GARY PADGETT'S MONTHLY GLOBAL TROPICAL CYCLONE SUMMARY JULY, 2002 (For general comments about the nature of these summaries, as well as information on how to download the tabular cyclone track files, see the Author's Note at the end of this summary.) SPECIAL NOTE: The July summary was originally disseminated in three installments. There have been some significant additions to the reports on Typhoon Rammasun and Tropical Storm Nakri in the Northwest Pacific basin since the first installment was disseminated. SPECIAL NOTE #2: Effective this issue, I am dropping the section of the individual storm narratives which was entitled "Comparisons Between JTWC and Other Centers". All the information previously contained in this paragraph is available in the cyclone tracks file which I prepare each month. The tracks file can be retrieved from the archival sites listed at the end of this summary. I do have a limited e-mail distri- bution list for the tracks, and if anyone would like for me to add him/ her to the list, please send me an e-mail and I'll be happy to do so. ************************************************************************* JULY HIGHLIGHTS --> Northwest Pacific extremely active with nine tropical cyclones --> Guam and Okinawa experience typhoon strikes --> Brief Category 5 hurricane in Northeast Pacific --> Some activity in Atlantic subtropics ************************************************************************* ***** Feature of the Month for July ***** TRACK FORECASTING WITH A KILO-MEMBER ENSEMBLE Recently I attended the 25th Conference on Hurricanes and Tropical Meteorology in San Diego, California, sponsored by the American Meteor- ological Society. I heard many outstanding presentations on various topics: seasonal and monthly tropical cyclone forecasting, new track forecasting methodologies, tropical cyclone-induced rainfall studies, hurricane climatologies for various locations, papers concerned with cyclone classification issues--just to name a few. All were very interesting, but three really grabbed my attention as they focused on current important and problematic issues challenging tropical cyclone forecasters and researchers while at the same time offering some new and exciting techniques that should ultimately provide better solutions for these problems. Since these particular developments will likely be of considerable interest to many readers, I plan to highlight them in the Feature of the Month spot in this and upcoming summaries. One of the most pressing challenges in tropical meteorology is the need for more accurate cyclone track forecasts, especially at longer lead times of 36 to 72 hours and beyond. One of the conference papers alluded to above presented a different sort of approach to this problem. The paper, entitled "Track Forecasting of 2001 Atlantic Tropical Cyclones Using a Kilo-Member Ensemble", was presented by Jonathan Vigh, a graduate student at Colorado State University. I would like to extend a special thanks to Jonathan for giving me permission to summarize some of the features of his kilo-member ensemble and for reviewing this article. Portions which are quoted directly from the conference preprint are enclosed in quotation marks. (Jonathan's e-mail address and website link are included below.) The accuracy of tropical cyclone forecasts has improved greatly over the past few decades as a result of improved data, a better understanding of storms and their interaction with the environment, and vastly increased computing power along with increasingly sophisticated numerical model guidance. Seventy-two hour track errors are now half of those of 30 years ago. However, it is highly desirable to press onward toward the goal of reducing forecast errors even more at longer lead times in view of the enormous cost of closing down businesses, resorts and industries in preparation for an impending tropical cyclone strike as well as the large amount of time necessary to evacuate mass numbers of of people from threatened coastal regions. "Track error is comprised of many sources including analysis error caused by inaccurate spatial and temporal sampling, incomplete represen- tation of physical processes, discretization error, and truncation error. In high resolution full physics dynamical forecasting systems, the latter of these sources has been largely minimized. The remaining error results from uncertainties in the initial analysis coupled with the inherent chaotic nature of atmospheric dynamics." One way to help reduce the uncertainties in model runs due to the chaotic nature of the atmosphere is to make multiple runs of a model with perturbed initial conditions. Studies have shown that, with carefully chosen perturbations, the mean of an ensemble of such runs is generally more accurate than a single deterministic forecast. Jonathan's paper "describes the development of an ensemble forecasting system simple and fast enough to be used operationally with minimal computing resources, yet with a sample size large enough to simulate the subspace of dynamical pathways available to the tropical cyclone." The particular model Jonathan chose to use is a multigrid barotropic vorticity equation model (MUDBAR). The MUDBAR model has been found to reproduce the accuracy of the operational LBAR model in 1/70 the computing time. On a 1-GHz Intel Pentium PC, each 120-hour model run takes approximately 1.4 seconds. The model is initialized with 2-degree resolution deep-layer mean wind fields obtained from the NCEP Global Forecast System (GFS). Winds from an appropriately chosen bogus vortex are blended with background fields to obtain the initial conditions. Five classes of perturbations are considered (the numbers in parentheses refer to the number of perturbations in each class): in the background environmental flow (11), in the deep layer mean averaging (4), in the equivalent phase speed (3), in the vortex size/strength (3), and in the storm motion vector (5). Unlike some previous studies in ensemble forecasting, the various perturbation classes are cross multiplied, yielding 1980 members (hence the name "kilo-ensemble"). Jonathan writes: "One of the central questions addressed by this research is whether cross multiplication of perturbations leads to increased skill of the ensemble mean and better generation of error statistics." The ensemble was run retroactively for all the Atlantic 2001 cyclones except for Tropical Storm Allison and TD-02, and is being run in near real-time for the 2002 season. A new feature for 2002 is the inclusion of Eastern North Pacific storms in ensemble runs. Various statistical characteristics of the ensemble can be examined, including the average error, bias, skill relative to a single control, and skill relative to climatology and persistence (CLIPER) as well as to other operational guidance. In a personal e-mail message, Jonathan states that the model's performance has been best for low-latitude westward-moving storms in a barotropic atmosphere. Since the model being used is a barotropic model, it is not surprising that it doesn't work as well for systems in a more baroclinic environment. As indicated above, Jonathan has a website where many details of the kilo-member ensemble technique are explained. Especially interesting are plots of the ensemble results for the 2001 Atlantic season and for Atlantic and Northeast Pacific storms so far this year. The URL of the website is: Jonathan is continuing to work on verification of the ensemble forecasts and will be updating his website from