Disentangling Complex Flows by Blurring the Data

Hussein Aluie, University of Rochester, Rochester, NY, United States, Matthew W Hecht, Los Alamos Nat'l Lab, Los Alamos, NM, United States, Mathew E Maltrud, Los Alamos National Laboratory, Los Alamos, United States, Shikhar Rai, Woods Hole Oceanographic Institution, Woods Hole, MA, United States, Mahmoud Sadek, University of Rochester, Mechanical Engineering, Rochester, NY, United States, Benjamin Aaron Storer, University of Waterloo, Applied Mathematics, Waterloo, ON, Canada and Geoffery K Vallis, University of Exeter, Exeter, United Kingdom
Abstract:
The multi-scale nature of oceanic flow presents a major difficulty in understanding, modeling, and predicting circulation and mixing. I will discuss an approach to unravelling these complex flows by blurring or coarse-graining the data. The approach is more versatile and powerful than the `mean-eddy' decomposition traditionally used in physical oceanography, and more broadly applicable than spectral, or Fourier-based, analysis. It allows for understanding and quantifying the dynamics + processes at different scales locally in space. I will present examples on how this has enabled us to probe the energy cascade, baroclinic instability, wind forcing, and also to measure local power spectra in realistic settings using data from altimetry, scatterometry, reanalysis, and high-resolution models.
Back to: Data Science for Modern Oceanography: Statistics, Machine Learning, Visualization, and More III Posters