Data and code for: Field morphometric fingerprint reveals the limits of a reduced stochastic-threshold model of honeycomb weathering
Description
This dataset and code package supports the manuscript “Field morphometric fingerprint reveals the limits of a reduced stochastic-threshold model of honeycomb weathering”. It contains analysis-ready morphometric measurements for 819 field cavities, OFAT and three-level full-factorial experimental designs and summaries, run-level metrics for 1,350 factorial simulations, source data for Figures 5 and 7, bootstrap and sensitivity-analysis outputs, four computational scripts, software-environment information, file and checksum manifests, and Supplementary Tables S5–S11. The field observations represent one sandstone surface in South China. The bootstrap analyses quantify within-surface sampling variability and do not constitute external validation. The nondimensional model parameters and the identified relatively low-mismatch region should not be interpreted as calibrated physical parameter estimates.
Files
Steps to reproduce
1. Download and extract the ZIP archive. 2. Create a Python 3.12 environment and install the dependencies listed in environment/requirements.txt. 3. Consult README.md for the complete directory structure and command-line entry points. 4. code/parameter_experiments.py reproduces the OFAT and three-level full-factorial experiments. 5. code/robustness_analysis.py reproduces the bootstrap, numerical-endpoint, observation-operator and statistical sensitivity analyses. 6. code/make_main_figures.py generates Figures 5 and 7 from the archived analysis-ready metrics. 7. File integrity can be checked against SHA256_MANIFEST.csv. The archived source-data tables allow the reported figure values to be inspected without rerunning all simulations.
Institutions
- East China University of TechnologyJiangxi, Nanchang