Climate Risk and Health Prediction Challenge
Climate Risk and Health Prediction Challenge supports machine-learning participants in building models to identify climate-sensitive deaths and improve understanding of climate-related health risks.
Funder: Zindi
Due Dates: October 18, 2026 (Submission close) | October 18, 2026 (Submission selection close)
Funding Amounts: $1,000 USD prize pool: $500 first place, $300 second place, $200 third place
Summary: A machine-learning competition to identify climate-sensitive deaths using health, demographic, geographic, and climate-related data.
Key Information: Winners must submit reproducible code for verification and assign worldwide copyright in the winning solution to Zindi.
Description
Run by Zindi, this challenge asks participants to build supervised machine-learning models that predict whether a recorded death falls into a climate-sensitive category. Using health, demographic, geographic, and climate-related data, participants explore how environmental conditions may contribute to mortality alongside age, living conditions, and geography. The challenge responds to concerns that changes in rainfall and temperature can increase health risks in low-resource settings, particularly for vulnerable populations.
The competition applies classification, prediction, and geospatial machine learning to a real-world public health problem. Models must produce both a binary classification and a probability of climate sensitivity. Evaluation balances precision and recall with the ability to rank climate-sensitive cases, supporting the broader goal of improving understanding of climate-related health risks.