time to time with new statistics, so interested parties should visit the website every now and then to see what's new. Also, for persons who wish to contact Jonathan via e-mail, he may be reached at: vigh@atmos.colostate.edu ************************************************************************* ACTIVITY BY BASINS ATLANTIC (ATL) - North Atlantic Ocean, Caribbean Sea, Gulf of Mexico Activity for July: 1 tropical storm 1 hybrid LOW Sources of Information ---------------------- Most of the information presented below was obtained from the various tropical cyclone products issued by the Tropical Prediction Center/National Hurricane Center (TPC/NHC) in Miami, Florida: discussions, public advisories, forecast/advisories, tropical weather outlooks, special tropical disturbance statements, etc. Some additional information may have been gleaned from the monthly summaries prepared by the hurricane specialists and available on TPC/NHC's website. All references to sustained winds imply a 1-minute averaging period unless otherwise noted. Atlantic Tropical Activity for July ----------------------------------- The first tropical storm of the Atlantic season, Arthur, formed on 15 July off the North Carolina coast from a disturbance which had originated in the Gulf of Mexico and moved across the southeastern U. S. into the Atlantic. Arthur rather uneventfully scooted off in the direction of Newfoundland. Earlier in the month there were some stirrings in the Atlantic subtropics. Late on 5 July one area of disturbed weather several hundred miles northeast of Bermuda was given a slight chance of developing into a depression before being absorbed by a frontal system, but this never materialized. Another area of disturbed weather located east-southeast of Cape Hatteras on 5 July merged with a frontal system the next day, but early in the morning of 7 July the LOW intensified and became quite organized while racing northeastward south of the Canadian Maritimes. Gale-force winds were recorded at Sable Island as the system swept past. (More on this interesting system below.) Also on 7 July a broad area of low pressure in the central Gulf of Mexico began to show signs of organization, and a reconnaissance aircraft flew into the system during the afternoon but did not find a closed low-level circulation. Conditions appeared favorable for some further strengthening and another plane investigated the disturbance on the 8th. However, the system had still not become sufficiently organized to warrant being classified as a tropical depression. The LOW continued moving westward and by midday on 9 July was moving inland into north- eastern Mexico and southern Texas. At the end of July there were also a couple of non-developing systems which should be mentioned. On the 30th a surface LOW several hundred miles east of Bermuda was interacting with an upper-level trough and producing widespread cloudiness and thunderstorms. While tropical or subtropical cyclone development was not really expected, there was at least a chance and this was noted in the Tropical Weather Outlooks from TPC/NHC. During the latter days of July a large tropical wave moved across the Atlantic and was approaching the Windward Islands on 27 July. An infrared satellite image taken at 27/0515 UTC shows a swirl in the clouds approaching the South American coast. No doubt it was probably just a mid-level twist, but looked interesting nonetheless. In the image it appears poised to moved into Guyana or Venezuela well southeast of Trinidad. Sable Island Storm of 7 July ---------------------------- During the first few days of July a broad surface trough extended northeastward from the east coast of Florida for several hundred miles. Widespread cloudiness and showers were present within the trough, and by 5 July an area of disturbed weather about 200 nm southeast of the North Carolina coast had begun to show some signs of organization. By the evening of the 5th the system was moving northeastward and its organization had continued to improve. The Tropical Weather Outlook at 06/0230 UTC noted that there was a potential for further development during the next 12 to 24 hours before it became absorbed by a frontal system. By the late morning of 6 July, however, the disturbance had merged with a front several hundred miles east-northeast of Cape Hatteras. On the evening of 7 July I received an e-mail (along with several other persons) from Chris Fogarty of Environment Canada, then stationed at the Newfoundland Weather Centre at Gander. Chris described a small disturbance which had passed near Sable Island earlier that morning, bringing gale-force winds to the island as well as to some oil rigs in the vicinity. Chris indicated that he had noted a mass of deep convection in infrared imagery the evening before in the southwesterly flow near Bermuda--clearly in a sheared environment--but it had held its form and was quasi-symmetric. Chris was of the opinion that it looked like a tropical cyclone in transition. Earlier on the morning of the 7th, Tony Cristaldi of the Melbourne NWS office had sent around an e-mail calling attention to a small, persistent CDO-like feature near 40N, 63W, and moving rapidly northeastward. Tony was of the opinion that it was traceable to the small circulation that was trying to form off Cape Hatteras on 5-6 July. Sable Island recorded its peak winds of 36 kts from the southeast, gusting to 43 kts, at 1200 UTC on the 7th. Winds gusted near or above gale force from around 1128 to 1300 UTC. A sounding taken there at 1200 UTC indicated three wind maxima exceeding 70 kts: 76 kts at 914 m (900-mb level), 72 kts at approximately 5800 m (500-mb level), and 72 kts at around 12800 m (200-mb level). Winds decreased to around 55 kts at intermediate altitudes between the maxima. Station LIPI (43.3N, 59.8W) recorded east winds of 40 kts at 1200 UTC. Tony Cristaldi sent me some observations from Canadian buoy 44137 taken as the center of the LOW passed to its west and north. At 1000 UTC the buoy reported southeast winds of 37 kts with a MSLP of 1007 mb. (The distance of the buoy from the LOW's center is unknown to the author.) Chris had sent a goodly amount of data to Jack Beven at TPC/NHC, and a couple of days later Jack offered his opinion of the system. Tropical characteristics exhibited by the cyclone included a small, tight center of circulation; the impressive convective mass with cloud tops exceeding -70 C; and evidence of spiral banding seen in radar imagery. On the extratropical side of the fence, satellite imagery on 6 and 7 July indicated that the center of the LOW was entangled with a cold front. Surface observations indicated that some degree of temperature gradient was present throughout the life of the cyclone. Jack also felt that the triple wind maxima noted in the Sable Island sounding was not very characteristic of a warm-core system, at least not over a deep layer. Also, the 500-mb temperature of -11.7 C was rather cold for a tropical cyclone. In the final analysis Jack's conclusion was that the system was some sort of frontal hybrid with winds of at least 40 kts and a central pressure below 1005 mb. The presence of the front would disqualify the cyclone from classification as a subtropical or tropical cyclone and it will not be included in the 2002 season totals. Jack notes that "even if it wasn't a tropical cyclone, it sure had ideas about becoming one." (Note: No track for this system was available for inclusion in the July cyclone tracks file.) (Report written by Gary Padgett) TROPICAL