-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Plate bending earthquakes and the strength distribution of the lithosphere #26
Comments
Model ReportThank you for submitting. Using Github actions, we have regenerated a report summarising information about your model
Parsed dataSection 1: Summary of your modelModel Submitter: Dan Sandiford (0000-0002-2207-6837) Model Creator(s):
Model name:
(this will be the name of the model repository when created) Model long name: Plate bending earthquakes and the strength distribution of the lithosphere License: Creative Commons Attribution 4.0 International Model Category:
Model Status:
Associated Publication title: Plate bending earthquakes and the strength distribution of the lithosphere Abstract: This study investigates the dynamics and constitutive behaviour of the oceanic lithosphere as it bends and yields during subduction. Two main observational constraints are considered: the maximum bending moment that can be supported by the lithosphere, and the inferred neutral plane depth in bending. We particularly focus on regions of old lithosphere where the ‘apparent’ neutral plane depth is about 30 km. We use subduction modelling approaches to investigate these flexural characteristics. We reassess bending moment estimates from a range of previous studies, and show a significant convergence towards what we call the ‘intermediate’ range of lithosphere strength: weaker than some classical models predict, but stronger than recent inferences at seamounts. We consider the non-uniqueness that arises due to the trade-offs in strength as well background (tectonic) stress state. We outline this problem with several end-member models, which differ in regard to relative strength in the brittle and ductile regimes. We evaluate the consistency of these models in terms of a range of constraints, primarily the seismic expression of the outer rise. We show that a 30 km neutral plane depth implies that net slab pull is not greater than about 2 TN m−1. In contrast, models with low brittle strength imply that regions with a 30 km neutral plane depth are under moderate net axial compression. Under these conditions, reverse faulting is predicted beneath the neutral plane at depths >30 km. We show that moderate variations in background stress have a large impact on the predicted anelastic dissipation. We suggest brittle reverse faulting is a marginal phenomenon which may be inhibited by moderate changes in background stress. Scientific Keywords:
Funder(s):
Section 2: your model code, output data** No embargo on model contents requested****Include model code:** True Model code existing URL/DOI: https://github.com/dansand/subduction_GJI2022 Model code notes: Model setup is provided by an ASPECT input file and a WorldBuilder file (https://github.com/GeodynamicWorldBuilder/WorldBuilder). Minor modifications to the ASPECT source code were implemented and are discussed in the associated publication as well as the Include model output data: True Model output data notes: Computations were done using the ASPECT code version 2.4.0. ASPECT output data from 2 simulations are included with this model. The reference model is the same model setup/data described in Sandiford and Craig, (2023). An alternative model is included in which the over-riding plate is welded to the left sidewall at the start of the simulation (whereas the initial temperature field in the reference model has a ridge). Note that both simulations develop a short-wavelength instability in the free surface of the over-riding plate, which begins approximately 3 Ma after the start of the simulation. The top level directories contains typical ASPECT output files, including log.txt and restart files. The primary output data consists of:
Section 3: software framework and compute detailsSoftware Framework DOI/URL: Found software: ASPECT v2.4.0 Software Repository: https://github.com/geodynamics/aspect Name of primary software framework: ASPECT v2.4.0 Software & algorithm keywords:
Section 4: web material (for mate.science)Landing page image: Filename: res_fig_final_ann.png Animation: Filename: animation Graphic abstract: Filename: gpe_fm26.png Model setup figure: Filename: s1a.png Errors and WarningsAssociated Publication Could not parse Embargo date. Check format is Next steps
|
Model ReportThank you for submitting. Using Github actions, we have regenerated a report summarising information about your model
Parsed dataSection 1: Summary of your modelModel Submitter: Dan Sandiford (0000-0002-2207-6837) Model Creator(s):
Model name:
(this will be the name of the model repository when created) Model long name: Plate bending earthquakes and the strength distribution of the lithosphere License: Creative Commons Attribution 4.0 International Model Category:
Model Status:
