INTRODUCTION
Recently evidence has been presented that the North Pacific atmosphere/ocean system fluctuates not only with periods of 2〜6 years in response to ENSO but also with decadal timescales (e.g., Nitta and Yamada 1989; Trenberth 1990; Tanimoto et al. 1993; Nakamura et al. 1997 and references therein). A marked decadal and interdecadal climatic event (DICE) occurred around 1976. Since then until the late 1980's, SST tended to be below normal over the North Pacific, accompanied by the weakened Aleutian low and intraseasonal fluctuations including blocking (Nakamura 1996) and by the southward-shifted stormtrack (Trenberth and Hurrel 1994).
Still, we do not fully understand the cause and mechanisms of the North Pacific (NP-) DICEs. It has been hypothesized that cumulus convection over the tropical Pacific activated by SST warming around 1977 (Graham 1994; Nitta and Kachi 1994) triggered a particular tele-connection pattern in the midlatitude atmosphere, which lowered SST over the North Pacific. This is analogous to the “atmospheric bridge” (Lau and Nath 1994), a scenario proposed to explain the observed response of extra-tropical SST to ENSO. If this scenario is really operative, decadal SST anomalies should be basin-wide. Yet, they may be confined in a single oceanic gyre if the extratropical oceanic processes play a primary role (Nakamura and Yamagata, 1998a). Indeed, recent global ocean/atmosphere GCM simulations suggest that the coupled system in the extratropical North Pacific may excite interdecadal variability by itself (Latif and Barnett 1994; Robertson 1996).
In this paper, we attempt to clarify the importance of midlatitude atmosphere/ocean interactions in NP-DICEs by analyzing observational data. Its brief summary has been given in Nakamura et al. (1997; cited as NLY97), and Nakamura and Yamagata (1998b; cited as NY98) should be referred to for a more complete description
DATA
