Volcanic hotspot tracks featuring linear progressions in the age of volcanism are typical surface... more Volcanic hotspot tracks featuring linear progressions in the age of volcanism are typical surface expressions of plate tectonic movement on top of narrow plumes of hot material within Earth’s mantle [1]. Seismic imaging reveals that these plumes can be of deep origin [2]—probably rooted on thermochemical structures in the lower mantle [3–6]. Although palaeomagnetic and radiometric age data suggest that mantle flow can advect plume conduits laterally [7,8], the flow dynamics underlying the formation of the sharp bend occurring only in the Hawaiian–Emperor hotspot track in the Pacific Ocean remains enigmatic. Here we present palaeogeographically constrained numerical models of thermochemical convection and demonstrate that flow in the deep lower mantle under the north Pacific was anomalously vigorous between 100 million years ago and 50 million years ago as a consequence of long-lasting subduction systems, unlike those in the south Pacific. These models show a sharp bend in the Hawaiian–Emperor hotspot track arising from the interplay of plume tilt and the lateral advection of plume sources. The different trajectories of the Hawaiian and Louisville hotspot tracks arise from asymmetric deformation of thermochemical structures under the Pacific between 100 million years ago and 50 million years ago. This asymmetric deformation waned just before the Hawaiian– Emperor bend developed, owing to flow in the deepest lower mantle associated with slab descent in the north and south Pacific.
In global convection models constrained by plume motions and subduction history over the last 230... more In global convection models constrained by plume motions and subduction history over the last 230 Myr, plumes emerge preferentially from the edges of thermochemical structures that resemble present-day Large Low Shear Velocity Provinces (LLSVPs) beneath Africa and the Pacific Ocean. It has been argued that Large Igneous Provinces (LIPs) erupting since 200 Ma may originate from plumes that emerged from the edges of the LLSVPs and numerical models have been devised to validate this hypothesis. Although qualitative assessments that are broadly in agreement with this hypothesis have been derived from numerical models, a quantitative assessment has been lacking. We present a novel plume detection scheme and derive Monte Carlo-based statistical correlations of model plume eruption sites and reconstructed LIP eruption sites. We show that models with a chemically anomalous lower mantle are highly correlated to reconstructed LIP eruption sites, whereas the confidence level obtained for a model with purely thermal plumes falls just short of 95%. A network of embayments separated by steep ridges form in the deep lower mantle in models with a chemically anomalous lower mantle. Plumes become anchored to the peaks of the chemical ridges and the network of ridges acts as a floating anchor, adjusting to slab push forces through time. The network of ridges imposes a characteristic separation between conduits that can extend into the interior of the thermochemical structures. This may explain the observed clustering of reconstructed LIP eruption sites that mostly but not exclusively occur around the present-day LLSVPs. This article is protected by copyright. All rights reserved.
The use of analog sensor arrays is often assumed to provide signal-to-ambient-noise improvements ... more The use of analog sensor arrays is often assumed to provide signal-to-ambient-noise improvements proportional to the square root of the number of sensors being summed. We determined via numerical modeling and field experiments that the improvements sought were significantly hindered once the ambient noise exhibited coherence over the array being summed. As a first step, a numerical model was developed to explore the optimal sensor spacing based on the average correlation coefficient between sensors. Field experiments were then carried out to measure ambient noise using closely spaced geophones at several sites in Perth, Western Australia. We show that the measured noise at six test sites was strongly coherent over distances of up to 10 m, with the level of coherency being inversely proportional to frequency. The resulting optimum geophone spacing under our field conditions was determined to be 7.5 m. These results offered further encouragement to reassess our use of analog arrays and to consider recording the output from individual sensors. In addition to the benefits from postacquisition noise suppression, this approach will enable increased trace density without degradation of the signal caused by strong coherent noise sources over acquisition spreads.
The acceleration of an existing linear feature
detection algorithm for 2D images using GPUs is d... more The acceleration of an existing linear feature detection algorithm for 2D images using GPUs is discussed. The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.
