Analysis of the 'trench pull force' in a numerical subduction zone model

This repository contains data and post-processing analysis for a 2D numerical model of subduction.

The focus of this analysis is to calculate the vertically integrated horizontal force balance.

The numerical model provides solutions to the stress equilibrium equations in a 2d Cartesian domain, on a uniform grid. The model was run with the open-source code MDOODZ7.0, developed by Thibault Duretz and collaborators, based on the input file (or SET) called AnneloreSubduction. This setup was designed to replicate the "forced subduction" model from https://doi.org/10.1093/gji/ggaa092.

This model analysis (jupyter notebook) was developed as part of the response to reviewer comments for the following manuscript: (https://doi.org/10.31223/X5TQ50). In this manuscript, a scaling relationship is developed for the net horizontal force due to the relative trench topography (the "trench pull force"). The scaling is based on assumptions about the depth at which the support of the non-isostatic topography occurs; this is equivalent to a statement about the depth at which vertical normal stress equilbrates, between a column of lithosphere beneath the trench, compared to an isostatic column in the trailing plate. The notebook also considerers other issues raised by the reviewers, such as the hydrostatic approximation in the asthenosphere.

The mathematical background is contained in this Jupyter notebook, so that I didn't have translate it to Github-compatable markdown. A article preprint can be found in the ./papers directory.

Static image of model

Overview of repository

• model_output_data
  Contains the H5 outputs from the model. These files have been compacted to include the necessary data fields required for the analysis.

• postprocessing_notebooks
  Contains the Jupyter notebook(s) that implement the analysis of the model outputs.

• model_code_inputs
  Contains the input files for MDOOD7.0 used to run the model.

• papers
  Contains relevant papers related to the modeling process and methodology.

To do

• Add MDOODZ7.0 input files
• Add conda env file
• Add licence
• Add papers
• Add Fig 2 from manuscript (domain)