Tornado Tracer

Work in Progress

Author: Joseph Cheatham Last Update: 5/16/2026 Version: 0.1.0

An open source tornado path detection tool for first responders, built on NOAA NEXRAD Doppler radar data. Tornado Tracer is being developed to ingest real-time and historical Level II radar scans, detect tornadic rotation signatures algorithmically, and provide actionable path prediction output for emergency response use.

---

Output Example

KSGF radar radial velocity field showing the Joplin EF5 mesocyclone rotation signature, May 22 2011 22:53 UTC. The tight red/blue couplet near 94.4°W, 37.05°N indicates strong counter-clockwise rotation — the tornadic vortex actively moving through the city

---

Setup

pip install -r requirements.txt

Usage

python3 joplin_2011.py

Downloads and plots radial velocity scans from the May 22, 2011 Joplin tornado event. Output PNGs are saved to the plots/ directory.

---

Current Progress

• Download NEXRAD Level II scans from NOAA's public AWS S3 archive
• Ingest and parse radar sweeps using Py-ART
• Plot georeferenced radial velocity fields with county/state overlays
• Test case May 2011 Joplin, MO EF 5 tornado

Future Work

• Implement Tornadic Vortex Signature (TVS) based detection algorithm
• Use TVS algorithm at each scan to predict path the tornado took
• User interface for first responder deployment

---

Data Source

NEXRAD Level II data is publicly available via NOAA's AWS archive — no API key required.

s3://noaa-nexrad-level2/YYYY/MM/DD/{SITE}/{filename}