Renewed interest in the thermal structure of the upper
crust has led to the development of a new... more Renewed interest in the thermal structure of the upper crust has led to the development of a new generation of tools capable of modelling the 3D heat conduction problem, incorporating the complex geology and physical properties of the crust. The numerical and resolution requirements of such models has necessitated a high- performance performance computing approach, utilising massively parallel machine architecture to obtain ~10m resolution of models over a basin-scale.
Here we demonstrate the application of the StGermain/Underworld geodynamic modelling framework to basin-scale geodynamic problems. The default code has been modified to incorporate importing of geometrically complex geological units, each with its own conductivity and heat production, and temperature- dependent thermal conductivity. Thermal models demonstrate the importance of incorporating 2 & 3D geometries, with large lateral variations in the subsurface temperature field and heat flow as a result of the heterogenous basin architecture.
We explore the effect of lithological resolution, and identify a critical level of detail required to adequately represent basin-scale temperature variations. The regional temperature field play a dominant role in determining tenement-level subsurface temperatures, and highlight the importance of understanding regional
Mantle plumes are proposed to explain a broad range of geological and
geodynamical phenomena, su... more Mantle plumes are proposed to explain a broad range of geological and
geodynamical phenomena, such as volcanic island chains showing an age pro-
gression. Numerical studies suggest that mantle plumes could originate deep
within the Earth, possibly at the core-mantle boundary, as a result of thermal
instabilities. Thermal buoyancy causes such plumes to rise through the man-
tle and reach the base of the lithosphere. Numerical models of mantle plumes
are routinely used to study their inception and subsequent ascent through the
mantle. Numerous 2D studies have been performed - however, the combined
effects of temperature dependent viscosity, thermal and chemical boundary
layers on plume generation and their subsequent ascent velocities are still
relatively poorly understood. Significant advances in computational capac-
ity now allow for systematic 3D studies to be undertaken at resolutions fine
enough to resolve highly convective plume features.
We first design simple 3D isoviscous regional plume models under the
Boussinesq approximation in CitcomS as a starting point, comparing results
against analytical solutions. We impose a hot sphere of radius 200 km with
an excess temperature of ~200 K - at the base of the lower mantle - to
initiate our plume models. We study the effects of mesh resolution on re-
solving fine structures of ascending plumes, along with associated return flow
as they reach the base of the lithosphere, and on that of the rate of conver-
gence to analytical solutions. Building onto these results we then investigate
the effects of thermal and chemical boundary layers, heat content and heat
distribution of the initial temperature anomaly and temperature-dependent
viscosity on the ascent velocity of plumes. In addition, we investigate the
predicted evolution of dynamic topography and surface heat flux. We will
use the results of our regional tests to devise appropriate parametrizations
to implement active upwellings in forward global mantle flow models with
compositionally distinct thermal boundary layers.
Volcanic hotspot tracks featuring linear progressions in the age of volcanism are typical surface... more Volcanic hotspot tracks featuring linear progressions in the age of volcanism are typical surface expressions of plate tectonic movement on top of narrow plumes of hot material within Earth’s mantle [1]. Seismic imaging reveals that these plumes can be of deep origin [2]—probably rooted on thermochemical structures in the lower mantle [3–6]. Although palaeomagnetic and radiometric age data suggest that mantle flow can advect plume conduits laterally [7,8], the flow dynamics underlying the formation of the sharp bend occurring only in the Hawaiian–Emperor hotspot track in the Pacific Ocean remains enigmatic. Here we present palaeogeographically constrained numerical models of thermochemical convection and demonstrate that flow in the deep lower mantle under the north Pacific was anomalously vigorous between 100 million years ago and 50 million years ago as a consequence of long-lasting subduction systems, unlike those in the south Pacific. These models show a sharp bend in the Hawaiian–Emperor hotspot track arising from the interplay of plume tilt and the lateral advection of plume sources. The different trajectories of the Hawaiian and Louisville hotspot tracks arise from asymmetric deformation of thermochemical structures under the Pacific between 100 million years ago and 50 million years ago. This asymmetric deformation waned just before the Hawaiian– Emperor bend developed, owing to flow in the deepest lower mantle associated with slab descent in the north and south Pacific.
