Description: The impact of remotely forced mantle flow on regional subduction evolution is largely unexplored. Here we investigate this by means of 3D thermo-mechanical numerical modeling using a regional modeling domain. We start with simplified models consisting of a 600 km (or 1400 km) wide subducting plate surrounded by other plates. Mantle inflow of ∼3 cm/yr is prescribed during 25 Myr of slab evolution on a subset of the domain boundaries while the other side boundaries are open. Our experiments show that the influence of imposed mantle flow on subduction evolution is the least for trench-perpendicular mantle inflow from either the back or front of the slab leading to 10–50 km changes in slab morphology and trench position while no strong slab dip changes were observed, as compared to a reference model with no imposed mantle inflow. In experiments with trench-oblique mantle inflow we notice larger effects of slab bending and slab translation of the order of 100–200 km. Lastly, we investigate how subduction in the western Mediterranean region is influenced by remotely excited mantle flow that is computed by back-advection of a temperature and density model scaled from a global seismic tomography model. After 35 Myr of subduction evolution we find 10–50 km changes in slab position and slab morphology and a slight change in overall slab tilt. Our study shows that remotely forced mantle flow leads to secondary effects on slab evolution as compared to slab buoyancy and plate motion. Still these secondary effects occur on scales, 10–50 km, typical for the large-scale deformation of the overlying crust and thus may still be of large importance for understanding geological evolution.

Description: One of the lowest geoid anomalies on Earth lies in the Indian Ocean just south of the Indian peninsula. Several theories have been proposed to explain this negative geoid anomaly, most of which invoke past subduction. Some recent studies have argued that high-velocity anomalies in the lower mantle coupled with low-velocity anomalies in the upper mantle are responsible for these geoid lows. However, there is no general agreement regarding the source of the geoid low in the Indian Ocean. We investigate the source of this anomaly by using instantaneous models of density-driven mantle convection. Our study is the first to successfully explain the presence of this anomaly using a global convection model driven by present-day density anomalies derived from seismic tomography. We test various tomography models in our flow calculations using different radial and lateral viscosity variations. Although quite a few of them produce a fairly high correlation to the observed geoid globally, only a few (SMEAN2, GyPSuM, SEMUCB, and LLNL-JPS) could match the exact location and pattern of the geoid low in the Indian Ocean. The source of this low is a low-density anomaly stretching from a depth of 300 km down to ∼900 km in the northern Indian Ocean region. This density anomaly potentially originates from plume material rising along the edge of the African Large Low Shear Velocity Province, which moves toward the northeast, along with the movement of the Indian plate in the same direction.

Description: Glacial-isotactic adjustment (GIA) is one of the key processes considering relative sea-level (RSL) and paleo-topography during the last glacial cycle. Especially in former ice-covered regions the subsidence of the solid Earth due to ice loads can reach more than 500 m and contributes to the stability of ice-sheets (e.g. position of grounding line and ice-sheet elevation), whereas at the coasts of the world oceans the deformation is governed by global RSL fall of more than 100 m. Because the viscoelastic response of the solid Earth is governed by its viscosity structure, the effect of lateral viscosity variations on deformations due to GIA has to be estimated. The importance was already shown for the differences in earth structure below the glacial ice sheets of Fennoscandia and Laurentide, as well as for a number of peripheral and far-field regions. One open question arises: Can the 3D earth properly be parameterized by locally optimized 1D earth structures? In this study, we apply a 3D Earth structure which we derived from seismic tomography and further geodynamic constraints as an a priori estimation of the Earth viscosity distribution. Applying a standard glaciation history, we compare the response characteristics of 1D and 3D earth parameterizations and discuss the limits of optimized 1D earth parametrizations. We will focus on reconstructions of RSL during the last deglaciation in view of sea level index points which are generally used for validating the GIA process.