STORM ARTHUR (TC-01) 13 - 19 July ----------------------------------------- Before setting out to write a summary for Arthur, I discovered that happily the official NHC storm report on this cyclone is already available on NHC's website. The report, authored by Miles Lawrence, can be accessed at: Since the official report is already available, there's no need to write a whole lot about Arthur. Indeed, there's not a whole lot to write about it anyway. According to Miles' report, Arthur stemmed from a weak LLCC first noted in the eastern Gulf of Mexico around 9 July, likely spawned by a decaying frontal zone which had persisted in the region for several days. Michael Pitt sent me the Navy JMV track for Arthur, which I used as the basis for the early portion of the storm's track in the cyclone tracks file for July. The maximum winds were 20-25 kts for several days as the weak LOW remained quasi-stationary. I began the track at the point where the MSW picked back up to 25 kts and remained at or above that intensity. That point was about 40 nm southeast of Apalachicola, Florida, at 1200 UTC on 13 July. The system had already begun to move northeastward and by 1800 UTC was inland just south of Lake City. The center was back over the Atlantic about 100 nm northeast of Charleston, South Carolina, at 14/1200 UTC, and by 1800 UTC had developed enough convection and warm-core characteristics to be classified as a tropical depression. The first advisory on TD-01 was issued by NHC at 2100 UTC on 14 July, placing the center about 40 nm south-southwest of Cape Hatteras. At the time it was classified, the depression was moving at 17 kts to the east-northeast, and the translational speed increased with almost every subsequent advisory. Even though the system was within a sheared environment, its rapid east-northeasterly motion helped to minimize the relative shear and it slowly strengthened. During the early morning of the 15th reports from U. S. Navy ships and buoy 41001 indicated sustained winds ranging from 35 to 47 kts, so at 1500 UTC the depression was upgraded to Tropical Storm Arthur, located about 250 nm east- northeast of Cape Hatteras. Maximum winds were estimated at 45 kts and the storm was moving east-northeastward at 20 kts. Miles' report indicates that a mid-level cut-off LOW dropped southward over the Canadian Maritimes and deepened. Arthur scooted east- northeastward around the southern periphery of this LOW, later turning northward around the eastern side of the LOW. The storm reached its peak intensity of 50 kts at 16/0000 UTC when it was centered about 350 nm south of Nova Scotia. Arthur's forward motion had increased to 34 kts by 1200 UTC on the 16th. By 2100 UTC the storm had become sufficiently extratropical that NHC issued its final advisory. The post-Arthur extratropical storm slowed as it turned northward around the eastern periphery of the cut-off LOW and headed toward Newfoundland. The storm crossed eastern Newfoundland on 17 July where winds to 34 kts were recorded at Bonavista. Maximum rainfall amounts were about 25 mm. The storm continued northward and became quasi-stationary between Newfoundland and Greenland as it slowly weakened. I would recommend that interested persons check out Miles' report on the TPC/NHC website. The Best Track for the storm is included as well as a table of selected ship reports of gale-force or higher winds associated with Arthur. (Report written by Gary Padgett) ************************************************************************* NORTHEAST PACIFIC (NEP) - North Pacific Ocean East of Longitude 180 Activity for July: 1 tropical storm 2 hurricanes Sources of Information ---------------------- Most of the information presented below was obtained from the various tropical cyclone products issued by the Tropical Prediction Center/National Hurricane Center (TPC/NHC) in Miami, Florida (or the Central Pacific Hurricane Center (CPHC) in Honolulu, Hawaii, for locations west of longitude 140W): discussions, public advisories, forecast/advisories, tropical weather outlooks, special tropical disturbance statements, etc. Some additional information may have been gleaned from the monthly summaries prepared by the hurricane specialists and available on TPC/NHC's website. All references to sustained winds imply a 1-minute averaging period unless otherwise noted. Northeast Pacific Tropical Activity for July -------------------------------------------- Over the period 1971-2001, the Northeast Pacific basin has produced an average of 4 tropical storms, 2 hurricanes, and 1 major hurricane during the month of July. This year was pretty much close to average with 3 named storms, 2 reaching hurricane intensity, and one of those becoming intense. The highlight of the month was the occurrence of the first Category 5 hurricane in the basin in five years (the last one being the most intense Eastern Pacific hurricane on record--Hurricane Linda of 1997). However, all three storms moved westward away from the Mexican coast with little effect on land. The following reports on Eastern North Pacific tropical cyclones were written by John Wallace of San Antonio, Texas. A special thanks to John for his assistance. TROPICAL STORM CRISTINA (TC-04E) 9 - 16 July ------------------------------------------- The cyclonic feature that became Cristina was noticeable in infrared and visible imagery as early as 6 July, probably originating from a disturbance that crossed Central America that day. The disturbance organized slowly, and was ragged-looking even at its upgrade to Tropical Depression Four-E at 1500 UTC on 9 July about 320 nm south of Acapulco. Conditions were not ideal--an upper-level LOW over the Bay of Campeche was generating moderate westerly shear over the system. Though forecasts predicted strengthening--the GFDL to 115 kts--shear prevented Four-E from intensifying significantly over the next two days as it tracked to the west-northwest south of a ridge. The existence of a closed circulation was doubtful early on the 10th, but the depression made a minor comeback later that day as it turned more to the west. It had deteriorated again by early on the 11th, however, when ship MZFK4 reported a west wind of only 10 knots. A British ship, the Harrier, recorded a sustained wind of only 15 kts that day when located 60 nm south-southwest of the putative center. On the whole, the depression's convection oscillated in a diurnal pattern common in weak tropical systems. Beginning late on the 11th the synoptic situation improved, and the depression became Tropical Storm Cristina at 0300 UTC on 12 July when it was 575 nm west-southwest of Manzanillo. It intensified only slightly as it tracked westward on the 12th. Early on the 13th, Cristina became quasi-stationary as steering currents briefly collapsed before commencing a slow northwestward track due to interaction with an upper-level trough to its north. The storm changed little in intensity until the following day when it suddenly strengthened to its peak MSW of 55 kts, with an estimated CP of 994 mb, at 1500 UTC on 14 July approximately 810 nm west of Manzanillo. Cristina weakened thereafter, first due to dry air, and later to cooler SSTs. By early on the 15th, it was devoid of deep convection and was accordingly downgraded to a depression later that day. After a brief north-northwestward track oscillation, Cristina turned increasingly westward as it came under the influence of low-level easterlies. The last advisory on Tropical Depression Cristina was issued at 2100 UTC on 