Modeling turbulent flows is one of the most challenging problems in science, attributed to the highly nonlinear multi-scale nature of the system. Numerically resolving the various scales of flow phenomena in practical turbulent problems in engineering and nature requires high spatial and
temporal resolution. These numerical simulations demand extreme computational resources. Moreover, the complex interactions amongst vortical structures in fluid flows create more challenges, particularly
in turbulence. In an era where data exerts a dominant influence on the analysis of complex systems in science and engineering, data-driven methodologies have evolved, leading to tremendous potential in
tackling such challenging problems. When combined with traditional domain science knowledge, effective interpretable modeling frameworks originate.

This seminar will cover research work utilizing graph theory and machine learning to model canonical turbulent flows in engineering and nature. The interaction-based framework of graph theory is used to implement unsupervised clustering techniques for modifying turbulent flows. Deep learning is used to formulate surrogate models for subgrid-scale turbulence closures in oceanographic and cloud-like flows. Efforts to analyze problems in physical and biological sciences using such data-driven methods will also be discussed. The applications include reduced order modeling of climate events, clustering in particle-laden turbulence, biomedical image segmentation, identifying pivotal interactions in various fungal communities for biofertilizer development, revealing microbe-host interactions, and the application of quantum computing to solve problems related to ideal flows. Finally, an overview of the computational resources at Oak Ridge Leadership Computing Facility, that enabled these works, will also be discussed, inviting potential users from the New Rummy Game research community.