of the overlying lithosphere that cause surface vertical motions and heat-flux variations. This study is inspired by the SE Canadian Cordillera, where a hot upper mantle and thin lithosphere of the Cordillera stands in stark contrast with the continental lithosphere of the stable north American craton. The Cordillera is referred to as a hot orogen that has been documented from elevated surface heat-flow data and low seismic velocities. Furthermore, significant post-orogenic intraplate uplift and denudation (i.e., epiorogeny) have been inferred that may require a dynamic mantle flow contribution. We test the implications of various rheological structures (i.e., variant rheologies for the stagnant mantle lithosphere and convecting mantle) for the overall model response and also the surface heat flow and surface deflection signal. We show, based on the geodynamic models, how the response of the lithosphere--i.e., amplitude and wavelength of the surface motions--is controlled by the interplay of the sub-lithospheric mantle flow and the spatial and temporal variations in the rheological and thermal structure at the transition zone.
Uniform Hot Thin Lithosphere for the Cordillera and Most Other Backarcs
Some continental areas have been interpreted to have thin lithospheres as the result of local removal of the normal lower lithosphere. Although delamination and other removal processes may occur, we suggest caution in interpreting thin lithosphere in local regions as due to this process. We have found that almost all current and recent backarcs, including the North America Cordillera, have quite uniform high temperatures and thin lithospheres, about 60 km; this is not just for extensional regions such as the Basin and Range. Quite uniform thin hot backarc lithospheres are expressed regionally by high heat flow, low velocities in the upper mantle, thin Te, susceptibility to deformation and seismicity, and sporadic volcanism with high water content. If the conclusion of hot thin lithosphere backarcs is accepted, the regional problem becomes how backarc lithospheres become thin globally with the onset of subduction. This initial backarc lithosphere thinning process may involve a form of removal through delamination or gravitational instability. An important factor may be that subduction inputs substantial amounts of water into the backarcs, weakening the upper mantle. In order to argue the case for local lithosphere removal we need to show that the region is anomalous compared to the average characteristics of continental backarcs, which have high temperatures, thin ~60 km weak lithospheres that are readily deformed by tectonic forces, and sporadic wet volcanism.
The Influence of Surface Erosion-Deposition on the Evolution of the Mantle Lithosphere Collisional Root at South Island, New Zealand
There are various first- and second-order factors that determine the growth and stability of mantle lithosphere (sub-crustal lithosphere) continental orogenic roots. In this study, the relationship between how the surface processes, mostly erosion and deposition, influence the evolution of the deep lithosphere as continental plate collisions transpire is examined in the context of South Island, New Zealand. Here, the erosion/deposition rates are relatively high and collision is fairly young. Recent findings suggest that the influence of erosion may reach much deeper into the mantle lithosphere than previously been appreciated. We will consider the coupling between surface processes and the evolution of the crust and lower lithosphere as orogenesis proceeds. Specifically, the eroded material from the orogen is transported then deposited in the flanks of the island-in the Westland and Canterbury Basins. The impact of an erosion-transport-deposition algorithm in the model is evaluated using numerical geodynamic experiments. Moreover, sub-aerial erosion rates are traditionally used in the empirical erosional laws; however these laws are revised using submarine erosion rates within the off-shore environment.