The BO is an interesting phenomenon by its own right, but also is important in association with climatic regime shifts over the North Pacific. In the last decade, a number of analyses showed the significant changes of the physical environments in association with the 1970s climatic regime shift (e.g., Nitta and Yamada 1989; Trenberth 1990; Graham 1994; Tanimoto et al. 1993; Trenberth and Hurrell 1994; Miller et al. 1994, 1998, Polovina et al. 1995, Lagerloef 1995, Yasuda and Hanawa 1997; Nakamura et al. 1997; Zhang and Levitus 1997; Schneider et al. 1999; Tourre et al. 1999; Zhang and Liu 1999; Suga et al. 2000). A climatic regime shift is defined as a transition from one climatic state to another within a period substantially shorter than the lengths of the individual epochs of each climate states (e.g., Minobe 1997). Several analyses indicate that similar regime shifts also occurred in the 1920s and 1940s with alternating polarities (Kondo 1988; Hare and Francis 1995; Dettinger and Cayan 1995; Zhang et al. 1997; Minobe 1997; Mantua et al. 1997). Minobe (1999, 2000) proposed that the quasi-simultaneous phase reversal of BO and a 50-70 year oscillation (referred to as Pacific Pentadecadal Oscillation, PPO) resulted in the three climatic regime shifts in the 1920s, 1940s and 1970s, and that the super position of the BO and PPO gives the Pacific (inter-)Decadal Oscillation, which was proposed by Mantua et al., (1997). The evidence of the PPO was also found in the tree-ring records (Minobe 1997, Ware 1995; Ware and Thomson 1999; Shabalova and Weber 1999). Minobe (1999, 2000) concluded that the BO and PPO are essentially different phenomena because of the difference of their seasonality. Their different spatial structures of the BO and PPO was also documented by Chao (2000). The interpretation that climatic regime shifts are resulted from the resonance between the BO and PPO suggests that the changes of either the BO or PPO can cause significant changes in the nature of the climatic regime shifts. Therefore, the BO is also be of interest from the aspect of the BO influence to the climatic regime shifts.
An interesting feature in the BO is that the amplitude and period has changed through the 20th century; The period is shorter and the amplitude is weaker in the early 20th century, and in the mid and late 20th century the period is longer and the amplitude is stronger. Such changes were first pointed-out by Royer (1989) for time series of Alaska air- and water-temperatures. Minobe (1999, 2000) analyzed large scale climate indices over the North Pacific, which are North Pacific Index (NPI) and Pacific Decadal Oscillation Index, and described that the period is about 1 5 years in the beginning of 20th century and have increased to 20 years from 1 930-50, during which the P0 amplitudes also increased. A question arises at this point. Does the BO has a constant spatial pattern in regardless with the changes in the period? In order to understand time varying structure of the BO, we have developed a Multivariate Wavelet Filter (MWF) technique.
The rest of the present paper is organized as follows; In section 2, the data and methodology of the analysis are described. In section 3.1, the results of the MWF applied to northern hemisphere SLPs are shown, and features captured by the MWF are also examined by different methodologies. In 3.2, we analyze the SST and Land-surface Air-Temperatures (LAT) fields, in order to know whether BO signatures in these fields varied or not in a consistent manner of the BO changes in the SLPs. Summary and discussion will be given in section 4.
