PhD position in High-Resolution Mapping Fuel Loads, Fuel Moisture and Microclimates for Next-Generation Wildfire Risk Assessment

Role Overview
The sGlobe lab at KU Leuven (Belgium) and PXL BIO-Research are seeking a highly motivated PhD student to work on the FireRisk project: “High-Resolution Mapping Fuel Loads, Fuel Moisture and Microclimates for Next-Generation Wildfire Risk Assessment”. This position focuses on the second core objective of the project: the generation of near-real-time, high-resolution maps of fuel loads, fuel moisture, microclimate conditions, and fire weather indices for improved wildfire risk assessment and management.

The project addresses the lack of formal risk assessment systems in many regions, including Flanders, where coarse spatial resolution often underestimates local fire dynamics. The PhD candidate will develop next-generation fire weather products that integrate ecological field measurements, drone observations, airborne LiDAR, satellite remote sensing, and artificial intelligence. The goal is to create high-resolution maps across Flanders to support improved wildfire forecasting systems and decision-support tools for land managers and policymakers.

Responsibilities
The PhD will focus on three interconnected research themes:
1. Mapping Fuel Loads: Develop methods to map fuel loads across landscapes using a combination of airborne and UAV LiDAR data, field inventories, and satellite imagery. Utilize deep-learning approaches, such as Convolutional Neural Networks (CNNs), to upscale local measurements to regional scales.
2. Quantifying and Predicting Live Fuel Moisture Content (LFMC): Combine field measurements collected across various ecosystems with Sentinel-1 radar imagery, Sentinel-2 optical imagery, topographic variables, and advanced machine-learning approaches. Explore the use of state-of-the-art multimodal foundation models for Earth observation to improve predictions in data-limited environments.
3. Modeling Microclimate Conditions: Develop high-resolution models of microclimate temperature and relative humidity using extensive logger networks, UAV-derived LiDAR and multispectral data, weather observations, and satellite imagery. Create AI-based models capable of predicting local environmental conditions at unprecedented spatial detail. Integrate these microclimate predictions into fire weather indices such as the Fire Weather Index (FWI) and the Hot-Dry-Windy Index (HDWI).

The role combines intensive fieldwork, drone-based data collection, geospatial analysis, machine learning, remote sensing, and ecological modelling. The candidate will work within an interdisciplinary team of ecologists, remote sensing specialists, fire scientists, and AI researchers.

Requirements and Qualifications
Education:
- MSc degree in a relevant field (e.g., Ecology, Biology, Bioscience Engineering, Environmental Sciences, Physical Geography, Remote Sensing, or a related discipline), or expected to be obtained by the start of the position.
- Excellent grades.

Skills and Experience:
- Strong interest in biodiversity and ecosystem functioning.
- Background in terrestrial ecology and ecological modelling.
- Solid programming skills (e.g., R), remote sensing, and experience with spatial data analysis.
- Fluency in English, both written and spoken.
- Collaborative team player with strong communication skills.

Assets:
- Interest in working with UAVs.

What They Offer
- A full-time PhD fellowship (4 years) following a positive evaluation after one year.
- Preferred starting date: November 2026.
- Location: Division Forest, Nature and Landscape in Leuven, Belgium.
- Competitive salary following KU Leuven assistant scales.
- Benefits include ecocheques, a bicycle and a bicycle allowance, or full reimbursement of public transport costs for commuting.
- Additional benefits including holidays and bonuses.
- Collaboration in a young and dynamic international scientific team.
- Emphasis on work-life balance.
- Access to state-of-the-art UAV platforms, LiDAR systems, high-performance computing facilities, and extensive environmental monitoring networks.