Description: Mantle plumes upwelling beneath moving tectonic plates generate age-progressive chains of volcanos (hotspot chains) used to reconstruct plate motion. However, these hotspots appear to move relative to each other, implying that plumes are not laterally fixed. The lack of age constraints on long-lived, coeval hotspot chains hinders attempts to reconstruct plate motion and quantify relative plume motions. Here we provide 40Ar/39Ar ages for a newly identified long-lived mantle plume, which formed the Rurutu hotspot chain. By comparing the inter-hotspot distances between three Pacific hotspots, we show that Hawaii is unique in its strong, rapid southward motion from 60 to 50 Myrs ago, consistent with paleomagnetic observations. Conversely, the Rurutu and Louisville chains show little motion. Current geodynamic plume motion models can reproduce the first-order motions for these plumes, but only when each plume is rooted in the lowermost mantle.

Description: For at least 120 Myr, the Kerguelen plume has distributed enormous amounts of magmatic rocks over various igneous provinces between India, Australia, and Antarctica. Previous attempts to reconstruct the complex history of this plume have revealed several characteristics that are inconsistent with properties typically associated with plumes. To explore the geodynamic behavior of the Kerguelen hotspot, and in particular address these inconsistencies, we set up a regional viscous flow model with the mantle convection code ASPECT. Our model features complex time-dependent boundary conditions in order to explicitly simulate the surrounding conditions of the Kerguelen plume. We show that a constant plume influx can result in a variable magma production rate if the plume interacts with nearby spreading ridges and that a dismembered plume, multiple plumes, or solitary waves in the plume conduit are not required to explain the fluctuating magma output and other unusual characteristics attributed to the Kerguelen hotspot.

Description: Mantle plumes are hot upwellings of rock thought to originate at the core‐mantle boundary. As they rise through the mantle, their conduits may become tilted due to lateral large‐scale mantle flow. Recent tomographic images have revealed a strongly tilted plume conduit starting at the core‐mantle boundary beneath northern Baja California rising toward the Yellowstone hot spot from the southwest. Here we perform numerical computations of plumes deflected in large‐scale mantle flow with the aim of finding if realistic model parameter ranges exist that yield a good fit with the tomographically observed conduit. We restrict ourselves to models that yield reasonable results for plume conduit tilt and hot spot motion globally. These models require high viscosity ≈1023 Pa s at some lower mantle depths. For a plume head reaching the surface 17 Ma, corresponding to the start of the Columbia River Basalts, our models require rise times ≈80 Myr or longer to match the tilt of the conduit observed by tomography. We used several tomography models to determine mantle density with almost all models predicting southwestward flow in the lowermost mantle beneath the western United States. Exact details of the shape of the predicted conduit's southwesterly tilt vary, depending on the density and viscosity structures we used. In many cases we find comparatively strong tilts in two depth ranges, in the upper and lower portions of the mantle, which is also a characteristic of the tomographically observed conduit. We expect that future models may help to constrain large‐scale flow by matching these corresponding depth ranges.

Description: Mantle plumes are likely initiated by large plume heads. Because mantle viscosity and its dependence on temperature and stress is poorly known, it is uncertain how long plumes take to rise through the mantle. Knowing this is important for constraining the total time until subducted material may resurface in mantle plume heads. Here we apply observational constraints to narrow down plume rise times. Firstly, the margins of Large Low Shear Velocity Provinces (LLSVPs) which probably represent piles of hot chemically dense material in the lowermost mantle are likely locations for plume generation. Here we model, as a function of rise time, the locations where plume heads start rising, such that plumes reach the surface at the observed hotspots. We find that these source locations agree well with LLSVP margins only for rise times of about 30 Myr or less. Different from other hotspots with likely deep source, Yellowstone is close to subduction zones and far from LLSVPs. Yet a recent tomography model shows a tilted plume conduit beneath, rising from the lowermost mantle. Here we compare modelled plume conduit shape, as a function of rise time, with tomography and find the best agreement for rise times of about 90 Myr or more. Comparatively slow rising could be due to a small plume head (corresponding to Columbia River Basalts being smaller than other Large Igneous Provinces). Faster rising of plumes near LLSVPs could also be due to hotter mantle, causing upward ambient mantle flow and reduced viscosity.