uw-ssec/gaia-data-downloader

v0.1.0BSD-3-Clause

Generate and develop hydroclimatological data download scripts for GAIA geoscience research

GAIA Data Downloader Plugin

Generate and develop hydroclimatological data download scripts for the GAIA project.

Overview

This Claude Code plugin encapsulates knowledge from the gaia-data-downloaders repository — 18 Jupyter notebooks implementing reproducible data download pipelines for 10+ environmental data sources. It provides:

Data Downloader Agent — Generates complete download scripts from natural language prompts
Download Script Dev Skill — Assists in developing and debugging download scripts

Supported Data Sources

Source	Type	Description
CONUS404	Weather reanalysis	4 km daily/hourly historical weather (USGS)
HRRR	Weather model	3 km hourly rapid refresh forecasts (NOAA)
WRF-CMIP6	Climate projections	Dynamically downscaled CMIP6 scenarios
PRISM	Climate normals	4 km daily/monthly climate data
Stage IV	Precipitation	Multi-sensor precipitation analysis
USGS	Streamflow	Real-time and historical stream gage data
ORNL DAYMET	Surface weather	1 km daily weather for North America
DEM/SRTM	Elevation	30-90 m global elevation data
Synoptic	Weather stations	Real-time weather station observations
IRIS	Seismic	Seismic waveform data (FDSN)

Installation

Claude Code (via rse-plugins marketplace)

# Add the marketplace (if not already added)
/plugin marketplace add uw-ssec/rse-plugins

# The gaia-data-downloader plugin is now available

For local development:

claude --plugin-dir ./community-plugins/gaia-data-downloader

GitHub Copilot CLI

# Add marketplace
copilot plugin marketplace add uw-ssec/rse-plugins

# Install plugin
copilot plugin install gaia-data-downloader@rse-plugins

Or install directly:

copilot plugin install uw-ssec/rse-plugins:community-plugins/gaia-data-downloader

Usage

Agent: Generate a Download Script

The agent uses a three-phase interaction to ensure scientists have full control over parameters:

1. Describe what you need:

Download HRRR temperature and wind data for the Skagit watershed, January 2024

2. Review the proposed configuration:

Source: HRRR
Model: hrrr
Product: sfc
Date Range: 2024-01-01 to 2024-01-31
Parameters:
  TMP:2 m → temperature_2m
  UGRD:10 m → u_wind_10m
  VGRD:10 m → v_wind_10m
AOI: ../data/GIS/SkagitBoundary.json
Output: ../data/weather_data/hrrr_skagit_202401.zarr
Max Workers: 8
Estimated size: ~2.5 GB

Would you like to adjust any parameters?

3. Adjust and generate:

Add precipitation. Change frequency to 6h. Reduce workers to 4.

The agent generates a Python script with all parameters in a CONFIG dict at the top.

Skill: Develop a Script

Invoke the skill when developing or debugging download scripts:

/gaia-data-downloader:download-script-dev

The skill provides templates, configuration validation, and debugging guidance for all supported data sources.

Prerequisites

Python 3.9+ with xarray, geopandas, rioxarray
Source-specific: herbie-data (HRRR), pyPRISMClimate (PRISM), obspy (IRIS), boto3 (WRF/S3), s3fs (CONUS404)
System: wgrib2 for HRRR (install via conda-forge)
Credentials: NASA Earthdata .netrc for ORNL, SYNOPTIC_API_TOKEN env var for Synoptic

License

BSD-3-Clause