Associated Publication title: Plate bending earthquakes and the strength distribution of the lithosphere Abstract: This study investigates the dynamics and constitutive behaviour of the oceanic lithosphere as it bends and yields during subduction. Two main observational constraints are considered: the maximum bending moment that can be supported by the lithosphere, and the inferred neutral plane depth in bending. We particularly focus on regions of old lithosphere where the ‘apparent’ neutral plane depth is about 30 km. We use subduction modelling approaches to investigate these flexural characteristics. We reassess bending moment estimates from a range of previous studies, and show a significant convergence towards what we call the ‘intermediate’ range of lithosphere strength: weaker than some classical models predict, but stronger than recent inferences at seamounts. We consider the non-uniqueness that arises due to the trade-offs in strength as well background (tectonic) stress state. We outline this problem with several end-member models, which differ in regard to relative strength in the brittle and ductile regimes. We evaluate the consistency of these models in terms of a range of constraints, primarily the seismic expression of the outer rise. We show that a 30 km neutral plane depth implies that net slab pull is not greater than about 2 TN m−1. In contrast, models with low brittle strength imply that regions with a 30 km neutral plane depth are under moderate net axial compression. Under these conditions, reverse faulting is predicted beneath the neutral plane at depths >30 km. We show that moderate variations in background stress have a large impact on the predicted anelastic dissipation. We suggest brittle reverse faulting is a marginal phenomenon which may be inhibited by moderate changes in background stress. Scientific Keywords:
Funder(s):
Section 2: your model code, output data** No embargo on model contents requested****Include model code:** True Model code existing URL/DOI: https://github.com/dansand/subduction_GJI2022 Model code notes: Model setup is provided by an ASPECT input file and a WorldBuilder file (https://github.com/GeodynamicWorldBuilder/WorldBuilder). Minor modifications to the ASPECT source code were implemented and are discussed in the associated publication as well as the Include model output data: True Model output data notes: Computations were done using the ASPECT code version 2.4.0. ASPECT output data from 2 simulations are included with this model. The reference model is the same model setup/data described in Sandiford and Craig, (2023). An alternative model is included in which the over-riding plate is welded to the left sidewall at the start of the simulation (whereas the initial temperature field in the reference model has a ridge). Note that both simulations develop a short-wavelength instability in the free surface of the over-riding plate, which begins approximately 3 Ma after the start of the simulation. The top level directories contains typical ASPECT output files, including log.txt and restart files. The primary output data consists of:
Section 3: software framework and compute detailsSoftware Framework DOI/URL: Found software: ASPECT v2.4.0 Software Repository: https://github.com/geodynamics/aspect Name of primary software framework: ASPECT v2.4.0 Software & algorithm keywords:
Section 4: web material (for mate.science)Landing page image: Filename: res_fig_final_ann.png Animation: Filename: animation Graphic abstract: Filename: gpe_fm26.png Model setup figure: Filename: s1a.png Errors and WarningsAssociated Publication Model Repository Slug Could not parse Embargo date. Check format is Next steps
|
Review RequestedA review of this submission has been requested from @ModelAtlasofTheEarth/model_reviewers |
Model repository created at https://github.com/ModelAtlasofTheEarth/sandifordcraig-2023-subduction |
Model repository created at https://github.com/ModelAtlasofTheEarth/sandifordcraig-2023-subduction |
Model repository created at https://github.com/ModelAtlasofTheEarth/sandifordcraig-2023-subduction |
Model repository created at https://github.com/ModelAtlasofTheEarth/sandifordcraig-2023-subduction-1 |
-> submitter ORCID (or name)
0000-0002-2207-6837
-> slug
sandifordcraig-2023-subduction
-> license
CC-BY-4.0
-> alternative license URL
No response
-> model category
model published in study, forward model
-> model status
None
-> associated publication DOI
https://doi.org/10.1093/gji/ggad230
-> model creators
No response
-> title
No response
-> description
Tectonic plates are recycled into the mantle through subduction, where they bend and deform in various ways, such as brittle failure. This process creates deep sea trenches and results in characteristic earthquake patterns and gravity anomalies. In this study, we used a numerical model to investigate plate bending dynamics, complementing simpler approaches like flexural yield strength envelopes. We focused on the competition between bending stress and sources of net in-plane stress, such as slab pull, which influences the plate's neutral plane depth. It is difficult to reconcile the 'apparent' neutral plane depth with a net slab pull force greater than about 2 TN/m. Deviatoric compression in subducting plates more easily explains reverse earthquakes at depths of 20-50 km in the bending plate.
-> abstract
No response
-> scientific keywords
Dynamics of lithosphere, Lithospheric flexure, Subduction, Earthquakes
-> funder
https://ror.org/05mmh0f86, DP150102887
https://ror.org/03wnrjx87, URF\R1\180088
https://ror.org/02b5d8509
-> model embargo?
No response
-> include model code ?
-> model code/inputs DOI
https://github.com/dansand/subduction_GJI2022
-> model code/inputs notes
Model setup is provided by an ASPECT input file and a WorldBuilder file (https://github.com/GeodynamicWorldBuilder/WorldBuilder). Minor modifications to the ASPECT source code were implemented and are discussed in the associated publication as well as the
model_code_inputs/README.md
directory.-> include model output data?
