Great project — you're tackling exactly the right problem. Making ray tracing accessible to everyday operators (especially POTA / SOTA / IOTA activators) would be a real contribution to the community.
A few things that might help from our end:
Data feeds:
We've built a set of open-source MCP servers (
qso-graph.io) that provide exactly the data you're pulling manually — solar indices (SFI, Kp, Bz) updated every 15 minutes, WSPR beacon data, and POTA park/spot information. They're pip-installable and work with Claude, ChatGPT, Copilot, or any MCP client. If you're using AI to help with development, these tools let the AI query propagation data directly.
Pre-built datasets:
We publish propagation signature datasets on SourceForge (
sourceforge.net/projects/ionis-ai/) — 175M+ aggregated signatures from WSPR, RBN, PSK Reporter, and contest logs, with solar conditions joined. SQLite format, ready to use. These might save you the ETL work of building your own pipeline from raw spots.
IRI ionospheric data:
PHaRLAP needs ionospheric electron density profiles. We maintain a 164M-row IRI lookup table that maps (lat, lon, hour, month, SFI) to ionospheric parameters. Happy to share the methodology if that's useful for your ray trace inputs.
Web hosting:
For getting your web service public, Vercel (
vercel.com) does free static site hosting with auto-deploy from GitHub. We just deployed a DXpedition propagation analysis demo this way — Next.js frontend, static data files, zero server cost. If your PyLap frontend is web-based, same approach would work. (
https://dxpedition-demo.vercel.app/predictions )
POTA integration:
Our pota-mcp server provides live activator spots, park info, and scheduled activations via API. Could feed directly into your ray trace visualization — show the activator's park location, then overlay the predicted ray paths.
We're also working on comparing ML-based propagation predictions against VOACAP and real observations. Our first analysis (3Y0K Bouvet Island) just went live at (
https://dxpedition-demo.vercel.app/predictions) if you want to see propagation data presented interactively.
Would be glad to collaborate or share data. The more people working on accessible propagation tools, the better.
73, Greg, KI7MT