Enviromental Modeling in the era of HPC

The past two decades have witnessed a profound transformation in numerical weather prediction, climate, and ocean modeling, largely driven by the widespread availability of massively parallel computing resources. 

This presentation will explore the key advancements that have propelled this evolution, the underlying factors that necessitated these changes, and the enduring computational and numerical challenges as we transition into and beyond the exascale computing era. 

A central focus will be on the numerical modeling of natural hazards, specifically tropical cyclones and tsunamis. We will identify the computational and numerical requirements essential for their accurate and effective simulation. Regarding tropical cyclones, which pose significant societal and economic threats, advancements in numerical techniques have led to substantial improvements in track forecasting. However, the mechanisms governing rapid intensification (RI) remain an area of active research. 

This talk will demonstrate how recent progress in very high-resolution large eddy simulations is enhancing our understanding of the fundamental processes that control RI dynamics. With respect to tsunami modeling, the discussion will center on simulating their interaction with coastal features, particularly within the context of natural protection strategies.

 Finally, while exascale computing now enables simulations at resolutions previously unattainable, the emerging field of quantum computing is also exploring applications in fluid dynamics. 

We will consider the potential for these advanced computational paradigms to significantly reduce the time required for large-scale simulations in the future.

Speaker:

Dr. Simone Marras is an Associate Professor of Fluid Dynamics in the Department of Mechanical and Industrial Engineering at the New Jersey Institute of Te chnology (NJIT). After graduating with a "Laurea" (v.o.) in Aerospace Engineering from Polimi in 2005, he earned his Ph.D. in Scientific Computing (focusing on numerical weather prediction) from the Universitat Politecnica de Catalunya and the Barcelona Supercomputing Center in Spain (2012). Before joining NJIT, he was a research scientist at Stanford University and the Naval Postgraduate School in California. His research interests span computational fluid dynamics, atmospheric turbulence, and computational neuroscience (studying the brain and memory). Dr. Marras is a co-developer of several open-source software packages for scientific computing and serves as an Associate Editor for the Quarterly Journal of the Royal Meteorological Society and Geoscientific Model Development. He is currently on sabbatical leave in the Department of Mathematics and Physics at Charles University in Prague and is always willing to collaborate with new students and scientists on innovative topics.

Free admission, open to all members of the university community and the public, subject to availability.