GeoAI and Satellite Fusion for Early Peatland Wildfire Prediction and Carbon Loss Monitoring

Project

PeatWatch — PFE-06 — Climate, GeoAI, Remote Sensing, Climate Tech

Predict underground peat fires days before they reach the surface using satellite radar and soil-moisture data.

Objectives

• Build a satellite-data pipeline using Sentinel-1/2, CHIRPS, and moisture indices.
• Model peat water-table depth from radar data.
• Generate a daily fire-risk score on 1 km grids.
• Develop a live risk heat map and threshold-based alerts dashboard.

Key challenge

Isolate the groundwater signal from vegetation and surface roughness.

Target

Provide days of advance warning before real fires at a stated false-alarm rate.

Tech stack

Python, PyTorch, Streamlit, Earth Engine, Rasterio, Copernicus, GeoPandas, XGBoost, Leaflet.js.

Profile

Student in Computer Science, data science, or GIS.

Format and location

• 1 intern
• Full-time, on-site internship
• 4 to 6 months
• Knowledge Place, Monastir, Tunisia
• Market: Indonesia and Malaysia

General requirements

• Willingness to learn and take on the challenge
• Relevant skills and interests
• Adequate written and spoken English
• Ability to work alongside people from diverse backgrounds and disciplines
• Availability for long-term work after the internship