AN UNUSUALLY ACTIVE TROPICAL CYCLONE SEASON IN THE MARSHALL ISLANDS

The Marshall Islands experienced a very active tropical cyclone season during 1997 as 18 tropical cyclones formed near or moved through the area. These storms (that developed east of 160E) represented more than half of the 33 tropical cyclones reported in the Western North Pacific Ocean(WPAC). This was also a year for very intense storms with 11 of the WPAC storms becoming super typhoons. This set a new record for the most super typhoons (STY) in any season. The previous record was seven, and the 37-year average is four per year. As a result of this record, the Joint Typhoon Warning Center designated 1997 as "The Year of the Super Typhoon"(JTWC, 1998).

Nine of the tropical cyclones that formed in the area east of 160E became super typhoons. Two of the Super Typhoons, Oliwa (02C) and Paka (05C), originated in the Central Pacific. The other seven developed near the Marshall Islands. This list included Super Typhoons: Nestor (07W), Winnie (14W), Bing (19W), Ginger (24W), Ivan (27W), Joan (28W), and Keith (29W) (Table 1). The genesis region for tropical cyclones during 1997 was well east of the normal area and was closer to the equator than normal. Twelve tropical cyclones formed south of 10N.

The 1997 WPAC tropical cyclones were stronger than normal with many being fast developers. The total number of all WPAC typhoons was 23 (five above the WPAC annual average). The total number of tropical storms and tropical depressions was eight (seven below the annual average).

Most signs suggest the ENSO event was the prime contributing factor in producing the record number of Super Typhoons. Earlier studies have shown there is a tendency for tropical cyclone activity to increase in the vicinity of the Marshall Islands during ENSO events (Lander, 1994). This increase has been attributed to an eastward shift of the genesis area in response to the warmer than normal water temperatures associated with an ENSO event. 1997 proved to be fairly typical for an ENSO year with the following events being observed. 1.) A pronounced eastward shift of the tropical cyclone genesis region. 2.) The formation and subsequent development of tropical cyclones closer to the equator than normal. 3.) The occurrence of westerly wind bursts near the equator (Phoebus, 1993). Westerly winds were observed near the equator with several storms includingPaka, Isa, Keith, and Winnie. 4.) The formation of twin cyclones in the Northern and Southern hemispheres (Keen, 1982). A cross-equatorial tropical cyclone pair was observed with the development of Paka and Pam in December.

1)  TS Hannah (01W)				50 kts2)  TS Jimmy (03W)				55 kts3)  TS Kelly (04W)				45 kts4)  TY Marie (06W)				90 kts5)  STY Nestor (07W)		               140 kts6)  TY Peter (09W)				65 kts7)  STY Winnie (14W)			       140 kts8)  TY Yule (15W)				65 kts9)  TD (16W)					30 kts10) STY Bing (19W)			       135 kts11) STY Oliwa (02C)			       140 kts12) TY David (21W)				95 kts13) TS Ella (23W)				40 kts14) STY Ginger (24W)			       145 kts15) STY Ivan (27W)			       160 kts16) STY Joan (28W)			       160 kts17) STY Keith (29W)			       155 kts18) STY Paka (05C)			       160 kts

Table 1. List of 1997 Tropical Cyclones that developed east of 160 E with maximum sustained winds.

The Tropical Cyclone season started early in the WPAC with Tropical Storm (TS) Hannah forming on 11 January. Several of the 1997 tropical cyclones had a significant impact on the weather in the Marshall Islands. The first impact was from STY Isa in April. Isa produced strong westerly winds and large waves which caused some damage and coastal flooding on Ebeye in the Kwajalein Atoll (and other exposed coastal areas). The second was from TS Kelly in early May. Kelly brushed the area with heavy showers and strong westerly winds. Typhoon(TY) Marie also brought gusty westerly winds to the Atolls during late May. STY Winnie (14W) developed in early August and went on to become a Super Typhoon on 11 August. Winnie brought a return of strong and gusty westerly winds across the Kwajalein Atoll in mid and late August. TY David (21W) developed just north of Kwajalein in early September producing heavy showers and gusty westerly winds over the Atoll in late September.

Between September 22 and October 27, Kwajalein experienced a series of rapidly developing cyclones which eventually became super typhoons. The list included Ginger, Ivan, Joan, and Keith. All of these storms intensified rapidly after passing Kwajalein. All reached typhoon and super typhoon status between 160E and 130E. STY Ginger produced strong westerly winds across the Atoll into late September. STY Joan and STY Keith also brought strong gusty winds to the area in late October. The season ended with Super Typhoon Paka (which proved to be the most damaging storm of the year).

Super Typhoon Paka (05C) formed in the Central Pacific near 6N 166W on 29 November. A westerly wind burst along the equator led to the formation of twin tropical cyclones in both hemispheres. Paka (05C) formed in the Northern Hemisphere and Pam (07P) formed in the Southern Hemisphere. Paka moved westward becoming a tropical storm on 2 December. The storm continued westward at 8 to 14 knots with winds increasing to 55 knots as the storm reached near 8N 168W on 4 December. Paka then weakened as it turned on a more west-southwestward track. Peak winds were only 35 knots as the storm neared the dateline on 6 December.

