Hybrid Data Assimilation Techniques and Applications

By Erik Van Vleck

University of Kansas

You must login to access the full resource.

Published on

Abstract

Data assimilation provides a framework for incorporating data or observations into models for both state space and parameter estimation. In this talk we discuss the development of some hybrid data assimilation techniques based upon shadowing refinement. These techniques employ dimension reduction motivated by time dependent stability theory and are designed in an equation free form for application from conceptual to large stand-alone models. We discuss some of the applications we are investigating using these techniques and illustrate their effectiveness on some model problems.

MCRN Colloquium on March 25, 2016

Submitter

Monica Romeo

RENCI, UNC Chapel Hill

Tags