OCTOBER 2000 ANNUAL SUMMARY 3695 ANNUAL SUMMARY Atlantic Tropical Systems of 1996 and 1997: Years of Contrasts LIXION A. AVILA,RICHARD J. PASCH, AND JIANN-GWO JIING NOAA/NWS/Tropical Prediction Center/National Hurricane Center, Miami, Florida (Manuscript received 13 August 1999, in ®nal form 14 January 2000) ABSTRACT A total of 62 and 63 tropical waves were counted in the Atlantic from May to November during 1996 and 1997, respectively. These waves led to the formation of 12 of the 13 total number of tropical cyclones in 1996 and only 3 of 7 tropical cyclones in 1997. All of the tropical depressions became tropical storms in 1996 and only one failed to become a named storm in 1997. On average, 62% of the Atlantic tropical depressions develop from tropical waves. These waves contributed to the formation of 92% of the eastern Paci®c tropical cyclones in 1996 and 83% in 1997. Tropical waves and their environment during the 1996 and 1997 seasons are discussed. 1. Introduction satellites and surface reports including 24-h pressure changes from ground stations, ships, and buoys. Radio- Tropical waves not only play a dominant role as pre- sonde data are used to construct time series of vertical cursors to tropical cyclone formation over the Atlantic wind pro®les (vertical time sections) from the sounding and eastern Paci®c Oceans but are also responsible for stations. the modulation of rainfall in the Caribbean Basin (Riehl In general, the analysis of tropical waves begins with 1954). The importance of tropical waves has been em- the examination of satellite animation of cloudiness over phasized in previous updates of this yearly article. Ref- northern Africa. Areas of concentration of cloudiness erences to publications about tropical wave structure and with cyclonic rotation (including vortices) were marked properties are included in Pasch et al. (1998). and tracked in time as they move westward. Satellite This article is concerned with operational tracking of HovmoÈller diagrams were examined to identify the time tropical waves in Atlantic and eastern Paci®c basins. continuity of those areas of cloudiness that exist for The primarily purpose is to tabulate and summarize cer- several days (a timescale greater than that of mesoscale tain weaker synoptic-scale systems of 1996 and 1997, clusters or squalls). These procedures assure the analysts namely tropical waves and tropical depressions. The that they are tracking synoptic-scale systems. Time sec- compilation is then used to update the statistics of such tions of vertical wind pro®les at western African radio- systems, which began in 1967. sonde stations were checked to identify the cyclonic wind shift as the troughs or vortices moved over those 2. Data and analysis stations. Once the tropical waves move into the Atlantic The Tropical Prediction Center/National Hurricane Ocean, they can be identi®ed by (a) areas of increased Center (TPC/NHC) performs surface synoptic analyses convection, (b) cyclonic curvature in low- to mid-level four times daily. The following data/tools are used by clouds, (c) increased concentration of stratocumuli on the analysts: animation of real-time satellite imagery the north side if the waves moved over cooler waters, from Meteosat, Geostationary Operational Environ- and (d) ship and buoy reports. Recently, high-density mental Satellite-8 and -10 (GOES-8 and -10) that, to- winds derived from satellite imagery (Velden 1997) gether, cover the earth from the Indian Ocean westward were also used to analyze and track the tropical waves. to the central Paci®c. Also the analysts use HovmoÈller Satellite HovmoÈller diagrams were examined for the (longitude vs time) diagrams of imagery from all three continuity of the waves as they propagate westward. The analysts can estimate the mean westward motion and period of the waves from the HovmoÈller diagrams. If there is no clear cloud pattern over a region where Corresponding author address: Dr. Lixion A. Avila, Tropical Pre- diction Center/National Hurricane Center, 11691 SW 17th Street, Mi- the waves are expected to propagate through, estimated ami, FL 33165-2149. positions were derived based on the derived mean prop- E-mail: [email protected] agation speed. q 2000 American Meteorological Society Unauthenticated | Downloaded 09/26/21 01:02 AM UTC 3696 MONTHLY WEATHER REVIEW VOLUME 128 When the waves propagated into the Caribbean Sea, in Fig. 3b. Hurricane Danny was the only tropical cy- time sections from the area were used to identify the clone that made landfall in the United States. In addition, passage of the waves and make adjustments to their there was a subtropical storm. A detailed summary of locations if necessary. Twenty-four-hour surface pres- the 1997 hurricane season is included in Rappaport sure changes from those stations were also used for the (1999). analysis. Additional discussion of the wave-tracking methodology is included in Pasch et al. (1998). 