KUROSHIO EXTENSION DYNAMICS FROM SATELLITE ALTIMETRY AND A MODEL SIMULATION

Altimeter data from the Geosat Exact Repeat Mission (ERM) are analyzed with the aid of a simulation from an eddy-resolving primitive equatio n model of the North Pacific basin in the region of the Kuroshio and K uroshio Extension. The model domain covers the Pacific Ocean north of 20 degrees S and has a resolution of 0.125 degrees latitude and 0.176 degrees longitude. The model is synoptically driven by daily 1000-mbar winds from the European Centre for Medium-Range Weather Forecasts (EC MWE) which encompass the Geosat time period. Model output is sampled a long Geosat ground tracks for the period of the ERM. Additionally, the model and the Geosat data are compared with climatological hydrograph y and satellite IR frontal position analyses. Analyses compared includ e maps of sea surface height (SSH) mean and variability, eddy kinetic energy (EKE), seasonal transport anomaly, and time-longitude plots of SSH anomaly. The model simulation provides annual mean SSH fields for 1987 and 1988 which reproduce the four quasi-permanent meanders seen i n hydrographic climatology (cyclonic at 138 degrees E and anticyclonic at 144 degrees E, 150 degrees E, and 160 degrees E). These are linked to the bottom topography. In the model simulation, Geosat altimeter d ata, and climatology, we observe four peaks in SSH variability associa ted with meander activity and two peaks in EKE, with the strongest abo ut 3200 cm(2) s(-2) along the mean Kuroshio path in the Geosat data. T he local maxima in SSH variability tend to occur where relatively stro ng, topographically steered meridional abyssal currents intersect the zonally oriented Kuroshio Extension. Westward propagation of SSH anoma lies at phase speeds of 2 to 3 cm s(-1) in the region east of 155 degr ees E is observed in the model simulation and Geosat observations. A l ate summer maximum in the upper ocean transport anomaly of the Kuroshi o Extension is inferred from changes in the cross-stream differential in SSH from the simulation and Geosat observations.