Data-driven Models for Turbulent Submesoscale Flows in the Ocean

Explore data-driven modeling approaches for understanding turbulent submesoscale flows in ocean systems through this comprehensive lecture by Jim Thomas. Learn how machine learning techniques can be applied to model complex oceanic processes that occur at scales smaller than traditional mesoscale phenomena, focusing on the turbulent dynamics that govern these critical ocean flows. Discover the challenges of capturing submesoscale turbulence in traditional ocean models and examine how data-driven methodologies offer new solutions for representing these processes more accurately. Understand the intersection of oceanography and advanced computational methods, including the development of surrogate models that can efficiently simulate turbulent ocean dynamics. Gain insights into the practical applications of these modeling approaches for improving Earth System Models and their ability to represent localized oceanic processes that significantly impact larger-scale climate patterns.