Begin Main Content

Description

The Multivariate Optimal Interpolation (MvOI) uses static multivariate statistics calculated from an ensemble of ocean runs. All components of the ocean state are corrected during assimilation. A major improvement in the MvOI and EnKF has been the introduction of an algorithm to filter spurious large lag information from the background error covariances due to small ensemble sizes. In the filtering algorithm, documented in Keppenne and Rienecker (2001), the resulting error covariances have compact support.

Testing of the MvOI to date has concentrated on the Pacific basin. Initial results are presented in Rienecker et al. (1999), Borovikov and Rienecker (2002) , and Borovikov et all (2004).

Borovikov A., M. M. Rienecker, C. L. Keppenne, and G. C. Johnson, 2004: Multivariate error covariance estimates by Monte-Carlo simulation for assimilation studies in the Pacific Ocean. Submitted to Mon. Wea. Rev. (ftp://nsipp.gsfc.nasa.gov/pub/assim/ayb/covar.pdf)