Thesis subject
MSc thesis topic: Assessing savannah vegetation structure and biomass with Sentinel-1 radar and spaceborne GEDI lidar data in Kruger National Park
Savannah vegetation in Kruger National Park (South Africa) varies from dense woodlands to open grasslands, driven by fire, rainfall, and herbivore pressure, influencing the tree-grass balance, biomass accumulation, and ecosystem resilience over time. Savannah ecosystems play a critical role in carbon storage, biodiversity, and ecosystem services, yet accurately assessing vegetation structure and biomass remains challenging due to the high variability in vegetation and seasonal dynamics.
Synthetic Aperture Radar (SAR) from Sentinel-1 offers all-weather, high-resolution observations, while Global Ecosystem Dynamics Investigation (GEDI) lidar provides direct measurement of the three-dimensional vegetation structure. This study integrates Sentinel-1 backscatter information with GEDI-derived metrics related to vegetation structure and biomass. By leveraging machine learning and statistical modeling, the research aims to evaluate how well dense Sentinel-1 C-band radar data can be used to assess and map vegetation structure and biomass.
Background
This thesis will utilize temporally dense Sentinel-1 C-band radar data available in Google Earth Engine to analyze its correlation with forest height and biomass information derived from GEDI lidar data. Seasonal dynamics, such as leaf-on versus leaf-off conditions, can be considered. The final model will be used to map vegetation structure and biomass, and potentially change dynamics.
The GEDIDB package (https://github.com/simonbesnard1/gedidb) will be used to access GEDI data.
Software: Google Earth Engine, R/python, ArcGIS/QGIS
Objectives and Research questions
- Analyze the relationship between Sentinel-1 C-band backscatter and GEDI-derived vegetation structure metrics and biomass.
- Assess vegetation structure and biomass across Kruger National Park.
Requirements
- Geo-scripting course (required)
- Advanced Earth Observation (required)
- GEE tutorials, such as (https://developers.google.com/earth-engine/tutorials/edu)
Literature and information
- Wang et al., 2025 https://www.sciencedirect.com/science/article/pii/S1470160X24015188
- David et al., 2022 https://www.sciencedirect.com/science/article/abs/pii/S0034425722003388
- Urban et al., 2020 https://koedoe.co.za/index.php/koedoe/article/view/1621/2479
- Heckel et al., 2021 https://koedoe.co.za/index.php/koedoe/article/view/1679/2916
Theme(s): Sensing & measuring; Modelling & visualisation