1) TROPICAL DEPRESSIONS Figure 1 displays vertical time sections of the wind for Dakar and Trinidad from 15 to 30 August 1996 and The number of tropical depressions in 1997 was sig- 1997. Figure 2 shows sequences of twice-a-day satellite ni®cantly below average with a total of eight. This num- images for 15±31 August of 1996 and 1997. This sample ber is less than half of the long-term average of 19. July represents the period in 1996 when tropical waves were was the most active month with a total of four tropical the strongest and most convectively active. depressions. There was one tropical depression in June, another one in September, and two in October. Re- markably, no tropical depressions developed during Au- 3. Summaries of Atlantic systems gust. Tropical depression ®ve, the only cyclone triggered a. The 1996 Atlantic systems by a tropical wave in July, failed to become a tropical storm and it is the only system described in this paper. Tropical cyclone activity in the Atlantic during 1996 Figure 4 shows the Atlantic tropical depression tracks was above the long-term average of 10 tropical storms for 1997. and six hurricanes. There were 13 named storms, 9 of which became hurricanes (Pasch and Avila 1999). Two tropical storms and two hurricanes made landfall in the 2) TROPICAL DEPRESSION FIVE continental United States, and all but Gustav and Isidore Tropical depression ®ve formed from a tropical wave affected land. Nine of the systems reached hurricane that emerged from the west coast of Africa on 11 July. strength in the deep Tropics between 108 and 208N. The wave moved westward across the tropical Atlantic Moreover, six of them, reached category 3 or higher and ®nally began to show consistent evidence of a cloud intensity on the Saf®r±Simpson hurricane scale (Simp- system center. Deep convection associated with the son 1974). All 13 tropical depressions intensi®ed into wave became concentrated on satellite images, and it tropical storms and 12 of the tropical cyclones devel- was estimated that the disturbance became tropical de- oped from tropical waves. Those tropical cyclones are pression ®ve around 0600 UTC 17 July while centered represented in Fig. 3a by lines attached to the main about 880 km east of Barbados. ``stream'' of waves. The tropical depression moved toward the west-north- west at 6 m s21. The ®rst aircraft reconnaissance ¯ight TROPICAL DEPRESSIONS investigated the depression from a ¯ight level near 455 m around 2100 UTC on 17 July and found a cyclonic A tropical depression is of®cially de®ned as a tropical circulation with 19 m s21 peak winds both north and cyclone (a nonfrontal, synoptic-scale cyclone originat- south of the center. Although satellite intensity estimates ing over tropical or subtropical waters with organized never exceeded 15 m s21, the fact that 19 m s21 winds deep convection and de®nite cyclonic surface wind cir- were measured by the aircraft at ¯ight level to the south culation) in which the maximum 1-min sustained sur- of the center suggests that tropical depression ®ve might face wind is 17 m s21 or less. In terms of tropical de- have been a minimal tropical storm earlier. However pression frequency, 1996 was below average with a total because of the reduction of wind speeds from ¯ight level of only 13. The average for the period of 1967±96 is to the surface, and the uncertainties inherent in intensity 19. There were tropical depressions in each month from estimates, this system has been kept as a tropical de- June to November. August was the most active month pression in the postanalysis. with four tropical depressions and all of them reached The cloud pattern of the depression soon lost its or- hurricane status. Because all the depressions became ganization in satellite imagery, and an aircraft recon- tropical storms in 1996, their history is provided in naissance ¯ight on the afternoon of 18 July had dif®- Pasch and Avila (1999). culty in ®nding a center, suggesting that the system had degenerated into a tropical wave. However, the distur- b. The 1997 Atlantic systems bance continued to show some signs of organization in satellite imagery, and another reconnaissance ¯ight In contrast with 1996, the number of tropical cyclones found a very weak circulation center at 1200 UTC on during the 1997 season was signi®cantly below average. 19 July in the northeastern Caribbean Sea. Convection There were six named tropical cyclones of which three associated with the wave moved over portions of the became hurricanes and three of the season's tropical Lesser Antilles, Puerto Rico, the Bahamas, and the Flor- cyclones developed from tropical waves, as indicated ida Straits. Unauthenticated | Downloaded 09/26/21 01:02 AM UTC OCTOBER 2000 ANNUAL SUMMARY 3697 FIG.