Disentangling the spatially combined and temporally lagged influences of climate oscillations on seasonal droughts in the East Asian monsoon influenced Poyang Lake Basin

Large-scale climate oscillations are the main forcings affecting regional meteorological droughts and being relevant to sources of their predictability. However, the physical mechanism of atmospheric teleconnections with respect to regional droughts is still not fully understood. In this study, a univariate-to-multivariate analysis framework is proposed to disentangle the spatially combined and temporally lagged effects of multiple oceanic-atmospheric oscillations on meteorological droughts at regional scale. Our study focuses on the largest freshwater lake basin of China, the Poyang Lake basin (PLB). Pearson's correlation coefficient and cross-wavelet transform are used to analyze the pair-wise linear and non-linear correlations between droughts and each climate oscillation. Random forests model is used to reveal the combined influences of multiple climate oscillations. The associated atmospheric mechanism for the identified combination of climate indices with changing lags is explored by performing composite analysis. Regarding the spatially combined influences, the concurrence of El Niño-Southern Oscillation (ENSO) and North Atlantic Oscillation (NAO) are the most important drought precursors. Regarding the temporally lagged influences, ENSO with lag of 11 months and NAO with lag of 2–3 months trigger meteorological droughts. The combined effect of preceding winter El Niño and late-summer negative NAO is the primary cause for triggering autumn droughts. The positive Eurasian teleconnection pattern, triggered by ENSO and NAO and favorable for anomalous northerly currents, is the main drought-prone circulation pattern for the PLB. These findings contribute to improved understanding of joint effects of lagged teleconnections for meteorological droughts, which could eventually lead to more skillful seasonal drought forecasting.