Aurora Troccoli
Hello
Aurora Troccoli
Ph.D. Student in Ecology and Marine Biology, University School for Advanced Studies IUSS Pavia
About Me

I am a PhD researcher in Sustainable Development and Climate Change (SDC) at the University School for Advanced Studies IUSS Pavia. I hold a Bachelor’s degree in Natural Sciences and a Master’s degree in Marine Ecobiology from Sapienza University of Rome. I worked as a research fellow at the Italian Institute for Environmental Protection and Research (ISPRA), focusing on the vulnerability of marine and coastal ecosystems to climate change.


My research aims to develop a robust and replicable framework for coastal vulnerability assessment, focusing on the role of Essential Land and Ocean Variables. The study investigates the limitations of current coastal vulnerability approaches and explores how advanced Earth Observation techniques can improve the representation of coastal ecosystems in vulnerability and risk models. The research focuses on the Gulf of Oristano (Sardinia, Italy), a coastal system that includes several Ramsar sites and represents an ideal natural laboratory for studying climate-sensitive coastal environments.


A major component of this work involves the use of hyperspectral satellite data (e.g., PRISMA) combined with machine learning techniques, including Active Learning and Linear Spectral Mixture Analysis (LSMA). These methods allow the identification and mapping of complex wetland habitats, such as Mediterranean salt marshes, improving the accuracy of Land Cover (LC) products in highly heterogeneous coastal environments. By integrating field spectral measurements, satellite observations, and advanced classification approaches, the research aims to produce more accurate and detailed LC maps capable of capturing seasonal variability in coastal ecosystems.


In addition, this work aims to monitor large-scale marine dynamics in the Mediterranean Sea using Essential Ocean Variables, with a specific focus on sea level variability and its spatiotemporal patterns detected through machine learning techniques.


Looking ahead, this work could contribute to improving coastal monitoring systems and support the development of Earth Observation–based tools for climate risk assessment, with potential applications in Digital Twin environments and operational coastal services.


Click here to view my poster.