We are pleased to share that Jean de Dieu Tuyizere, MSc student at the University of Würzburg, has successfully defended his thesis entitled:
“Utilizing deep learning and Earth Observation data to predict land cover changes in Volcanoes National Park, Rwanda.”
The thesis was supervised by Dr. Insa Otte (NETCDA member) and Dr. Martin Wegmann (EORC).
Summary of the research
Jean’s work analyzed and projected land cover change (LCC) in Rwanda’s Volcanoes National Park (VNP) from 2019 to 2049. Using remote sensing, random forest classification, and a hybrid convolutional neural network (Hybrid-CNN), the study provides new insights into the future of land cover dynamics in this important protected area.
Key findings include:
- Between 2019–2024, alpine, subalpine, bamboo, mixed forest, Hagenia-Hypericum, and water areas declined, while herbaceous and bare land expanded significantly.
- The Hybrid-CNN model (97.9% accuracy) projects continued decreases in alpine, bamboo, mixed forest, Hagenia-Hypericum, subalpine, and water cover through 2049, with herbaceous and bare areas expected to expand further.
- Regression analysis revealed that bioclimatic variables influenced LCC patterns differently across land cover types.
Contribution
This research fills a gap in temporal LCC analysis for VNP, delivering high-resolution forecasts that are highly relevant for conservation and park management. It highlights the importance of sustained remote sensing monitoring to inform adaptive strategies in response to environmental and climatic changes.
We congratulate Jean on this important academic achievement and wish him continued success in his future career.