Straub Award Ceremony and Distinguished Lecture

Presentation of the 2017 Lorenz G. Straub award
Award Recipient: Allison Goodwell
, University of Colorado Denver

Distinguished Lecture: Identification of physically-interpretable dynamical modes from big data: from turbulence to climate
Distinguished SpeakerEfi Foufoula-Georgiou, Distinguished Professor, Henry Samueli Endowed Chair in Engineering, Departments of Civil and Environmental Engineering
 and Earth System Science, University of California Irvine

Coherent spatio-temporal patterns (modes) in complex systems are regular features whose evolution is expected to be more predictable than the general background variability, whose impact on other parts of the system is often significant, and which reveal physical aspects of the system and its response to change. Two specific systems of interest exhibiting multi-scale variability and coherent spatio-temporal patterns are turbulence and climate. Here we will present a new methodology recently developed in our group, called rotated spectral Principle Component Analysis (rsPCA), and demonstrate its ability to robustly identify physically interpretable propagating modes even in the presence of noise. The innovation of rsPCA lies on (1) a wavelet-based implementation of sPCA using the continuous Morlet wavelet as a robust estimator of the cross-spectral matrices with good frequency resolution, and (2) a rotation of the complex-valued eigenvectors to optimize their spatial regularity (smoothness) using a regularization of their spatial Laplacian.  The combination of these two innovations improves interpretability and reduces the sensitivity of the extracted modes to noise and sampling variability as well as unmixes competing modes with similar amplitude within the same frequency band.  In this talk, the theory and application of rsPCA to a wave propagating example and to extracting dynamical modes from climate observations will be discussed.

Category
Start date
Tuesday, Sept. 29, 2020, 3 p.m.

Share