Fluctuation in the distribution of low-salinity water in the North Equatorial Current and its effect on the larval transport of the Japanese eel

Surface water in the North Equatorial Current (NEC) is composed of southern low-salinity water diluted by precipitation to less than 34.2 psu and northern, high-salinity tropical water greater than 34.8 psu. Analyses of 27-year historical data, observed in winter and summer along the longitude 137 degreesE by the Japan Meteorological Agency, shows that an obvious salinity front (34.5 psu) generated by the two water masses was usually located around 15 degreesN. However, the salinity front has been moving northward during the past three decades. El Nino/Southern Oscillation (ENSO) affected salinity in the surface layer, while temperature changed in the middle layer. The salinity front sometimes moved southward, mainly south of 5 degreesN, and the movement was well correlated with the southern oscillation index (SOI). Because precipitation at Yap (9.5 degreesN, 138.1 degreesE) fluctuated with SOI, this spike-like southward movement of the salinity front was probably affected by reduction of low-salinity water during El Nino in the north-western Pacific Ocean. However, ENSO only induced such large southward movements of the salinity front when the time lag between the low precipitation and low SOI was short (within four months). This salinity front is quite important for long-distance migrating fish such as the Japanese eel because the eels spawn just south of the salinity front in the NEC. This behaviour suggests that the movement of the salinity front associated with ENSO may control the success of larval transport from the spawning ground in the NEC to the nursery ground in East Asia. In fact, catch of the Japanese eel larvae in Japan was well correlated with fluctuation of SOI and the location of the salinity front, and lower catch occurred during El Nino. The salinity front has moved from 13 degreesN to 17 degreesN during the past three decades. Considering that conditions of larval transport are worse north of 15 degreesN, we suggest that decadal-scale linear decrease of glass eel catch during the past three decades also can be explained by the displacement of the salinity front.