Wind Field

Chapter 1 teaches vectors by moving one thing toward the mouse. This expansion uses the same vector math, but the vectors are today’s weather.

The page asks Open-Meteo for wind speed and direction at nine points across the US northeast. Each speed and direction pair becomes a vector; bilinear interpolation fills the space between the nine samples so every pixel has a wind vector. Then hundreds of test points get dropped each frame and stepped along the field, tracing the streamlines you see. Line color follows the wind strength along each streamline, so calm regions read cool and fast ones warm.

Data source: Open-Meteo Forecast API
Endpoint: https://api.open-meteo.com/v1/forecast?latitude=...&longitude=...&hourly=wind_speed_10m,wind_direction_10m
Access: Public, no key. Loaded from an hourly server-side refresh when possible, with a bundled snapshot fallback, and a live browser fetch only when the API allows CORS. Data window: Jul 18, 2026 (pulled Jul 18, 2026 UTC)

Because it is a forecast for a specific hour, the pattern is a real snapshot of the atmosphere, not a noise field dressed up to look like one.