Nonlinear and Stochastic Problems in Atmospheric and Oceanic Prediction (17w5061)

Arriving in Banff, Alberta Sunday, November 19 and departing Friday November 24, 2017

Organizers

(University of Northern British Columbia)

(Indiana University)

Wansuo Duan (Chinese Academy of Sciences)

Objectives

It is proposed to hold a Banff Workshop on “Nonlinear and Stochastic Problems in Atmospheric and Oceanic Prediction”. This broadly interdisciplinary workshop aims to: (1) bringing together nonlinear and stochastic researchers from Mathematics/Statistics, and environmental scientists from Atmospheric Science, Oceanography, and Climate Science, so that environmental scientists will learn the latest nonlinear and stochastic methods, while mathematicians and statisticians will learn about challenging environmental problems for which new nonlinear and stochastic methods need to be developed; (2) crystallizing, through seminars and discussions, new lines of research and research collaborations among the workshop participants; and (3) enhancing the training of junior scientists, by balancing participants between mature scientists and post-doctoral researchers. The proposed workshop will contribute in promoting, enhancing, and stimulating cross-continental research interactions and collaborations at the cross-intersection of nonlinear stochastic dynamics, nonlinear PDEs, and atmospheric and oceanic dynamics. It is anticipated that the Workshop will contribute significantly to the advancement of international atmospheric and oceanic prediction and applied mathematics communities.



The specific problems the workshop will try to address include the following: (i) How can we estimate the model states for high-dimensional nonlinear and non-Gaussian systems, and which nonlinear and stochastic method will be best for the particular type of problem? (ii) What new nonlinear optimal methods or strategies need to be developed to detect the optimal prediction error growth? (iii) What new measurement metrics should be developed to optimally quantify the prediction utility for a nonlinear and stochastic system? (iv) How can we measure and differ the contribution of different error sources to predictability, and further reduce the errors? (v) How should nonlinear and stochastic methods complement the dynamical models used in the atmospheric and oceanic prediction community? (vi) How can full advantage be taken, in atmospheric and oceanic sciences, of the powerful tools that have been developed by mathematicians in the framework of the theory of dynamical systems (dynamic transitions, Lyapunov exponents and vectors, fluctuation-dissipation theorem, invariant and parameterizing manifolds).