Climate Connections This forecasting system could predict exactly which streets will flood
Nov 17, 2025
Felix Santiago Calazo, a researcher in hyperlocal flood forecasting in San Juan, Puerto Rico, blends community knowledge with advanced technology to predict flooding at the block level. He discusses the vital role of local input in shaping accurate flood models, drawing on residents' memories of past events. The team deploys solar-powered sensors throughout the city to monitor flood conditions in real-time. This innovative approach aims to enhance preparedness as climate change intensifies storm impacts.
AI Snips
Chapters
Transcript
Episode notes
Hyperlocal Flood Forecasting Improves Preparedness
- Hyperlocal flood forecasts can predict flooding down to specific streets and times during storms.
- This improves preparedness beyond general rainfall amounts by linking rain to expected street-level impacts.
Local Knowledge Shapes Flood Maps
- Felix Santiago Calazo describes telling people which street will flood when a certain rainfall occurs.
- He gathers residents' memories of historic floods to map where water goes during storms.
Use Sensors And Community Data Together
- Deploy inexpensive solar-powered flood sensors across a city to track flooding in real time.
- Combine sensor data with resident reports to produce more detailed forecasts of future floods.