After crossing the dateline, Paka again strengthened with peak winds reaching 60 knots on 7 December near 7.5N 178E. Paka then weakened again on 8 December. The storm continued moving westward into an area of warmer water with sea surface temperatures gradually increasing from 28 to near 30 degrees Celsius. This track also took the storm into an area with more favorable outflow at the 250-mb level.

Meanwhile, a strong low-level anticyclone to the north of the storm had shifted southward, increasing the pressure gradient and winds just north of the developing storm. A westerly wind burst also developed near the equator. This combination of events contributed to the subsequent intensification of Paka near the Marshall Islands. Paka moved west and became a typhoon near 6.8N 171E on 10 December. During the period 10-12 December, Paka intensified rapidly with peak winds reaching 115 knots.

The Kwajalein DWSR-93S Doppler Radar captured this two-day period of intensification (Burgett and Trapp, 1998). During this period, the storm was in a constant state of change. Several mesoscale features were observed including an eyewall replacement cycle, a significant increase in convection, and the convergence of several wind speed maxima into the storm center (Burgett and Trapp, 1998).

By 12 December, the outflow pattern near Paka was less favorable as the storm approached a mid-latitude trough. Peak winds dropped to 95 knots on 13 December. After passing the trough, peak winds increased to 140 knots on 15 December (with Paka continuing west-northwest). Peak winds decreased briefly to 125 knots early on 16 December, before reaching 160 knots on 18 December (just after devastating Guam).

The storms that formed near the Marshall Islands proved to be fast developers. The time required for intensification of these storms from tropical storm intensity (35kts) to typhoon intensity (65 kts) and then to super typhoon intensity (130 kts) was relatively short. The average time for these storms to increase from 35 to 65 knots was 40 hours. The average time to increase from 65 to 130 knots was 39 hours. Four of the Super Typhoons increased from 65 to 130 knots in 36 hours or less.

The most rapid intensification took place over the ocean area where water temperatures were warmest. The ENSO event provided abnormally warm water temperatures across much of the Pacific (from the dateline eastward between the equator and 10N) through most of the season. The unusually warm ocean water played a very important role in maintaining the very active ITCZ across the Western and Central Pacific. All disturbances in the vicinity of the Marshall Islands were generated in, or developed near, the ITCZ (including both Paka and Oliwa). Storms intensified most rapidly in the area between 165E and 150E.

Paka, Keith, and Winnie all intensified over 30 degree Celsius water. The sea surface temperatures (SST) along the path of Paka (from near the dateline to the vicinity of the Marshall Islands) increased by two to three degrees Celsius. This SST gradient combined with favorable surface and upper level dynamics to enhance tropical cyclone development near the Marshall Islands. In addition, the persistence of a very active ITCZ and a large cyclonic gyre in the vicinity of the Marshall Islands combined with the warm ocean temperatures to create an environment favorable for tropical storm development.

Paka intensified rapidly as it approached Guam (even as it was accelerating). Paka's westward motion was enhanced by the large anticyclone to the north as it intensified in response to the passage of a strong jet stream. In this case, the dynamics in the mid-latitudes had a major impact on the motion of Paka. The intensification of Paka in the vicinity of the Marshall Islands was also enhanced by the large anticyclone to the north that appeared to be shifting westward with the storm. The westward motion of the anticyclone helped to maintain a favorable outflow pattern for the intensification process. The upper-level flow was primarily away from the storm center to the northwest then turning northward near an approaching mid-latitude trough. This same type of feature was observed during the 1994 Central Pacific Hurricane Season (which also established a record for the greatest number of Category Five storms in the Central Pacific).

The increased pressure gradient produced by a stronger than normal anticyclone in the mid-latitudes combined with the warm SST and the lower surface pressure in the active ITCZ to enhance the development of tropical activity. This helped to increase available thermal energy to provide the necessary fuel for the rapidly developing tropical cyclones. In the case of Paka, the increasing pressure gradient and resultant surface wind speed maxima north of the storm combined with the westerly surge (to the south) to enhance near surface vorticity and low level convergence.

The sea surface temperatures in the vicinity of the equator in the Central Pacific were 4 to 5 degrees Celsius above normal throughout much of the 1997 tropical cyclone season. There is no doubt that this extremely warm water played an important role in setting the stage for a very active tropical cyclone season in the eastern portion of theWPAC. The warm SST anomaly extended a very active ITCZ well into the Central Pacific area. This had the effect of expanding the area of ocean surface capable of generating tropical cyclones, thus allowing storms to develop further east than normal. This shifted the genesis area east and south. As the tropical cyclones developed in the eastern portion of the WPAC area, much closer to the equator than normal, they had much more time to intensify and grow over the warm ocean before moving out of the tropics and dissipating.