Multivariate analysis of multiscale periodicity in North American annual mean sea level based on wavelet coherence and multiple wavelet coherence

Abstract

Abstract The relationships between sea levels and single climate indexes have been widely explored. However, sea level is controlled by multiple climate factors simultaneously with differences among places and time scales. Despite this, few studies have addressed the relationships between sea levels and multiple climate indexes. Here, the interrelations between the annual mean sea level (AMSL) and individual climate indexes and combinations of climate indexes were characterized by wavelet coherence (WTC) and multiple wavelet coherence (MWC). The results showed that six climate indexes had a significant correlation with AMSL, among which the Atlantic Multidecadal Oscillation (AMO) had the least significant correlation, but Niño3.4 had the significant correlation. The average significant coherence area (PASC) value for Niño3.4 was 14.03% and its average wavelet coherence (AWC) value was 0.79. By combining climate indexes, the average values of PASC and AWC can be significantly increased. The average PASC and AWC values of the five climate index combinations were the largest, 95.87% and 0.965, respectively, followed by four, three, and two index combinations. A single climate index is not sufficient to explain sea level change in North America. Sea level changes in North America require between three and five climate indexes to explain, depending on the region. By calculating PASC and AWC values, this study provides the possibility to understand the impact of the combined effects of multiple climate indexes on AMSL, is of great help to screen the best predictor of sea level, and provides a new method to reveal the complex mechanism of sea level change.