Estimating the accelerated sea level rise along the Korean peninsula using multiscale analysis

An accelerated sea level rise has recently been detected in many parts of the world. It has been acknowledged that isolation of the acceleration in the long-term is difficult due to the fact that there is an embedded low frequency associated with the climate variability. In light of this, this paper aims to provide a new multiscale assessment tool for the estimation of sea-level acceleration for tide gauge data along the Korean peninsula. This study employed the wavelet transform (WT) and the Ensemble Empirical Mode Decomposition (EEMD) methods to better isolate time varying oscillation and estimate reliable sea-level acceleration of the trend. This paper first examined the WT and the EEMD methods to fully understand their capability using synthetic data, and then further investigated the acceleration rates of the historical tide gauge records. It was found that the EEMD showed more reliable estimates than the WT in terms of the trend error. The smooth increase in the trend at all stations, except Wido, and acceleration in southern coastlines of South Korea was identified through the EEMD. Future work must be focused on developing a Bayesian-based nonstationary frequency analysis model in which the estimated acceleration rates are used to evaluate the risk associated with sea-level rise.