web

Weather Forecast Tracker

Dockerized service that tracks how NWS forecasts drift over time

Weather Forecast Tracker

Tech Stack

PythonFastAPISQLAlchemyAPSchedulerDocker

About This Project

Overview

Weather Forecast Tracker polls the weather.gov API on a schedule and stores a complete 72-hour forecast snapshot for each configured location on every cycle, instead of overwriting the last one. A FastAPI layer exposes a forecast-range query that shows exactly how much the predicted temperature for a given hour swung between polls — useful for measuring forecast reliability, not just reading today's weather.

Motivation

I was curious how much NWS forecasts actually change as the target hour approaches, and there was no easy way to see that — most weather apps just overwrite yesterday's forecast.

Lessons Learned

Storing full snapshots instead of just the latest value turned out to be the entire point of the project — the moment I started keeping history, previously invisible forecast drift became obvious.