Our Team
REW brings together complementary expertise across climate science, humanitarian systems, and operational practice:
Columbia Climate School · Data Science Institute F4H leads the REW initiative, contributing climate forecasting capacity, food systems network science, and institutional infrastructure for translating research into operational platforms. Columbia's 30-year legacy in seasonal-to-subseasonal climate forecasting underpins the initiative's analytical core. In collaboration with Columbia's Data Science Institute, REW is developing interpretable AI systems that explain their reasoning, trace back to source data, and keep human decision-makers in control — no black boxes.
Tulane University Tulane contributes deep expertise in humanitarian systems, disaster risk reduction, and the institutional dynamics of early warning and early action — bringing critical knowledge of how regional and national actors actually make decisions under uncertainty.
Euro-Mediterranean Center on Climate Change (CMCC) CMCC brings advanced climate modeling capacity and extensive regional experience across Africa and the Mediterranean, strengthening REW's ability to generate high-resolution, actionable climate intelligence for food-insecure contexts.
Independent Humanitarian Practitioners Senior practitioners with operational backgrounds in FEWS NET and crisis response bring ground-level knowledge of what frontline decision-makers actually need — ensuring REW is designed around real operational constraints rather than theoretical frameworks.
The Problem
Current early warning systems are built primarily to inform donor decisions, not the regional, national, and local actors who control most humanitarian response capacity. Information products are optimized for donor board meetings rather than the district official asking: "Should I pre-position emergency supplies this week?" They operate at the wrong scale, move too slowly, and require technical expertise that frontline actors don't have — and weren't built with them.
Our Approach
REW reframes the challenge. Rather than producing more data or better composite scores, F4H is developing a translation layer between world-class humanitarian research and the frontline decision-makers who need it. Three core innovations distinguish the approach:
Co-creation from day one — We map how partners actually make decisions: what information they need, when, and in what format. Local actors define their own triggers and thresholds based on their response capacity and institutional authority.
Brokerage, not platform — REW connects to existing systems — HDX, WFP Hunger Map, FEWS NET, IPC, GAIN's Food Systems Dashboard — synthesizing across them to answer specific decision questions rather than competing with or replacing them.
Nuanced intelligence over composite scores — Instead of reducing risk to a single number, REW provides interpretable intelligence: what specific hazards are likely where and when, which populations are most exposed, and what response gaps exist.
Status
REW is in active development. F4H is pursuing funding to support a proof-of-concept deployment with motivated early-adopter partners at regional and national levels. Interested institutions and researchers are welcome to reach out.
Contact: Professor Michael Puma