16 July when it was located 1065 nm west of Manzanillo. The remnant vortex had dissipated by the 18th. No damages or casualties are known to have resulted from Tropical Storm Cristina. (Report written by John Wallace) HURRICANE DOUGLAS (TC-05E) 20-26 July ------------------------------------- A. Storm Origins ---------------- The system that became Douglas was first evident in the Pacific south of Mexico on the 19th. The disturbance organized swiftly, from an amorphous, barely recognizable LOW at 1200 UTC on the 19th to its upgrade to Tropical Depression Five-E at 1500 UTC on 20 July roughly 380 nm south-southwest of Manzanillo, Mexico. The system thus set its future pace even in its infancy. B. Track and Intensity History ------------------------------ Five-E had strengthened to Tropical Storm Douglas by the second advisory, issued at 2100 UTC. The fledgling storm was located 360 nm south-southwest of Manzanillo. The strengthening was something of a surprise in the face of modest northeasterly shear. The storm tracked northwestward on the 21st as it experienced minor binary interaction with a disturbance that circulated around its northern periphery. This second system was absorbed by Douglas later that day, and the storm assumed a west-northwesterly track. The binary interaction did not affect Douglas's intensification, nor did the initial modest shear, and the storm strengthened rapidly. By 1915 UTC on the 21st an eye was present in TRMM imagery, only a day after becoming a tropical storm, and at 0300 UTC on 22 July Douglas was upgraded to a hurricane while located approximately 380 nm west-southwest of Manzanillo. At 2100 UTC the same day Douglas reached its peak MSW of 90 kts, with an estimated CP of 970 mb, when centered about 515 nm west of Manzanillo. The radii of the storm-force winds were adjusted at this time, based on a report from ship UGTY of 40 kt winds 210 nm east-southeast of the center. This report made Douglas a large storm for the northeastern Pacific. The hurricane maintained its peak intensity of 90 kts for 18 hours. Douglas's CP fell roughly 0.94 mb/hr after its upgrade to a hurricane, with an average rate of 0.71 mb/hr after it became a tropical storm--not officially rapid deepening, but impressive nonetheless. The forecaster who wrote the 22/0900 UTC discussion stated that rapid intensification following a westward turn is not uncommon. Similar events occurred in the Atlantic in association with Hurricane Andrew in 1992 and Hurricane Floyd in 1999. Douglas tracked nearly due westward at its peak. A weakening trend began on the 23rd as it turned to the west-northwest and accelerated. Cool SSTs and stable air took their toll on Douglas, but the storm declined slowly. The hurricane underwent a concentric eyewall replacement cycle on the 23rd and into the 24th, interesting for a system that was apparently weakening. The question of whether the weakening trend or eyewall replacement cycle came first might be a topic for future research--if the answer is not obvious to those in the know already. In any event, the remnant inner eyewall persisted until 0900 UTC on the 24th. Douglas dropped below hurricane strength late on the 24th, and turned to the west as it became a shallow vortex. Its track was unusually rapid--easterlies south of a strong ridge pushed Douglas to 17-19 kts on the 25th. By late that day, it was nearly devoid of convection. The final advisory on Tropical Depression Douglas was issued at 1500 UTC on 26 July when it was located about 1790 nm west of Manzanillo. Douglas's remnant vortex persisted until around the 29th. C. Damage and Casualties ------------------------ No casualties or other damages are known to have resulted from Hurricane Douglas. (Report written by John Wallace) HURRICANE ELIDA (TC-06E) 23 - 30 July ----------------------------------- Elida was noteworthy in two respects--one was that it was the latest first intense hurricane in a Northeast Pacific season since 1996. The second will be revealed later in this report. A. Storm Origins ---------------- The origin of Elida can apparently be traced back to a vigorous tropical disturbance that was first noted in the southwest Caribbean on 19 July. There were faint signs of cyclonic structure evident by the 20th as the disturbance tracked northwestward along the Central American coast. The system failed to thrive, however, and by later that day its organization collapsed. Convection was sporadically vigorous while the disturbance crossed the Isthmus of Tehuantepec. Vague cyclonic structure developed again on the 22nd, but it was unremarkable until mid-day on 23 July when its organization increased in dramatic fashion. From an amorphous convective mass at 0000 UTC, it had become Tropical Depression Six-E by 1500 UTC about 320 nm south- southeast of Acapulco. At its upgrade, it was already just below tropical storm strength. The depression tracked westward, south of the same ridge that had guided Hurricane Douglas, then located about 1000 nm to the west-northwest. B. Track and Intensity History ------------------------------ The depression was upgraded to Tropical Storm Elida on the next advisory at 23/2100 UTC when it was located about 280 nm south- southwest of Acapulco. At its "upgrade", the MSW was already up to 50 kts. Synoptic conditions were extremely favorable, and Elida showed every indication of following the rapid intensification trend that Douglas had. Unlike Douglas, though, its intensification was much more aggressive. By 0037 UTC on the 24th, an SSM/I overpass showed a banding-type eye, while other infrared imagery indicated a possible eyewall in the middle of a very cold CDO. Ship H9LA reported a sustained wind of 36 kts with 4.6 m seas around this time well away from the center, which led the NHC to expand the radius of storm-force winds and high seas. At 0900 UTC on 24 July, Elida was upgraded to a 95-kt hurricane when centered about 410 nm south of Manzanillo, Mexico (the TAFB analyst had assigned a Dvorak T-number as high as 6.0, or 115 kts). Elida reached Category 4 intensity at 1500 UTC some 395 nm south of Manzanillo. It is interesting to note that the SHIPS model forecast a peak MSW as high as 157 kts. Between its upgrade to Six-E (23/1500 UTC) and 1500 UTC on the 24th, the CP had fallen 66 mb, representing an average drop of 2.2 mb/hr, well into the rate for official rapid deepening (1.75 mb/hr for 24 hrs). However, this is only an average--between 0300 UTC and 1500 UTC on the 24th, its pressure fell an average of 4.5 mb/hr, solid explosive deepening (2.5 mb/hr for 12 hrs). After 1500 UTC the deepening rate slackened somewhat, but continued while the hurricane turned west-northwestward the following day. Its track wobbled somewhat, probably due to the trochoidal oscillations that are common in intense hurricanes. Enhanced TRMM imagery from 0257 UTC on the 25th indicated concentric eyewalls, which persisted until at least 1748 UTC. Elida reached its peak intensity of 140 kts near 0600 UTC on 25 July (1) about 415 nm southwest of Manzanillo. Using the CP-MSW relationship for the Atlantic and NEP, this suggests a minimum CP of 921 mb. The estimate was based on T-numbers of 6.9 and (roughly) 7.0 from GOES-8 and GOES-10 imagery, respectively. This made Elida not only the first Category 5 hurricane in the NEP since Linda in 1997, it also made it the first official one in the Western Hemisphere since Mitch of 1998. Though there are several synoptic situations that can lead to rapid deepening, Elida's may have been due to