-> data creators
No response
-> model output data DOI
No response
-> model output data notes
Computations were done using the ASPECT code version 2.4.0. ASPECT output data from 2 simulations are included with this model. The reference model is the same model setup/data described in Sandiford and Craig, (2023). An alternative model is included in which the over-riding plate is welded to the left sidewall at the start of the simulation (whereas the initial temperature field in the reference model has a ridge). Note that both simulations develop a short-wavelength instability in the free surface of the over-riding plate, which begins approximately 3 Ma after the start of the simulation. The top level directories contains typical ASPECT output files, including log.txt and restart files. The primary output data consists of:
./solution
sub-directory (e.g. solution-00000.0000.vtu). At each output step, there are 48 vtu files written. These can be opened with Paraview using the solution.pvd file in the top level. Quantities generally have SI units. Velocities are output as meters/year.-> model output data size
47 Gb
-> software framework DOI/URI
https://doi.org/10.5281/zenodo.6903424
-> software framework source repository
https://github.com/geodynamics/aspect
-> name of primary software framework (e.g. Underworld, ASPECT, Badlands, OpenFOAM)
No response
-> software framework authors
https://orcid.org/0000-0003-2311-9402
https://orcid.org/0000-0003-0357-7115
https://orcid.org/0000-0003-0035-7723
https://orcid.org/0000-0001-7098-8198
https://orcid.org/0000-0002-9481-1749
https://orcid.org/0000-0002-8137-3903
https://orcid.org/0000-0001-9489-5236
https://orcid.org/0000-0002-5697-7203
-> software & algorithm keywords
C++, finite-element, adaptive-mesh-refinement, particles
-> computer URI/DOI
https://dx.doi.org/10.25914/608bfd1838db2
-> add landing page image and caption
Downdip component of strain rate tensor and resolved stress difference from the numerical model, focusing on features within the plate/slab. The resolved stress difference is defined as (
-> add an animation (if relevant)
https://github.com/ModelAtlasofTheEarth/model_submission/assets/10967872/f9e647c5-3341-400c-bac3-d5546f1d7226$\dot\epsilon_{II} = \sqrt{J2} = \sqrt{0.5(\dot\epsilon_{i,j}: \epsilon_{i,j})}$ . Upper panel shows the evolution of the model topography (a true free surface). The topographic profile reveals the long-wavelength isostatic thermal subsidence, as well as the flexural topography associated with the subduction zone. The model exhibits a very short-wavelength instability in the free surface of the over-riding plate, which begins approximately 3 Ma after the start of the simulation.
Animation shows the model domain at 2x vertical exaggeration. The scalar field is the effective strain rate, i.e.
-> add a graphic abstract figure (if relevant)
The main panel shows the variation in terms that arise in a 2D "vertically integrated" form of the force balance (or stress equilibrium) equations. Assuming a traction-free surface, the force balance states that across a horizontal section of the lithosphere, the following terms must sum to zero: 1) integrated basal shear traction, 2) the difference in the vertically-integrated deviatoric stress and 3), the difference in the vertically-integrated vertical normal stress (often called the GPE). In the figure, the overbar symbols represent vertical integration across the lithosphere. Specifically, integration from a reference height, (taken here as the mean ridge height) down to a reference depth (taken here as 150 km beneath the reference height). In the main panel, the black line shows the horizontal variation in the vertically integrated deviatoric stress difference (
-> add a model setup figure (if relevant)
The main panel shows the full model domain and initial temperature field. The texture is generated with a line integral convolution of the velocity field. Contours show evolution of the slab during the 10 Myr simulation. Velocity arrows show convergence rates at 5 Myr into the simulation. Inset panels show details of the adaptive mesh refinement during the simulation.
-> add a description of your model setup
The subduction model comprises a rectangular domain with a depth of 2900 km, and an aspect ratio of 4. The initial conditions comprise an adiabatic mantle with a potential temperature of 1350 C and two plates, whose age and thermal structure follows the cooling 1d cooling profile for a half-space (infinite in the depth direction). One of these plates is attached to a slab that extends to 660 km depth, and has an age of 100 Myr at the trench. The upper plate is modelled with a younger thermal age, 25 Myr at the trench. Imposing an initial slab that reaches the transition zone was found to be a more stable initial configuration in terms of instabilities of the free surface. 7 levels of mesh refinement were used, with the largest (Q2) elements having an edge length of 45 km, and the smallest elements have an edge length of ∼ 700 m. The interface is modelled through an entrained weak layer approach. A thin layer (here 2 km thick) represented by a separate composition is imposed on the top of the subducting plate, as well as between the subducting slab and upper plate. This composition has a low coefficient of friction, providing a shear stress that varies between between about 10 - 20 MPa throughout the plate interface domain. See the included model input file (.prm) for further details.
Please provide any feedback on the model submission process?
No response
The text was updated successfully, but these errors were encountered: