Interdecadal climate variability over the North Pacific is investigated using a reconstructed 298-year-long annual Pacific Decadal Oscillation Index during 1700–1997. Here, we show that its interdecadal component can be decomposed into the bi-decadal (~ 20 years), the tri-decadal (~ 30 years), and the multi-decadal variations (~ 60 years) through bandpass filtering. The phase reversals of the bi-decadal and the tri-decadal variations were synchronized throughout the data, while the synchronization of the bi-decadal and the multi-decadal variations was maintained during 1780–1880 and 1920–1997. Monte Carlo simulation reveals that this synchronous temporal structure is statistically significant and could not occur from random processes. We further demonstrate that these three phase-locked variations play an important role in the regime shifts in the North Pacific, both during the data period (1700–1997) and thereafter. The bi-decadal variation, which is self-sustained through the North Pacific mid-latitude air–sea interaction, is also found to be possibly linked to the lunar nodal cycle. Moreover, the tri-decadal and the multi-decadal variations, whose periods are approximately 1.5 and 3 times longer than that of the bi-decadal variation, may be linked to the bi-decadal variation through nonlinear process. This simple relationship could largely contribute to understanding and predicting the interdecadal climate variability.

使用重建的298年太平洋太平洋年代际振荡指数（1700-1997年），调查了北太平洋年代际气候变化。在这里，我们表明，它的年代际分量可以通过带通滤波分解为双年代（〜20年），三年代（〜30年）和多年代变（〜60年）。在整个数据中，双年代际变化和三年代际变化的相位反转是同步的，而在1780-1880年和1920-1997年间则保持了双年代际变化和多年代际变化的同步。蒙特卡洛模拟表明，这种同步时间结构具有统计学意义，不可能发生于随机过程。我们进一步证明，在数据时期（1700年至1997年）及其后，这三个锁相变化在北太平洋的政权转移中起着重要作用。通过北太平洋中纬度海-海相互作用而自我维持的双年代际变化，也可能与月球节拍周期有关。此外，周期为双年代际变化的大约1.5倍和3倍的三年代和多年代变化可以通过非线性过程与双年代变化联系起来。这种简单的关系在很大程度上有助于理解和预测年代际气候变化。它是通过北太平洋中纬度海-海相互作用而自我维持的，也可能与月球节拍周期有关。此外，周期为双年代际变化的大约1.5倍和3倍的三年代和多年代变化可以通过非线性过程与双年代变化联系起来。这种简单的关系在很大程度上有助于理解和预测年代际气候变化。它是通过北太平洋中纬度海-海相互作用而自我维持的，也可能与月球节拍周期有关。此外，周期为双年代际变化的大约1.5倍和3倍的三年代和多年代变化可以通过非线性过程与双年代变化联系起来。这种简单的关系在很大程度上有助于理解和预测年代际气候变化。