Investigating the spatiotemporal variations of extreme rainfall and its potential driving factors with improved partial wavelet coherence

Abstract

Extreme rainfall can be affected by various climatic factors such as the large-scale climate patterns (LCPs). Understanding the changing LCPs can improve the accuracy of extreme rainfall prediction. This study explores the variation trend of extreme rainfall in the middle and lower reaches of the Yangtze River Basin (MLRYRB) and the telecorrelation with four LCPs, namely WPSHI (Western Pacific Subtropical High Index), EAMI (East Asia Monsoon Index), ENSO (El Niño-Southern Oscillation) and PDO (Pacific Decadal Oscillation), through modified Mann-Kendall (MMK) analysis, Pearson correlation coefficient, wavelet coherence analysis (WTC) and improved partial wavelet analysis (PWC). Previous studies have ignored the interdependence between these climate indices when analyzing their effects on precipitation. This study introduces the improved PWC, which can remove the correlations between them and reveal the influence of a single LCP. The results show that: 1) extreme rainfall in the MLRYRB has an obvious increasing trend and has a significant correlation with the LCPs; 2) the LCPs have a significant cyclical relationship with extreme rainfall, which can be significantly affected by the intergenerational variation of the LCPs; and 3) the improved PWC can accurately reveal the influence of a single LCP. EAMI is the main influencing factor in the 1-year cycle, while WPSHI is the main influencing factor in the 5-year cycle. ENSO and PDO can always influence extreme rainfall by coupling WPSHI or EAMI.