warm SST anomalies present along its track, noted in SST maps at the time. Then again, part of the intensification "roller coaster" may have resulted from concentric eyewall cycles. Indeed, these cycles seem to have played a role in its post-peak intensity fluctuations as a 100+ kt hurricane. There are a few individuals in the tropical meteorology community who doubt Elida's peak intensity and explosive deepening, noting that satellite estimates often lead true intensity. A good example of this bias was the case of Hurricane Carlotta in 2000 (2). Without reconnaissance data, however, Elida's fit into this controversy will never be known. Elida began a weakening trend right after its peak, though only slowly. Interestingly, the eye actually contracted while it weakened to only 5 nm at 2100 UTC on the 25th. By 0300 UTC on the 26th, Elida's inner eyewall had weakened and dissipated while its track bent back toward the west due to enhanced ridging to its north. Later on the 26th, the storm's track turned west-northwestward once again and Elida fell below intense hurricane strength as it went through another eyewall cycle. Elida was still a large system--3.7 m seas extended as far as 400 nm from the center in the northern semicircle. At 1500 UTC on the 27th, Elida dropped below hurricane intensity. The storm's convection made a comeback early on the 28th, which enabled Elida to hold its own considerably longer than most NEP storms that cross into cool waters. Excellent upper-level outflow and the system's large circulation also gave it a reprieve. Even the NHC was impressed by Elida's tenacity, which one forecaster hypothesized might have been due to cooler air over its center that helped to maintain instability. Elida turned northwestward on the 28th, then north-northwestward the following day as a trough eroded the ridge to its north. On the 29th convection finally decreased enough for it to be downgraded to a depression over 19 C waters. Tropical Depression Elida weakened apace, and the final advisory was issued on the almost convection-free vortex at 0300 UTC on 30 July, then located about 1000 nm west-northwest of Cabo San Lucas. Elida still had some bite--ship reports indicated that high seas extended some 200 nm north of the center near the time of the last advisory. The remnant vortex dissipated off the California coast the following day. C. Casualties and Damage ------------------------ No casualties or other damages are known to have resulted from Hurricane Elida. D. References ------------- (1) The peak intensity given in the advisories was 135 kts at 0900 UTC on the 25th. Cloudtop temperatures had warmed slightly since the synoptic hour of 0600 UTC, so it was felt like the storm had already peaked and was beginning to weaken. (2) http://www.nhc.noaa.gov/2000carlotta.html (Report written by John Wallace) ************************************************************************* NORTHWEST PACIFIC (NWP) - North Pacific Ocean West of Longitude 180 Activity for July: 1 tropical depression ** 3 tropical storms ++ 2 typhoons 3 super typhoons ** - system was not classified as a tropical depression by JTWC nor HKO, but was by JMA and some of the other Asian TCWCs ++ - one of these was treated as a tropical storm by JTWC only for a 6-hour period--another was classified as a tropical storm only by JMA and NMCC Sources of Information ---------------------- Most of the information presented below is based upon tropical cyclone warnings and significant tropical weather outlooks issued by the Joint Typhoon Warning Center of the U. S. Air Force and Navy (JTWC), located at Pearl Harbor, Hawaii. In the companion tropical cyclone tracks file, I normally annotate track coordinates from some of the various Asian warning centers when their center positions differ from JTWC's by usually 40-50 nm or more. All references to sustained winds imply a 1-minute averaging period unless otherwise noted. Michael V. Padua of Naga City in the Philippines, owner of the Typhoon 2000 website, normally sends me cyclone tracks based upon warnings issued by the Japanese Meteorological Agency (JMA) and the Philippines' Atmospheric, Geophysical & Astronomical Services Administration (PAGASA). Also, Huang Chunliang of Fuzhou City, China, sends me each month tracks obtained from warnings issued by the National Meteorological Center of China (NMCC), the Central Weather Bureau of Taiwan (CWBT) and the Hong Kong Observatory (HKO). A very special thanks to Michael and Chunliang for the assistance they so reliably provide. In the title line for each storm I have referenced all the cyclone names/numbers I have available: JTWC's depression number, the JMA-assigned name (if any), JMA's tropical storm numeric designator, and PAGASA's name for systems forming in or passing through their area of warning responsibility. The summaries for Tropical Storms Nakri and Juan were written by Kevin Boyle of Stoke-on-Trent in the UK. Kevin does full-time voluntary work for the Newchapel Observatory, keeping meteorological records for the observatory from an automatic weather station situated on site. The data is sent to the Climatological Observers Link with a monthly summary of atmospheric phenomena which Kevin prepares. Kevin was an avid fan of Jack Beven's weekly tropical cyclone summaries during the early 1990s, and after Jack ceased writing them, Kevin began writing similar reports of the Northern Hemisphere basins for his own personal use. A very special thanks to Kevin for writing the reports on the two cyclones. Northwest Pacific Tropical Activity for July -------------------------------------------- Tropical activity in the Northwest Pacific basin literally exploded during the month of July. Two tropical storms which were named at the end of June, Chataan and Rammasun, matured into typhoons with Chataan reaching super typhoon intensity (130 kts) on JTWC's scale. In its early tropical storm stage, Chataan dropped torrential rains on Chuuk which resulted in several dozen deaths and much damage. Later, the storm crossed Guam as a Category 2 typhoon, causing significant damage. In its latter moments as a tropical cyclone Chataan sideswiped southern Honshu, causing flooding rains and several fatalities. Typhoon Rammasun brushed the eastern coast of China before making landfall as a weakening tropical storm in South Korea. Several deaths and widespread scattered damage was reported in and around Shanghai and in Zhejiang Province to the south. Two more tropical cyclones came to life during the first week of the month. Typhoon Halong developed in the eastern Caroline Islands near where Chataan had formed and followed a similar track to the earlier storm. Halong passed less than 100 nm south of Guam, briefly reached super typhoon intensity, and then moved across the southern end of Okinawa as a Category 2 typhoon. Like Chataan, Halong also brushed southern Honshu. Weaker Tropical Storm Nakri moved northward off the west coast of Taiwan, turning eastward near the island's northern tip and moving to near Okinawa before turning northward and weakening. Although none of the tropical cyclones directly affected the Philippines, their presence to the north and northeast of the archipelago helped to enhance and intensify the southwest monsoon, leading to heavy rains for several weeks. Press reports indicated that more than 60 persons lost their lives in the nation during July from the monsoonal rains. Shortly before mid-month, Tropical Storm Fengshen sprang to life in the Marshall Islands area and