Dissolved Gaseous Mercury Measurements in Seawater with Analysis Code

This repository contains data and code for estimating the effect sizes of solar radiation and temperature on dissolved gaseous mercury in seawater. It includes raw measurement data, pre-processed data, and 10 sequential analysis modules for Bayesian modeling and visualization. The code is provided in both R and Python.

Use Cases

• Apply the provided Bayesian models from module_4_model.rmd to estimate causal effect sizes of solar radiation on mercury levels.
• Use the simulation data generated by module_2_simulation.rmd to test and validate causal inference frameworks.
• Perform posterior predictive checks using the code in module_5_postpredict.rmd to assess model fit for environmental variables.
• Visualize the results of the causal analysis using the plotting functions in module_6_result_plotting.ipynb and module_7_postpredict_plotting.ipynb.

Strengths

• Includes 10 modular code files for a complete, reproducible analysis pipeline from data selection to visualization.
• Provides both raw measurement data and pre-processed data, supporting transparency and different analysis starting points.
• Code is provided in both R and Python, catering to users of different programming environments.

Limitations

• The core dataset size, row count, and specific variables are not documented, making initial assessment of statistical power difficult.
• The 'data' folder intended for final model outputs is empty, requiring users to execute the code pipeline to generate results.
• The analysis relies on specific R and Python package versions as defined in requirement files, which may pose dependency challenges.

Provenance

Source

Harvard Dataverse, authored by Hans-Martin Heyn.

Collection Method

Raw measurement data from environmental sampling, processed through a defined modular code pipeline.

Freshness

Last updated on 2026-03-12.