Sampling Strategies and Model Assimilation of Altimetric Data for Ocean Monitoring and Prediction

摘要

Numerical forecasts using simulated altimeter data are generated to examine assimilation of altimeter-derived sea surface heights into numerical ocean circulation models. A one-layer reduced gravity primitive equation circulation model of the Gulf of Mexico is used; the Gulf of Mexico is chosen because of its amenability to modeling and the ability of low vertical mode models to reproduce many of the observed dynamic features of gulf circulation. The simulated data are obtained by flying a hypothetical altimeter over the model ocean and sampling the model sea surface in a manner similar to a real altimeter. The data are used to examine spatial sampling requirements for accurate resolution of oceanic eddies and, in forecast mode, the assimilation of asynoptic altimeter data into numerical models. Results indicate that, for a stationary circular eddy, approximately two tracks (either ascending or descending) across the eddy are sufficient to ensure adequate spatial resolution. An irregularly shaped eddy may require 3 or 4 tracks. The study also reveals that, if track spacing is sufficient to resolve the height field of an eddy, the along-track geostrophic velocity component is determined with equivalent accuracy to that of the cross track component.