摘要

We here propose a new statistical method to interpret nonstationary running correlations by decomposing them into a stationary part and a first-order Taylor expansion approximation for the nonstationary part. Then, this method is applied to investigate the nonstationary behavior of the El Nino-Southern Oscillation (ENSO)-East Asian winter monsoon (EAWM) relationship, which exhibits prominent multidecadal variations. It is demonstrated that the first-order approximation of the nonstationary part can be expressed to a large extent by the impact of the nonlinear interaction between the Atlantic Multidecadal Oscillation (AMO) and ENSO (AMO*Nino3.4) on the EAWM. Therefore, the nonstationarity in the ENSO-EAWM relationship comes predominantly from the impact of an AMO modulation on the ENSO-EAWM teleconnection via this key nonlinear interaction. This general method can be applied to investigate nonstationary relationships that are often observed between various different climate phenomena.