Mirror of WRF-Chem Tutorials from NOAA

[Will H]

Today I took an old, mirrored WRF-Chem tutorials and documentation site, that was taken down by NOAA: National Oceanic & Atmospheric Administration about 3 years ago, and turned it into a reliable, modern, GitHub Pages-hosted resource that still preserves the original content and URLs.

The goal was not just to “host HTML,” but to make sure the tutorials are actually usable end to end, including the parts that typically break on older sites: missing assets, dead FTP links, large downloads, and pages that are hard to navigate.

I built a clean GitHub Pages deployment from the repository’s /docs folder and validated that the site is truly serving the mirrored files. After the initial publish, I verified deployment status through Actions and confirmed live serving by directly checking URLs and response headers for key pages and assets.

I addressed the biggest practical issue for real users: large tutorial datasets and archives that cannot reasonably live inside a Pages repo. I moved those large files into GitHub Releases, including handling multi-gigabyte tarballs by splitting them into parts that stay under size limits. I generated and published SHA-256 checksum files so users can validate integrity after download. Then I added a dedicated Downloads page that clearly explains what is hosted in Releases, how to reconstruct split archives, and how to verify checksums.

I handled legacy FTP dependencies. Some tutorial exercises historically depended on FTP-hosted directories, which modern browsers either do not support or treat as insecure. I replaced those FTP links with working links on the site or with pointers to the Downloads page, and I created browsable index pages for each exercise folder so users can see exactly what files exist and download them directly. This included adding exercise folder browse pages for exe001, exe002, exe003, and exe004, and linking them from the Downloads page for quick access.

I also implemented a “modern portal” as a front door for the project. I generated a portal scaffold with a search index, added a simple client-side search experience, and moved the default homepage to redirect users to the portal while preserving the full classic mirror at a stable URL. This keeps the original content intact while giving new users a much better navigation and discovery experience.

Finally, I tagged a stable checkpoint before modernization so there is a known good state that can always be referenced or restored if needed.

The end result is a functional WRF-Chem tutorials site that helps users find the right tutorial year, locate supporting materials, and download the actual exercise inputs and large datasets without fighting broken links or outdated hosting patterns.

https://hathewaywill.github.io/wrf-chem-site/