-----------------------------------------------------------

Applied Math Colloquium

Friday, Sep 7, 2012, 11:30 AM
Cullimore Lecture Hall, Lecture Hall II
New Jersey Institute of Technology

Optimal Control in Data Assimilation

Richard Moore

Department of Mathematical Sciences, NJIT

Abstract

Data assimilation refers to the combination of field measurements with a physical model to produce an estimate of the state of a physical system, such as the spatiotemporal distribution of climate variables or the velocity field in the surface layer of the ocean. These systems are often very high-dimensional relative to the observational data available, and controlling the observers can be an important tool in improving the state estimate. We discuss recent work that explores the use of optimal sampling of an oceanic velocity field in the context of Lagrangian data assimilation.

This is joint work with Damon McDougall and Christopher Jones.