In global convection models constrained by plume motions and subduction history over the last 230... more In global convection models constrained by plume motions and subduction history over the last 230 Myr, plumes emerge preferentially from the edges of thermochemical structures that resemble present-day Large Low Shear Velocity Provinces (LLSVPs) beneath Africa and the Pacific Ocean. It has been argued that Large Igneous Provinces (LIPs) erupting since 200 Ma may originate from plumes that emerged from the edges of the LLSVPs and numerical models have been devised to validate this hypothesis. Although qualitative assessments that are broadly in agreement with this hypothesis have been derived from numerical models, a quantitative assessment has been lacking. We present a novel plume detection scheme and derive Monte Carlo-based statistical correlations of model plume eruption sites and reconstructed LIP eruption sites. We show that models with a chemically anomalous lower mantle are highly correlated to reconstructed LIP eruption sites, whereas the confidence level obtained for a model with purely thermal plumes falls just short of 95%. A network of embayments separated by steep ridges form in the deep lower mantle in models with a chemically anomalous lower mantle. Plumes become anchored to the peaks of the chemical ridges and the network of ridges acts as a floating anchor, adjusting to slab push forces through time. The network of ridges imposes a characteristic separation between conduits that can extend into the interior of the thermochemical structures. This may explain the observed clustering of reconstructed LIP eruption sites that mostly but not exclusively occur around the present-day LLSVPs. This article is protected by copyright. All rights reserved.
The use of analog sensor arrays is often assumed to provide signal-to-ambient-noise improvements ... more The use of analog sensor arrays is often assumed to provide signal-to-ambient-noise improvements proportional to the square root of the number of sensors being summed. We determined via numerical modeling and field experiments that the improvements sought were significantly hindered once the ambient noise exhibited coherence over the array being summed. As a first step, a numerical model was developed to explore the optimal sensor spacing based on the average correlation coefficient between sensors. Field experiments were then carried out to measure ambient noise using closely spaced geophones at several sites in Perth, Western Australia. We show that the measured noise at six test sites was strongly coherent over distances of up to 10 m, with the level of coherency being inversely proportional to frequency. The resulting optimum geophone spacing under our field conditions was determined to be 7.5 m. These results offered further encouragement to reassess our use of analog arrays and to consider recording the output from individual sensors. In addition to the benefits from postacquisition noise suppression, this approach will enable increased trace density without degradation of the signal caused by strong coherent noise sources over acquisition spreads.
The acceleration of an existing linear feature
detection algorithm for 2D images using GPUs is d... more The acceleration of an existing linear feature detection algorithm for 2D images using GPUs is discussed. The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.
Renewed interest in the thermal structure of the upper
crust has led to the development of a new... more Renewed interest in the thermal structure of the upper crust has led to the development of a new generation of tools capable of modelling the 3D heat conduction problem, incorporating the complex geology and physical properties of the crust. The numerical and resolution requirements of such models has necessitated a high- performance performance computing approach, utilising massively parallel machine architecture to obtain ~10m resolution of models over a basin-scale.
Here we demonstrate the application of the StGermain/Underworld geodynamic modelling framework to basin-scale geodynamic problems. The default code has been modified to incorporate importing of geometrically complex geological units, each with its own conductivity and heat production, and temperature- dependent thermal conductivity. Thermal models demonstrate the importance of incorporating 2 & 3D geometries, with large lateral variations in the subsurface temperature field and heat flow as a result of the heterogenous basin architecture.
