Statistical inversion of oceanographic tracer data

Ian McKeague

Department of Biostatistics
Columnia University

This talk will discuss open issues and new research directions in the development of statistical methods to improve our understanding of deep ocean circulation. A Bayesian inversion approach will be discussed in detail, and illustrated by analyzing tracer data collected from oceanographic field work in the South Atlantic. Further research is needed to refine the combination of physical and statistical models that underlie the inversion, to assimilate other data sources (e.g. Lagrangian data from floats) into the inversion, and to find better ways of assessing the uncertainties involved.

Last Update: February 11, 2005