Detecting Surface Kuroshio Front in the Luzon Strait From Multichannel Satellite Data Using Neural Networks

An objective classification method is developed to distinguish the water masses of Kuroshio and South China Sea (SCS) by using an artificial neural network (ANN). Sea surface temperature (SST) and ocean-color data obtained from the Moderate Resolution Imaging Spectroradiometer in two specified areas to the east and west of Luzon, representing the Kuroshio and SCS waters, respectively, are used to train, validate, and test the ANN model. The water masses of Kuroshio and SCS can be distinguished correctly with a high success rate of over 99%. The model is then applied to the Luzon Strait, and the result of water mass classification agrees well with the temperature-salinity characteristics derived from a cruise in May and June of 2006. The performance is good in summertime when the SST or ocean color has a rather uniform spatial distribution and the traditional method of front detection by using a threshold value is inappropriate.