We explore the effect of lithological resolution, and identify a critical level of detail required to adequately represent basin-scale temperature variations. The regional temperature field play a dominant role in determining tenement-level subsurface temperatures, and highlight the importance of understanding regional
Mantle plumes are proposed to explain a broad range of geological and
geodynamical phenomena, su... more Mantle plumes are proposed to explain a broad range of geological and
geodynamical phenomena, such as volcanic island chains showing an age pro-
gression. Numerical studies suggest that mantle plumes could originate deep
within the Earth, possibly at the core-mantle boundary, as a result of thermal
instabilities. Thermal buoyancy causes such plumes to rise through the man-
tle and reach the base of the lithosphere. Numerical models of mantle plumes
are routinely used to study their inception and subsequent ascent through the
mantle. Numerous 2D studies have been performed - however, the combined
effects of temperature dependent viscosity, thermal and chemical boundary
layers on plume generation and their subsequent ascent velocities are still
relatively poorly understood. Significant advances in computational capac-
ity now allow for systematic 3D studies to be undertaken at resolutions fine
enough to resolve highly convective plume features.
We first design simple 3D isoviscous regional plume models under the
Boussinesq approximation in CitcomS as a starting point, comparing results
against analytical solutions. We impose a hot sphere of radius 200 km with
an excess temperature of ~200 K - at the base of the lower mantle - to
initiate our plume models. We study the effects of mesh resolution on re-
solving fine structures of ascending plumes, along with associated return flow
as they reach the base of the lithosphere, and on that of the rate of conver-
gence to analytical solutions. Building onto these results we then investigate
the effects of thermal and chemical boundary layers, heat content and heat
distribution of the initial temperature anomaly and temperature-dependent
viscosity on the ascent velocity of plumes. In addition, we investigate the
predicted evolution of dynamic topography and surface heat flux. We will
use the results of our regional tests to devise appropriate parametrizations
to implement active upwellings in forward global mantle flow models with
compositionally distinct thermal boundary layers.
Uploads
detection algorithm for 2D images using GPUs is discussed.
The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each
component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.
crust has led to the development of a new generation of
tools capable of modelling the 3D heat conduction
problem, incorporating the complex geology and physical
properties of the crust. The numerical and resolution
requirements of such models has necessitated a high-
performance performance computing approach, utilising
massively parallel machine architecture to obtain ~10m
resolution of models over a basin-scale.
Here we demonstrate the application of the
StGermain/Underworld geodynamic modelling
framework to basin-scale geodynamic problems. The
default code has been modified to incorporate importing
of geometrically complex geological units, each with its
own conductivity and heat production, and temperature-
dependent thermal conductivity. Thermal models
demonstrate the importance of incorporating 2 & 3D
geometries, with large lateral variations in the subsurface
temperature field and heat flow as a result of the
heterogenous basin architecture.
We explore the effect of lithological resolution, and
identify a critical level of detail required to adequately
represent basin-scale temperature variations. The
regional temperature field play a dominant role in
determining tenement-level subsurface temperatures, and
highlight the importance of understanding regional
temperature variations in geothermal exploration.
geodynamical phenomena, such as volcanic island chains showing an age pro-
gression. Numerical studies suggest that mantle plumes could originate deep
within the Earth, possibly at the core-mantle boundary, as a result of thermal
instabilities. Thermal buoyancy causes such plumes to rise through the man-
tle and reach the base of the lithosphere. Numerical models of mantle plumes
are routinely used to study their inception and subsequent ascent through the
mantle. Numerous 2D studies have been performed - however, the combined
effects of temperature dependent viscosity, thermal and chemical boundary
layers on plume generation and their subsequent ascent velocities are still
relatively poorly understood. Significant advances in computational capac-
ity now allow for systematic 3D studies to be undertaken at resolutions fine
enough to resolve highly convective plume features.
