in green, summer in orange, fall in red, winter in cyan, any season in gray, and year-round in purple) using a deep learning model relying on the species observations (black) (from Vanalli et al., under review).
Understanding ecosystems and their dynamics is essential for sustainable management and biodiversity conservation. Recent technologies — from widespread satellite imagery to AI models — are transforming the way scientists observe and predict ecological processes.
In a warming world, our projects combine remote sensing, environmental data, and machine learning to monitor, map, and predict the spatial distributions of species, land covers, and ecosystems across multiple scales.
Papers
- Zermatten, V., Castillo-Navarro, J., Marcos, D. and Tuia, D., 2025. Learning transferable land cover semantics for open vocabulary interactions with remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing (paper).
- Zermatten, V., Castillo-Navarro, J., Jain, P., Tuia, D. and Marcos, D., 2025. EcoWikiRS: Learning Ecological Representation of Satellite Images from Weak Supervision with Species Observations and Wikipedia. In Proceedings of the Computer Vision and Pattern Recognition Conference Workshops (paper).
- Zbinden, R., Van Tiel, N., Kellenberger, B., Hughes, L. and Tuia, D., 2024. On the selection and effectiveness of pseudo-absences for species distribution modeling with deep learning. Ecological Informatics (paper).
- van Tiel, N., Fopp, F., Brun, P., van den Hoogen, J., Karger, D.N., Casadei, C.M., Lyu, L., Tuia, D., Zimmermann, N.E., Crowther, T.W. and Pellissier, L., 2024. Regional uniqueness of tree species composition and response to forest loss and climate change. Nature Communications (paper).
- Zermatten, V., Lu, X., Castillo-Navarro, J., Kellenberger, T., and Tuia, D., 2024. Land cover mapping from multiple complementary experts under heavy class imbalance. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (paper).
