Functional ecosystem properties estimated from satellite PRISMA and Sentinel-2 data in natural European sites
Year: 2026
Type: article
Abstract: Science-based indicators are needed to monitor ecosystems at different scales, especially with increasing climate change and anthropogenic pressures, and varying impacts according to the scale of analysis. Ecosystem functional properties report on key ecosystem processes and dynamics at multiple scales, from community to biome level, providing a dynamic view of ecosystem carbon- and energy-related processes, useful for monitoring short-term changes. These quantities can be derived from flux measurements, such as those collected by the Integrated Carbon Observation System flux tower network. Here, modeling of selected ecosystem functional properties with hyperspectral satellite data was carried out at fifteen European sites belonging to five different plant functional types. The results, compared to those obtained using a dense Sentinel-2 time series, highlight the potential of hyperspectral narrowbands. Gross primary productivity, light use efficiency and net ecosystem exchange have been predicted with reasonable accuracy, independently by the plant functional type or site latitude . Two different modeling approaches have been compared: Random Forests and Extreme Gradient Boosting. The potential of monitoring ecosystems using the EFPs approach is discussed in consideration of the pros and cons in data use, and the future increased availability of hyperspectral missions.
Institutions: National Research Council, Consiglio Nazionale delle Ricerche - Istituto di Ricerca sugli Ecosistemi Terrestri, National Academies of Sciences, Engineering, and Medicine, National Agency for New Technologies, Energy and Sustainable Economic Development, Eurac Research, Institute of Atmospheric Pollution Research
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
SDGs: Life on land, Climate action
Open Access status: gold
Estimated APC paid: $1,060
OpenAlex ID: W7160629989