We first design simple 3D isoviscous regional plume models under the
Boussinesq approximation in CitcomS as a starting point, comparing results
against analytical solutions. We impose a hot sphere of radius 200 km with
an excess temperature of ~200 K - at the base of the lower mantle - to
initiate our plume models. We study the effects of mesh resolution on re-
solving fine structures of ascending plumes, along with associated return flow
as they reach the base of the lithosphere, and on that of the rate of conver-
gence to analytical solutions. Building onto these results we then investigate
the effects of thermal and chemical boundary layers, heat content and heat
distribution of the initial temperature anomaly and temperature-dependent
viscosity on the ascent velocity of plumes. In addition, we investigate the
predicted evolution of dynamic topography and surface heat flux. We will
use the results of our regional tests to devise appropriate parametrizations
to implement active upwellings in forward global mantle flow models with
compositionally distinct thermal boundary layers.
detection algorithm for 2D images using GPUs is discussed.
The two most time consuming components of this process are implemented on the GPU, namely, linear feature detection using dual-peak directional non-maximum suppression, and a gap filling process that joins disconnected feature masks to rectify false negatives. Multiple steps or image filters in each
component are combined into a single GPU kernel to minimise data transfers to off-chip GPU RAM, and issues relating to on-chip memory utilisation, caching, and memory coalescing are considered. The presented algorithm is useful for applications needing to analyse complex linear structures, and examples are given for dense neurite images from the biotech domain.
crust has led to the development of a new generation of
tools capable of modelling the 3D heat conduction
problem, incorporating the complex geology and physical
properties of the crust. The numerical and resolution
requirements of such models has necessitated a high-
performance performance computing approach, utilising
massively parallel machine architecture to obtain ~10m
resolution of models over a basin-scale.
Here we demonstrate the application of the
StGermain/Underworld geodynamic modelling
framework to basin-scale geodynamic problems. The
default code has been modified to incorporate importing
of geometrically complex geological units, each with its
own conductivity and heat production, and temperature-
dependent thermal conductivity. Thermal models
demonstrate the importance of incorporating 2 & 3D
geometries, with large lateral variations in the subsurface
temperature field and heat flow as a result of the
heterogenous basin architecture.
We explore the effect of lithological resolution, and
identify a critical level of detail required to adequately
represent basin-scale temperature variations. The
regional temperature field play a dominant role in
determining tenement-level subsurface temperatures, and
highlight the importance of understanding regional
temperature variations in geothermal exploration.
geodynamical phenomena, such as volcanic island chains showing an age pro-
gression. Numerical studies suggest that mantle plumes could originate deep
within the Earth, possibly at the core-mantle boundary, as a result of thermal
instabilities. Thermal buoyancy causes such plumes to rise through the man-
tle and reach the base of the lithosphere. Numerical models of mantle plumes
are routinely used to study their inception and subsequent ascent through the
mantle. Numerous 2D studies have been performed - however, the combined
effects of temperature dependent viscosity, thermal and chemical boundary
layers on plume generation and their subsequent ascent velocities are still
relatively poorly understood. Significant advances in computational capac-
ity now allow for systematic 3D studies to be undertaken at resolutions fine
enough to resolve highly convective plume features.
We first design simple 3D isoviscous regional plume models under the
Boussinesq approximation in CitcomS as a starting point, comparing results
against analytical solutions. We impose a hot sphere of radius 200 km with
an excess temperature of ~200 K - at the base of the lower mantle - to
initiate our plume models. We study the effects of mesh resolution on re-
solving fine structures of ascending plumes, along with associated return flow
as they reach the base of the lithosphere, and on that of the rate of conver-
gence to analytical solutions. Building onto these results we then investigate
the effects of thermal and chemical boundary layers, heat content and heat
distribution of the initial temperature anomaly and temperature-dependent
viscosity on the ascent velocity of plumes. In addition, we investigate the
predicted evolution of dynamic topography and surface heat flux. We will
use the results of our regional tests to devise appropriate parametrizations
to implement active upwellings in forward global mantle flow models with
compositionally distinct thermal boundary layers.