quickly intensified into a typhoon. Fengshen reached super typhoon intensity at 18/0000 UTC and maintained that status for 120 hours, setting a new record for super typhoon longevity. The previous record of 114 hours was set by Typhoon Joan in 1997. Fengshen passed well north of the Marianas Islands and approached Japan, but had weakened significantly by the time it neared Kyushu. The weakening residual depression limped ashore in eastern China. While Fengshen was still well east of the Marianas, Typhoon Fung-wong formed northwest of the islands and the two storms engaged in a Fujiwhara interaction with Fung-wong describing a large counterclockwise loop. While Fengshen and Fung-wong were performing their do-si-do, Tropical Depression Juan moved through the central Philippines. JTWC upgraded the system to minimal tropical storm status for just one warning cycle. Tropical Storm Kalmaegi was a weak, short-lived tropical storm which sprang up near and just west of the International Dateline on 20 July. JTWC classified the system only as a tropical depression, but both JMA and NMCC upgraded the system to tropical storm status. During the final days of the month, a low-pressure system in the Gulf of Tonkin was treated as a tropical depression by JMA and by the Meteorological Department of Thailand, but not by any of the other warning centers. There was one other disturbance which was labeled a tropical depres- sion by JMA. The system was located roughly 475 nm east of Iwo Jima at 25/0000 UTC, and had moved west-northwestward to a point about 250 nm east-northeast of the island by 26/0000 UTC. The system was relegated to the summary portion of JMA's High Seas Bulletins, implying maximum winds less than 30 kts. JTWC mentioned this disturbance for a couple of days in their STWOs, but only gave it a poor potential for develop- ment. No track was included for this system in the companion cyclone tracks file. SUPER TYPHOON CHATAAN (TC-08W / TY 0206 / GLORIA) 28 June - 12 July ----------------------------------------------- Chataan: submitted by the United States, is the Chamorro word for 'rain' Gloria: PAGASA name, is a feminine name A. Storm Origins ---------------- What would become the first typhoon to directly strike Guam since the destructive Paka in December, 1997, began to develop southeast of Chuuk late in June simultaneously with Typhoon Rammasun farther to the west. The development of both these tropical cyclones was spawned/enhanced by a surge in monsoonal westerlies along and just north of the equator in the western Pacific. JTWC issued an interim STWO at 0100 UTC on 27 June, noting that an area of convection had developed approximately 150 nm east-southeast of Chuuk. A very broad circulation was embedded in the monsoon trough with an upper-level LOW to the north enhancing outflow. The disturbance was relocated to a position about 300 nm southeast of Chuuk at 0600 UTC. An area of deep convection was consolidating near the LLCC, and at 0900 UTC the development potential was upgraded to fair. At 27/2000 UTC JTWC issued a TCFA for the system, then located about 210 nm southeast of Chuuk. Deep convection was becoming more organized near the center, and an upper-level analysis depicted good divergence aloft with weak vertical shear. JTWC issued the first warning on Tropical Depression 08W at 28/0000 UTC with the center estimated to be about 190 nm southeast of Chuuk. The depression subsequently meandered slowly northward, reaching a point about 130 nm southeast of Chuuk by 1800 UTC. Deep convection weakened some during the day as a shortwave trough deepened north of the system, but had begun to make a comeback by 1800 UTC with the MSW being upped to 30 kts from the initial warning intensity of 25 kts. B. Track and Intensity History ------------------------------ JTWC upgraded the depression to tropical storm status at 29/0000 UTC with an initial MSW of 35 kts. The center was relocated to a point about 210 nm east-southeast of Chuuk at this time, and the warning inten- sity was based on CI estimates of 30 and 35 kts. Convective organ- ization had improved by 0600 UTC and the MSW was upped to 45 kts. At the same time, JMA and NMCC upgraded the system to tropical storm inten- sity with JMA assigning the name Chataan. Intensification leveled off, however, after 0600 UTC and the MSW remained at 45 kts for the remainder of the 29th. During this time Tropical Storm Chataan moved slowly and erratically in the region approximately 200-250 nm east-southeast of Chuuk. The MSW was increased to 50 kts at 0000 UTC on 30 June based on CI estimates of 45 and 55 kts. All through the 30th Chataan remained quasi-stationary roughly 250 nm east-southeast of Chuuk. SSM/I data showed that the LLCC was displaced slightly to the east of the deeper convection. (JMA also estimated the intensity at 50 kts during this period--NMCC was slightly lower at 45 kts.) Chataan's intensity remained static at 50 kts on 1 July with deep convection cycling to the west of the LLCC. During the day the storm drifted very slowly northwestward and by 01/1800 UTC was located about 180 nm east of Chuuk. Animated water vapor imagery depicted a short- wave trough deepening to the north-northwest of Chataan, weakening the ridge located west of the cyclone. The storm began a more pronounced westward motion after 1800 UTC and by 0000 UTC on 2 July was located about 100 nm east of Chuuk. This westerly movement continued and Tropical Storm Chataan passed almost directly over Chuuk around 1200 UTC. (More information on the torrential rainfall which fell on Chuuk and the accompanying disaster is included below.) The storm's intensity fluctuated somewhat on the 2nd between 45 and 50 kts (per JTWC's warn- ings) as satellite CI estimates remained generally between 45 and 55 kts. An upper-level LOW north-northeast of Guam appeared to be inhibiting the cyclone's outflow somewhat. (NMCC's estimated 10-min avg MSW remained at 45 kts during this period while JMA reduced their estimate to 40 kts on the 2nd.) JTWC upped the intensity once more to 50 kts at 03/0000 UTC and the storm finally began to exhibit a marked intensification trend. Chataan also began to track more northwestward under the influence of a ridge to the north which was re-orienting itself in a northwest-southeast direction. By 1800 UTC CI estimates had reached 65 and 77 kts, so JTWC upgraded Chataan to a typhoon, located approximately 350 nm southeast of Guam and moving north-northwestward at 8 kts. The MSW was bumped up to 75 kts at 0000 UTC on 4 July--a 03/2116 UTC TRMM pass had revealed an eye which was obscured by the cirrus shield. Chataan's north-northwesterly motion had increased to 14 kts by 0600 UTC, but later the storm took a jog to the west which placed Guam more in the line of fire. At 1200 UTC the typhoon was located about 125 nm east-southeast of the island, moving west-northwestward at 14 kts. Satellite imagery revealed that the eye had contracted, therefore, the intensity was upped to 90 kts. The MSW was further increased to 95 kts at 1800 UTC, based on CI estimates of 90 and 102 kts, as the eyewall began to track across the island of Guam. (NMCC did not upgrade Chataan to typhoon status until 04/1800 UTC, and JMA did not do so until 05/0000 UTC, after the storm had crossed Guam.) At 2200 UTC the eye of Typhoon Chataan was located over northern Guam about 24 km northeast of Hagatna. (More information on the typhoon's effects on Guam and some meteorological observations are included below.) The storm's eye had emerged off