Effects of stretching on maintaining the Kuroshio meander

Abstract

Using a two‐way nested‐grid ocean general circulation model driven by a repeat annual cycle forcing, the Kuroshio path variation is examined by varying the strength of wind forcing. For climatological or 10% weaker wind forcing, both large‐meander (LM) and non‐large‐meander (NLM) paths alternately appear, with each type of path continuing for a few years to a decade. This timescale and path transition processes are generally consistent with observations. For the LM path, the main balance in the depth‐integrated vorticity equation for the upper ocean is shown to be between the beta, advection, and stretching terms. The stretching term is comparable to and has the same sign as the beta term at the western side of the meandering part, indicating that the stretching term has the effects of shortening the wavelength and stabilizing the LM path. The stretching term is mainly determined by downwelling caused by crossing of the upper and deep flows. Contrary to the above two cases, only the NLM path appears for 10% stronger wind forcing. It is suggested that the strength of climatological wind forcing is near the upper limit that allows the LM path to occur. In this study, the control experiment is carried out by choosing parameterizations and parameter values to reproduce the Kuroshio path variation as realistic as possible. Therefore, the results from the two sensitivity experiments are expected to represent a realistic response of the Kuroshio path variation.