Extreme value prediction of current profiles in the South China Sea based on EOFs and the ACER method

Abstract

Empirical orthogonal functions (EOF) and the average conditional exceedance rate (ACER) method are used to perform extreme value prediction of the current profile based on prototype current data of the Liuhua (LH11-1) sea area in the South China Sea. Multi-year extreme profiles are calculated using the drag force as the objective function and are compared with the extreme envelopes of the current speed in each layer. The results indicate that the spatial correlation of the current is taken into account in the proposed method. It is important and necessary to consider the interlayer correlation in an extreme value model of the current profile. Finally, the influence of the mode order of the EOF on the prediction results of the extreme profile is discussed, and a selection method for the mode order is proposed. The proposed EOF-ACER method provides a new strategy for the analysis and prediction of extreme values of the current profile.