Groundwater model reproducibility - References

[Pauline Marsilla]

Hi everyone,

I'm looking into groundwater modeling reproducibility and I was wondering if you could recommend any (detailed) references /  guidelines  / articles? I’m particularly interested in standards that ensure a model can be audited or updated years after its initial development.

• I am already aware that the use of scripts/Python is a major benefits for ensuring better model reproducibility / transparency, and I (/within my team we) have been switching to Flopy. 
• However, I often work with models built by third parties and I would like to 'easily' be able to adjust input data or re-run their processing logic in the future.
• I already have access to basic (regional/internal) guidelines for groundwater modelling, so I am working on improving those.

Thank you,

Pauline

[Mahesh Maskey]

Hi Pauline,

This is a very important topic, especially as more groundwater models are transferred between teams and institutions.

In addition to scripted workflows (e.g., Python/Flopy), the following references provide concrete guidance on long-term reproducibility, documentation, and archiving:

1. USGS Policy on Documenting, Archiving, and Public Release of Groundwater Models
Practical structure for model archives, metadata requirements, and reproducibility expectations:
https://water.usgs.gov/ogw/policy/gw-model/

2. ModelArchiver (USGS Open-File Report 2017-1149)
Step-by-step guidance for creating standardized groundwater model archives:
https://pubs.usgs.gov/of/2017/1149/ofr20171149.pdf

3. USGS Metadata Guidance for Model Archives
Details on metadata preparation and DOI practices for long-term accessibility:
https://water.usgs.gov/ogw/policy/gw-model/modelers-prep-metadata.html

4. Reilly & Harbaugh (2004) – Guidelines for Evaluating Ground-Water Flow Models
Although focused on model evaluation, this report outlines documentation standards necessary for independent review and reconstruction:
https://pubs.usgs.gov/sir/2004/5038/PDF/SIR20045038_ver1.01.pdf

In my experience, the main reproducibility gap is often not the MODFLOW input files themselves, but undocumented preprocessing steps (GIS workflows, time-series aggregation, parameter scaling, etc.). Archiving preprocessing and postprocessing scripts alongside model inputs greatly improves long-term auditability.

I would also be interested to hear whether others have implemented containerization (e.g., Docker) or structured version tracking for MODFLOW executables.

Best regards,

Mahesh L. Maskey, Ph. D. in Hydrologic Sciences.

Research Computational Hydrologist

Agrohydrology, Hydroinformatics and Computational Hydrology

Cell: (530) 220-5562
Email: mma...@formerstudents.ucdavis.edu

Inclusive/Honest/Deligent/Transparent/Productivity/Research/Teaching/Mission