NEXRAD Storm Catalog: 584,039 NOAA Storm Events Paired with Nearby WSR-88D Radar Stations (2015-2024)

Provider
Silurian
License
Apache-2.0 (source)
Size
5 files
Format
Parquet, CSV (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/Silurian/nexrad-storm-catalog)
Access
Free and public. No account with the data provider is needed.
Category
Climate & Weather
Source
https://huggingface.co/datasets/Silurian/nexrad-storm-catalog

About this dataset

A metadata index that pairs storm reports from the NOAA Storm Events Database with nearby NEXRAD WSR-88D weather radar stations, built so a pipeline can fetch the matching Level-II radar volumes from the Unidata NEXRAD Level-2 archive on AWS (s3://unidata-nexrad-level2/). Each of the 584,039 rows in nexrad_event_catalog.parquet is one (event, station) pair with the station within 150 km of the event, for 2015 to 2024. Columns give the station ID (for example KTLX), NOAA event ID, one of 19 event types (thunderstorm wind, hail, tornado, flash flood, flood, heavy rain, winter storm, blizzard, ice storm, lake-effect snow, tropical storm, hurricane, lightning, waterspout, dust storm and others), a coarse six-way category (severe convective, heavy rain and flood, winter weather, tropical, other severe, general), EF scale for tornadoes, UTC begin and end times, great-circle distance, event latitude and longitude, state, magnitude and type, and the NWS episode narrative. nexrad_stations.csv lists 158 radar sites. The catalog holds no radar payloads. Read with pandas or polars. Released under the Apache License 2.0 by Silurian; the underlying NOAA data are US government works.

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("Silurian/nexrad-storm-catalog")
print(ds)

Data structure

File types in sampled files: Parquet, CSV

Files

Example file: nexrad_event_catalog.parquet

Analyze this dataset without downloading it

On Scigantic, NEXRAD Storm Catalog: 584,039 NOAA Storm Events Paired with Nearby WSR-88D Radar Stations (2015-2024) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Climate & Weather datasets