Sensitivity of the Interannual Kuroshio Transport Variation South of Japan to Wind Dataset in OGCM Calculation

Numerical experiments were carried out using OGCM (Ocean General Circulation Model), MOM2.2 (Modular Ocean Model Ver. 2.2), over realistic topography data, ETOPO5 (Earth Topography - 5 Minute), to investigate the interannual variability of the Kuroshio transport in 1960–2000 south of Japan; 1) the PN line located off the East China Sea, and 2) the ASUKA (Affiliated Surveys of the Kuroshio off Cape Ashizuri) line located off Cape Ashizuri. We adopted two wind datasets as driving forces of the OGCM: 1) the NCEP/NCAR (National Centers for Environmental Prediction/National Center for Atmospheric Research) reanalysis monthly mean wind stress data, and 2) the ECMWF (European Centre for Medium-range Weather Forecasts) daily wind data. In the ECMWF experiments we replaced the NCEP/NCAR data only in 1979–1993 because of the availability of the data. The OGCMs and observation basically agree on the temporal variation patterns of the transports until 1986 on the PN line with correlation coefficients of about 0.6. During the 1990s, when data were collected on the ASUKA line, the NCEP/NCAR experiments give lower correlation coefficients (less than 0.3), on both PN and ASUKA lines, while the ECMWF experiments have a higher value on the ASUKA line (0.5). One of the reasons for the disagreement between the observations and OGCMs during the 1990s might arise from the NCEP/NCAR data. An additional analysis of a wind-driven circulation was performed to examine the sensitivity of integrated Sverdrup transport along the western boundary to the propagation speed of a baroclinic Rossby wave, which is varied by stratification. A variation of the stratification, which might be induced by variability of air-sea heat and freshwater fluxes, cannot be a main cause of the disagreement. This revised version was published online in July 2006 with corrections to the Cover Date.