SiNCEM laboratory

A laboratory dedicated to marine and coastal environmental simulation, based on numerical modelling and the use of AI.

Album

Modelling the sea. Understanding the coast

Numerical simulation - Machine learning - Coastal resilience

From Global to Coastal

Capturing the full spectrum of scales

Data Assimilation

Integrating observations into high-resolution ocean models

Flood Modelling & Hazard Mapping

Protecting coastlines with high-resolution risk simulations

Coastal Resilience

Supporting communities at the land—sea interface

The Last Mile

Citizen science and real-time coastal monitoring in Rimini

Brief description

SiNCEM is a research laboratory dedicated to numerical simulation and artificial intelligence applied to marine and coastal systems. Research activities cover ocean circulation modeling, waves, storm surge, and coastal dynamics, with approaches ranging from traditional numerical models to machine learning techniques and their hybrids. The laboratory develops digital twins of coastal zones and marine environments, including the ocean soundscape, integrating high-resolution simulations, observational data assimilation, and non-stationary extreme value analysis for coastal risk mapping. Teaching activities are carried out through involvement of master's students, thesis researchers, and PhD candidates.

SiNCEM operates in strategic partnership with CMCC (Centro Euro-Mediterraneo sui Cambiamenti Climatici), contributing to European projects including OCEANITY and AdriaClimPlus. It also closely collaborates with the Decade Collaborative Center for Coastal Resilience (DCC-CR) within the UN Ocean Decade framework.

More info at https://site.unibo.it/sincem/en

Equipment

SiNCEM is equipped with computing and storage infrastructure dedicated to research in numerical modelling, ML, and computationally intensive data analysis:

Calypso high-performance server: a multiprocessor system with 1 TB of RAM, designed for parallel numerical simulations and large-scale climate and oceanographic dataset analysis. The system features 200 TB of storage in RAID configuration, ensuring redundancy, data integrity, and high-performance access for long-term scientific data archives.

Dedicated GPU-accelerated servers for the development and training of machine learning and deep learning models applied to oceanographic, climate, and coastal processes, including Graph Neural Network (GNN)-based emulators and autoregressive models for extreme value analysis.

Location: 4000-WP01-008

Responsibles

Lorenzo Mentaschi

Senior assistant professor (fixed-term)

Paolo Oddo

Associate Professor

Lab Contacts

Lorenzo Mentaschi

Senior assistant professor (fixed-term)

Paolo Oddo

Associate Professor

Nadia Pinardi

Full Professor