AI-powered climate risk mapping with QGIS : Predicting flood and drought zones

Learn to combine QGIS spatial analysis with machine learning models to predict and visualize climate hazard zones in real time.

Learning Objectives

By the end of this session, participants will be able to:

1. Import and preprocess climate datasets (ERA5, CHIRPS rainfall, DEM) directly in QGIS
2. Apply a Random Forest classification model to identify flood-prone and drought-risk zones
3. Generate predictive risk maps with temporal scenarios (2030, 2050)
4. Automate the workflow using the QGIS Model Builder and PyQGIS scripts
5. Export publication-ready cartographic outputs (PDF, Web Map)

Session Agenda

• 0:00 – 0:20 | Introduction: GIS + AI for climate risk — state of the art
• 0:20 – 0:50 | Dataset acquisition: ERA5, CHIRPS, SRTM DEM via QGIS plugins
• 0:50 – 1:20 | Random Forest model: training, validation, and QGIS integration
• 1:20 – 1:40 | Generating predictive risk maps and temporal scenarios
• 1:40 – 2:40 | Hands-on lab: participants build their own flood risk map
• 2:40 – 3:00 | Q&A, feedback, and next steps

Target Audience

GIS analysts, environmental engineers, urban planners, NGO field officers, and climate researchers working with spatial data.