Sai Ravela gives a Data Science for Atmospheric Sciences colloquim at  Ludwig Maximilian University of Munich (LMU)  on a Bayesian Theory for Coherent Fluids that non-parametrically incorporates amplitude and geometry in for robust inference, uncertainty quantification, sampling, and reduced modeling. The approach provides for a historically Bayesian alternative to modern interest in optimal transport while generalizing it to include both geometry and field amplitude errors. Full video coming soon!