the northwest coast of the island by 0000 UTC on the 5th and the typhoon continued to march west-northwestward away from Guam throughout the day. Chataan gradually intensified with the MSW reaching 110 kts by 1800 UTC. The storm was then located about 200 nm west-northwest of Guam and moving northwestward. Satellite inten- sity estimates were 115 kts, but animated imagery indicated weakening poleward outflow and warming of convective cloud tops. A 05/2308 UTC SSM/I pass revealed a very small eye on the northern edge of the deep convection while animated multi-spectral imagery indicated that the storm was experiencing some northerly shear. Chataan continued trekking northwestward on 6 July, generally between 10 and 14 kts. Even though CI estimates remained at 115 kts, JTWC lowered the MSW to 100 kts at 1200 UTC due to the observed shearing and filling of the eye. By 1800 UTC the storm had reached a point approximately 450 nm south-southwest of Iwo Jima, and animated infrared imagery indicated that the shearing over the storm had lessened. (The 10-min avg MSW estimates from JMA and NMCC on the 6th were 85 and 90 kts, respectively, and do not show the weakening indicated by JTWC.) Satellite imagery at 07/0000 UTC indicated that the eye was redevel- oping. JTWC increased the intensity to 115 kts at 0600 UTC, and to 120 kts at 1200 UTC--the typhoon at that time exhibited a well-defined 24-nm diameter eye. By 1800 UTC, CI estimates were 127 kts and JTWC upgraded Chataan's MSW to 130 kts, making it the third super typhoon of the year. The typhoon was then centered approximately 470 nm south- east of Okinawa, moving west-northwestward at 10 kts. Gales extended outward from the center about 165 nm, and the radius of storm-force winds was estimated at 90 kts. Chataan's direction of motion gradually became more poleward on the 8th. The storm continued tracking along the western periphery of a mid-level ridge situated southeast of Japan, but a short- wave ridge and trough moving off the coast of China began to re-orient the ridge in a more north-south direction. Super typhoon status was maintained for 18 hours--at 1200 UTC the intensity was decreased slightly to 120 kts. Satellite CI estimates remained at 127 kts, but the storm appeared to have embarked on a weak- ening trend. (The peak 10-min avg MSW estimate from JMA was 95 kts from 08/0000 through 09/0000 UTC, and NMCC's peak intensity was 100 kts from 07/1200 through 09/0000 UTC. The minimum CP estimated by JMA was 930 mb.) Typhoon Chataan passed about 250 nm east of Okinawa around 0000 UTC on 9 July, moving northward at 12 kts. The MSW had been lowered to 115 kts and continued to fall throughout the day. SSM/I data indi- cated weakening convection in the western semicircle as cool, dry air was entrained into the system. By 1800 UTC Chataan was located approx- imately 450 nm southwest of Tokyo Bay, moving north-northeastward at 6 kts with the MSW estimated at 85 kts (based on CI estimates of 77 and 90 kts). On the 10th Chataan accelerated to the northeast along the northwest periphery of the mid-level ridge as it began to feel the effects of the mid-latitude westerlies associated with a shortwave trough. By 1200 UTC the storm was located about 160 nm south-southwest of Tokyo Bay and moving northeastward at 25 kts. Deep convection had continued to erode and a 10/1108 UTC SSM/I pass revealed an exposed LLCC southwest of the remaining convection. JTWC and NMCC downgraded Chataan to a tropical storm at this time--JMA had done so twelve hours earlier. The cyclone crossed the extreme southeastern tip of Honshu (Boso Peninsula) and by 1800 UTC was located approximately 55 nm east of Tokyo. At 0000 UTC on 11 July Chataan was located about 160 nm east-southeast of Misawa, Japan, and racing northeastward at 37 kts. JTWC issued their final warning on the system at this time, reducing the intensity to 40 kts and declaring Chataan extratropical. A 10/2332 UTC SSM/I pass depicted comma-shaped convection wrapping around the LLCC, a characteristic of extratropical cyclones. JMA, however, estimated the 10-min avg MSW at 55 kts and maintained Chataan as a tropical cyclone through 1200 UTC when it was located over eastern Hokkaido. By 1800 UTC JMA had also classi- fied the former typhoon as extratropical, and the slowly weakening system thereafter moved north-northwestward, reaching southern Sakhalin Island by 12/1800 UTC as a weak 25-kt LOW. C. Meteorological Observations ------------------------------ On the islands of Chuuk, the story of Chataan was rain, rain, and more rain. On 2 July alone 506 mm of rain was recorded at the Chuuk International Airport, and over the three-day period 1-3 July a total of 622 mm was recorded. This followed on the heels of 317 mm which had fallen during the cyclone's formative stages (23-30 June). The peak wind gust reported on Chuuk was 44 kts at 03/0529 UTC. The heavy rains which Chataan produced on Chuuk were likely not orographically- induced, as the islands of Tol and Moen reach peaks of only 442.9 and 373.0 metres, respectively. (This information was taken from a webpage created by Jonathan Vigh, a graduate research student at Colorado State University. A special thanks to Jonathan for sharing the information with me. More information can be found at the following website: ) On Guam the minimum SLP reported during Typhoon Chataan's passage over the island was 961 mb. (Note: In the monthly tropical cyclone tracks file I prepared and disseminated earlier, the CP values given for Chataan at 04/1800 and 05/0000 UTC of 975 and 970 mb, respectively, were those estimated by JMA in their warnings. Obviously the well-substantiated minimum pressure of 961 mb from several stations on Guam indicate these estimates were a little high.) According to information received from Mark Lander, Chataan's eye was about 30 nm in diameter when it crossed northern Guam, requiring up to two hours to pass over some locales. NEXRAD base velocity readings were 100 kts inbound and outbound as the eye approached. Phil Smith of Hong Kong sent me a report a friend of his had received via telephone from an acquaintance in Guam. The gentle- man in Guam reported that the eye was gray and cloudy with no sunshine and some drizzle. Mark Lander also reported that a thick haze or fog was noted in the eye. Roger Edson sent a report of the peak winds recorded at Andersen AFB and the Guam NWS Airport station. Andersen AFB was in the eye for two hours with the minimum pressure of 961 mb recorded at 04/2255 UTC. The peak gust before eye passage was 72 kts at 04/1940 UTC, and the peak following the eye was 90 kts at 04/2355 UTC. At the airport (located 24 km southwest of Andersen AFB), the peak gust preceding the eye was 77 kts at 04/2140 UTC and the peak following eye passage was 78 kts at 04/2338 UTC. The minimum pressure there of 961.1 mb was also recorded at 04/2255 UTC. The JTWC warning at 05/0000 UTC remarked that the highest surface sustained wind recorded on the island was 70 kts. According to a report in the Pacific Daily News, the island recorded about 280 mm of rain associated with Chataan in a 42-hour period. How- ever, in typhoon winds much of the rainfall is not caught by standard rain gauges. Mark Lander reports that the rain had ended by the after- noon of the 5th but that winds were still of storm force out of the south-southwest. He observed stripes of sea spray racing across the tops of the waves and orographically-induced tornado-like eddies of sea spray spinning across Pago Bay in the lee of the