Tropical Intraseasonal Air-Sea Exchanges during the 1997 Pacific Warming

The Madden Julian Oscillations (MJO) and associated westerly wind (WW) events account for much of the tropical intraseasonal variability (TISV). The TISV has been suggested as an important stochastic forcing that may be one of the underlying causes for the observed irregularities of the El Nino-Southern Oscillation (ENSO). Recent observational studies and theories of interannual to interdecadal-scale variability suggest that ENSO may arise from different mechanisms depending on the basic states. The Pacific warming event of 1997, being associated with a period of strong MJO and WW events, serves as a natural experiment for studying the possible role of TISV in triggering an ENSO event. We have performed a combined statistical and composite analysis of surface WW events based on the assimilated surface wind and sea level pressure for the period of 1980-1993, the SSM/I wind for the period of 1988-1997, and OLR. Results indicates that extratropical forcing contribute significantly to the evolution of MJO and establishment of WW events over the Pacific warm pool. Following the major WW events, there appeared an eastward extension of equatorial warm SST anomalies from the western Pacific warm pool. Such tropical-extratropical interaction is particularly clear in the winter of 96-97 that leads to the recent warming event in 1997/98. From the above discussion, our current study on this subject is based on the hypothesis that 1) there is an enhanced air-sea interaction associated with TISV and the northerly surges from the extratropics in the initial phase of the 97/98 warming event, and 2) the relevant mechanisms are functions of the basic state of the coupled system (in terms of SST distribution and atmospheric mean circulation) that varies at the interannual and interdecadal time scale. We are analyzing the space-time structure of the northerly surges, their association with air-sea fluxes and upper ocean responses during the period of September 1996 to June 1997. The estimate of daily values of latent heat fluxes is based on NSCAT wind, SST, and ECMWF surface air temperature and SSM/I water vapor data (Chou et al. 1997). To understand the relevant mechanisms, we will analyze the origin of the northerly surges in terms of atmospheric instability associated with the extratropical circulation, and the mutual influence between the tropical heating and the extratropical circulation. In this meeting, we will report the analysis addressing the first part of the above hypothesis.