BEHAVIOR OF ONE TO 2 WEEK SUMMERTIME SUBTROPICAL WIND MAXIMA OVER THESOUTH-PACIFIC DURING AN ENSO CYCLE

The behavior of subtropical jet streaks over the Australian-South Paci fic region is investigated for four consecutive 6-month summer seasons , November to April, during the period 1985-89. The study is an extens ion of a recent one by Ko and Vincent (hereafter referred to as KV), i n which jet streaks during the summer season of 1984-85 were examined. Ko and Vincent found that the upper-tropospheric zonal wind behaved i n a quasiperiodic manner, with maxima recurring over the western South Pacific at intervals between one and two weeks. In the present study, wind maxima are found to exhibit a 1-2-week periodic oscillation over the South Pacific in all four years, thus substantiating the finding by KV. In addition, a second set of jet streaks, with a similar period icity, is found over the southern Australian region. It is also seen t hat the mean summertime position of the South Pacific jet and its acco mpanying jet streaks are much farther east during the El Nino event of 1986-87 than they are in the other three years. The jet is also stron gest in 1986-87. Based on point-to-point correlations of the 5-20-day filtered winds, it appears that the South Pacific jet streaks form ove r eastern and sometimes northern Australia, and propagate east-southea stward at about 10 degrees longitude per day ( 12 m s(-1)). The jet st reaks over southern Australia originate over the eastern Indian Ocean and propagate eastward at about the same phase speed as the South Paci fic jets. These two tracks of wind maxima appear to repeat their patte rns approximately every 12 days. In some years, the South Pacific and southern Australian jet streaks appear to be distinctly separate, whil e in other years it seems as though the South Pacific jets are a conti nuation of the jets that originated over the Indian Ocean and propagat ed across southern Australia. Finally, it is found that the jet streak s in both regions are advected eastward in close proximity to upper-tr opospheric midlatitude troughs, which also have a statistically signif icant period between one and two weeks. Consequently, these results co uld impact on medium-range, as well as short-term intraseasonal, forec asting. In the concluding remarks, the similarity is noted between the results found in this study and those found by other authors who have investigated jet streak activity in the Northern Hemisphere. A hypoth esis is suggested for the behavior of the jet streaks diagnosed in thi s study.