Grant: Discovery Launch Supplement

Turbulence, the chaotic motion of a fluid, is a phenomenon of direct importance to multiple facets of everyday life, from large-scale atmospheric dynamics to blood flow in the heart. However, despite their importance, turbulent processes remain poorly understood due to their inhomogeneous and stochastic nature over a vast range of spatial and temporal scales. This research program aims to develop an improved understanding of turbulence in stratified flows, scenarios in which the fluid contains regions of variable density as is encountered in a variety of geophysical, environmental, and industrial settings. The ocean is stratified, for example, with the average density gradually increasing with depth as relatively warm light surface waters transition to colder denser bottom waters. This research will address how turbulence enhances the mixing and transport of scalar quantities (such as heat, carbon and pollutants) throughout a stratified fluid. Of particular importance is developing more robust mathematical models for characterizing the turbulence-enhanced mixing of heat in the ocean, a leading area of uncertainty in global climate modeling and thus a topic of direct importance to Canadians. Turbulent mixing processes also play central roles in a variety of other areas of societal importance, including the modeling of wildfire spread and the dispersion of pollutants in our oceans and atmosphere. To develop a better understanding of turbulent mixing processes, the proposed research will analyze data collected from a variety of sources including cutting-edge numerical simulations run on supercomputers, laboratory experiments, and observational ocean measurements. A key objective will be to investigate how mixing mechanisms, and their associated efficiency, vary as a function of the underlying physical parameters governing the flow, such as the turbulence intensity, the stratification strength, and the fluid's molecular properties. By developing an improved mathematical description of the variety of mechanisms that may arise, and then quantifying their spatial distributions and probability of occurrence as a function of the underlying flow parameters, our research will inform the development of more universal and accurate mixing models. Further, as the data sources informing our research are growing exponentially larger and more complex, this research will leverage artificial intelligence techniques to efficiently draw trustworthy physical insight from datasets that are too massive to analyze by hand. This award will directly facilitate the training of new generations of Canadian researchers, allowing them to gain experience in a diverse set of skills in the mathematical, computational, and experimental sciences, and will promote their interactions with established international networks of multi-disciplinary researchers.