CAREER: Understanding the Physics of Turbulent Flow, Erosion and Depositional Patterns in River Systems

Rivers are geomorphologic features that play an essential role in landscape evolution. As the river landscape changes due to climate change, severe droughts, floods, and human interventions, the fluvial ecosystems and their ecological and economic values respond in unprecedented ways, and the majority of these cases cannot currently be predicted. Understanding and predicting transient dynamics in river systems through tools that accurately estimate flow and sediment transport is still limited, partially because of the difficulty of monitoring sediment but also because of the inability to understand the fluid dynamics.

This work aims to provide a theoretical and numerical framework to study the feedback between turbulent flow, sediment transport, and geomorphologic changes in river systems. The principal investigator and students will develop and implement state-of-the-art physically-based models aided by machine learning that allow the quantification and forecasting of the flow and sediment dynamics in field-scale rivers.

The education and outreach plan, integrated with the research objectives, focuses on (1) engaging young women at college, undergraduate, and graduate levels into Earth science, through participatory writing for the creation of a science comic book, followed by high school curriculum development, as tools to enhance Earth science pedagogy and promote gender equity, and (2) public outreach through the university art museum that is considered to be an informal learning environment.

This study addresses explicitly how convoluted fluid dynamics manifest in fluvial environments, such as regions of massive flow separation, secondary flows, high-velocity core plunges, velocity inversions, and free shear layers; and the role played by macro-turbulence in sediment transport and river morpho-dynamics. The overall objective is to transform the state of the art in quantifying and predicting the fundamental physics of the coupled fluid and sediment mechanisms that control the morpho-dynamic changes in fluvial systems.

A hybrid physics-based/ machine learning algorithm coupled with a sediment transport and morphodynamic solver will be developed and tested at different spatial scales, from laboratory to large river reaches. The hydro-morphodynamic model will use the Large Eddy Simulation (LES) techniques to resolve macro-turbulence and predict the sediment concentration and riverbed evolution in the computational domain.

A dynamically adaptive, process-based domain re-meshing, based on machine learning algorithms, will be applied to refine the complex topography in areas where turbulent structures are dominant and fundamental to understanding and quantifying erosion and depositional processes present in recirculation zones and plunging flows, thus ensuring a sufficient spatial scale resolution to represent geomorphologic processes. Once the fundamental framework is validated, it could be adapted to different river environments to test its spatio-temporal transferability.

The expected societal outcomes of the educational component are focused on: (1) enhancing Earth science learning among women and racial minorities, (2) modifying stereotypes of women in the Earth science community, and (3) increasing the representation of women in Earth science and creating new literacy in gender equity.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.