One of the biggest challenges in climate-health research isn't the science itself — it's the data infrastructure.
The problem
Climate data and health data live in completely different worlds. Climate datasets are gridded, temporal, and massive (think terabytes of reanalysis products). Health data is tabular, often messy, and collected at wildly different spatial and temporal scales.
When a researcher wants to ask "how did last month's heat wave affect hospital admissions in Moshi, Tanzania?" they need to:
- Find the right climate dataset for that region
- Extract and process the relevant variables
- Align it temporally and spatially with health records
- Deal with missing data, different time zones, and unit conversions
Each of those steps can take weeks if you're doing it from scratch.
What we're building
At CGHI, we're building systematic infrastructure to make this easier. That means:
- Data inventories — Cataloging what climate and health data exists across our observatory network
- Standardized pipelines — Reproducible workflows for extracting and harmonizing data
- Accessible tools — Making climate data approachable for health researchers who may never have worked with a NetCDF file
The goal is to reduce the barrier from "weeks of data wrangling" to "run a function and get a clean dataset."
Why it matters
Climate change is already impacting health outcomes globally. But if researchers can't efficiently access and combine the data they need, we're leaving critical questions unanswered. The infrastructure work isn't glamorous, but it's foundational.
More to come on the specific tools and packages we're developing. Stay tuned.