v2.0.1 (Dawn) - Metadata Patch

What's New:

• Citation: Added CITATION.cff for better integration with GitHub and Zenodo.
• Metadata: Updated README.md with a "Cite as" section and improved metadata.

Release History (Major Versions):

• 2.0 (Dawn): Numerical Engine Revision (2019-12-13)
  • Enhanced Occam's Razor: Improved Bayesian model selection logic
  • Memory Optimization: Implemented REAL4 (Fortran) precision for covariance matrices
  • Advanced Visualization: New 2D fault-surface plotting tools for posterior distributions
  • Key Publication: Used by paper Hallo and Gallovic (2020)
• 1.0 (Chasm): Initial Release (2018-01-30)
  • Internal-only version
  • This version served as the core computational framework for the author's PhD thesis

About this Software:

• A unique software package for Parametric Slip Inversion (PSI) of earthquake fault-slip models within a full Bayesian framework (Hallo and Gallovic, 2020). The framework is designed as transdimensional and data-driven, meaning the model complexity is inferred directly from the data. This is implemented by a mathematical Occam's razor inherent to the Bayesian formulation. The transdimensional model space is sampled using Markov Chain Monte Carlo (MCMC) with Parallel Tempering, enabling efficient exploration of high-dimensional and non-linear parameter spaces.
• High-Performance Computing: The code is parallelized in Fortran using MPI (CPU), specifically optimized for deployment on HPC clusters and large-scale seismic inversions.
• This software implements the framework described in https://doi.org/10.1029/2019JB018703

Environment & Dependencies:

• Validated for Fortran 90 Standard
• Optimized for HPC environments (tested on 128+ CPUs)
• Visualization: Included plotting scripts are validated for MATLAB R2016b and R2023b

v2.0 (Dawn) - Numerical Engine Revision

What's New:

• Legacy Archival Release: Contains the original and stable core developed between 2017-10 and 2019-12
• Enhanced Occam's Razor: Improved Bayesian model selection logic to more effectively infer optimal model complexity from observed data
• Memory Optimization: Implemented REAL4 (Fortran) precision for covariance matrices, significantly reducing memory footprint for large-scale HPC inversions
• Advanced Visualization: New 2D fault-surface plotting tools for posterior distributions

Release History (Major Versions):

• 2.0 (Dawn): Numerical Engine Revision (2019-12-13)
  • Enhanced Occam's Razor: Improved Bayesian model selection logic
  • Memory Optimization: Implemented REAL4 (Fortran) precision for covariance matrices
  • Advanced Visualization: New 2D fault-surface plotting tools for posterior distributions
  • Key Publication: Used by paper Hallo and Gallovic (2020)
• 1.0 (Chasm): Initial Release (2018-01-30)
  • Internal-only version
  • This version served as the core computational framework for the author's PhD thesis

About this Software:

• A unique software package for Parametric Slip Inversion (PSI) of earthquake fault-slip models within a full Bayesian framework (Hallo and Gallovic, 2020). The framework is designed as transdimensional and data-driven, meaning the model complexity is inferred directly from the data. This is implemented by a mathematical Occam's razor inherent to the Bayesian formulation. The transdimensional model space is sampled using Markov Chain Monte Carlo (MCMC) with Parallel Tempering, enabling efficient exploration of high-dimensional and non-linear parameter spaces.
• High-Performance Computing: The code is parallelized in Fortran using MPI (CPU), specifically optimized for deployment on HPC clusters and large-scale seismic inversions.
• This software implements the framework described in https://doi.org/10.1029/2019JB018703

Environment & Dependencies:

• Validated for Fortran 90 Standard
• Optimized for HPC environments (tested on 128+ CPUs)
• Visualization: Included plotting scripts are validated for MATLAB R2016b and R2023b