A Duke Global Health Institute Initiative
Open Source
Open-source tools for climate-health data processing, harmonization, and reproducible analysis.
A central goal of CGHI is getting robust climate and environmental data into the hands of health and global health researchers who may have little or no familiarity with climate data — and doing so in a way that is efficient, well-structured, reproducible, understandable, and practically useful for research and intervention.
The tools below are developed and maintained as part of CGHI’s analytic infrastructure effort. They are designed to lower the barrier to working with climate data and to promote reproducible, well-documented workflows across the observatory network and beyond.
Fast, reproducible, hard to mess up
Transforms raw weather data columns into clean, analysis-ready variables without requiring meteorological expertise. Handles derived comfort/stress metrics, wind operations, pressure utilities, and unit management.
Self-describing data that travels well
Creates self-describing Parquet files and Arrow datasets with embedded metadata — column meaning, units, and storage quirks — so files remain self-documenting as they move between projects, machines, teams, and time.
The DGHI-CGHI GitHub organization hosts observatory-specific repositories, shared infrastructure code, and organizational governance documentation. Code is MIT-licensed; documentation is CC BY 4.0.
Shared code and resources across all observatories
Southeast Asia observatory — Sri Lanka disease surveillance alignment
East Africa observatory — Madagascar sub-project
North Carolina / Southeast US observatory
Amazon Basin observatory
For technical questions about weathertools, dataplane, or CGHI’s data infrastructure, use the contact form and select a data or methods inquiry.