Tagachang promon- tory. D. Damage and Casualties ------------------------ The torrential rains of Tropical Storm Chataan were particularly devastating to the islands of Chuuk. A report dated 12 July from the Chuuk chapter of the Micronesian Red Cross Society indicates that the final death toll stands at 48 with 73 persons injured. Over 1300 people were left homeless and 130 houses were completely destroyed. Several weeks after the storm there was a growing concern about a possible increase in certain infectious diseases. Much of the devastation was caused by landslides unleashed by the heavy rains--a total of 62 slides were reported. Some ranged up to 400-500 metres in length and were from 200-300 metres wide. Many crops were destroyed, exacerbating a pre-existing food shortage. The worst affected islands were Weno, Tonoas, Fefan, Udot, Uman and Siis. (More information can be found at the following website: ) On Guam Typhoon Chataan was responsible for significant damage, but nothing like the devastation wrought by Super Typhoon Paka in December, 1997. The Pacific Daily News reported that the typhoon caused damage to power lines and utility poles, flooded roads, damaged buildings, and left debris strewn over the island. The entire island experienced a power outage with water service erratic to non-existent in some areas for several days. Mark Lander reported that a friend living in the south part of the island related that at the height of the storm vortices shedding from the edges of his house could be seen streaming out in the rain and shredding his large breadfruit tree branch by branch. Mark noted that a survey of the damage to vegetation indicates a Category 2 typhoon: extensive damage to broadleaf trees like breadfruit, flame trees, monkeypods, plumeria, and tangantangan. Also, bamboo groves were crushed. However, damage to coconut trees was limited with about 1 in 100 blown down. According to Mark, an average Category 3 typhoon would be expected to down about 10% of coconut palms. Also, no auto- mobiles were flipped, something which typically happens in a Category 3 typhoon. According to a report on the ReliefWeb website, the total damage to public facilities only was estimated at $59.9 million. Although far from Chataan's destructive winds, the Philippines experienced the effects of the storm in the form of monsoonal rains, enhanced by the passing cyclones. A report from the Philippine Daily Inquirer, dated 11 July, indicates that the death toll from the rains attributed to Chataan (known in the Philippines as Gloria) and from Tropical Storm Nakri (Hambalos) had risen to 30 with four still missing and 41 injured. Many areas were flooded to a depth of two metres. Most victims drowned or were killed in landslides. By 30 July the death toll from the monsoon rains had reached 64 with 75 injured and three missing. Press reports indicated that Chataan left seven persons dead or missing in southern Japan. Heavy rains caused the flooding of thou- sands of homes, triggered landslides, and forced the cancellation of many high-speed "bullet" trains and scores of domestic flights. Over 100,000 persons were forced to evacuate their homes in areas north of Tokyo. Well-known storm and volcano chaser, photographer, and videographer Geoff Mackley was in Japan when Typhoon Chataan brushed the coast. Several nice photos and a write-up of Geoff's experiences with Chataan can be found at the following URL: (Report written by Gary Padgett) TYPHOON RAMMASUN (TC-09W / TY 0205 / FLORITA) 28 June - 6 July ------------------------------------------------ Rammasun: contributed by Thailand, is the Thai god of thunder Florita: PAGASA name, is a feminine name A. Storm Origins ---------------- Typhoon Rammasun developed simultaneously with Typhoon Chataan farther east in the vicinity of Chuuk. The development of both systems was related to a strong surge in the monsoonal westerlies. An updated STWO issued by JTWC at 1700 UTC on 25 June noted that an area of convection had developed and persisted approximately 100 nm east of Palau. Various satellite sensor platforms revealed deep convection embedded in the mon- soon trough with some mid-level turning evident but with no apparent LLCC. A concurrent 200-mb analysis indicated that the region was experi- encing moderate vertical shear but also with moderate upper-level diver- gence. At 26/0600 UTC the disturbance was relocated to a position about 140 nm east-northeast of Palau. There were indications that a possible LLCC was embedded in a line of convergent flow associated with the enhanced equatorial westerlies and tropical easterlies. At 27/0100 UTC the system was relocated to about 40 nm east-northeast of Palau and the development potential was upgraded to fair. The disturbance gradually increased in organization as shear lessened and a weak ridge to the north provided good diffluence aloft. JTWC issued a TCFA at 27/2100 UTC, placing the weak LLCC about 75 nm west- northwest of Yap. Both PAGASA and JMA classified the system as a trop- ical depression at 28/0000 UTC with PAGASA naming the depression Florita. Tropical Depression Florita was centered roughly 100 nm west of Yap at the time. At 0600 UTC JTWC issued their first warning on TD-09W, placing the center approximately 150 nm west of Yap or about 130 nm north- northeast of Palau and moving westward at 9 kts. Deep convection was still cycling but was becoming more organized--a 28/0110 UTC TRMM pass had depicted a curved band of deep convection in the western quadrant. During the 28th Tropical Depression Florita/09W took on a north- westerly heading--by 1800 UTC the system was centered approximately 200 nm north-northwest of Palau. JTWC upped the intensity to 30 kts, based on CI estimates of 30 and 35 kts. Outflow over the depression was being enhanced by a longwave trough anchored over the Philippine Sea. Two significant things happened at 0600 UTC on 29 June. Based on a SSM/I pass and synoptic data, the center was relocated about 120 nm to the east of the previous warning position to a point approximately 160 nm north- west of Yap, moving east-northeastward at 6 kts. Secondly, JTWC, NMCC, and JMA all upgraded the depression to tropical storm status with JMA assigning the name Rammasun. B. Track and Intensity History ------------------------------ The east-northeastward motion turned out to be temporary--by 1200 UTC Tropical Storm Rammasun was drifting slowly northwestward. Several bursts of deep convection were seen occurring, but the LLCC remained broad and difficult to locate, apparently east of the deepest convection. At 1800 UTC the cyclone's center was approximately 210 nm northwest of Yap and still drifting northwestward at 5 kts. A SSM/I pass several hours earlier had indicated that the LLCC was under the northeastern edge of the deep convection, which was improving in organization. JTWC upped the MSW to 45 kts at 30/0000 UTC and to 50 kts at 1200 UTC, based on CI estimates of 45 and 55 kts and QuikScat data which indicated winds near 50 kts in the southern sector. The storm by this time was located approximately 700 nm east-southeast of the island of Luzon, moving north- westward at 8 kts. Earlier animated visible imagery had indicated that the LLCC was partially-exposed north of the deep convection. (JMA's 10-min avg MSW estimates during this period ran generally 5-10 kts higher than JTWC's 1-min avg MSW, and NMCC's and PAGASA's 10-min avg MSW values were roughly equivalent to or slightly higher than JTWC's intensities.) The tropical cyclone gradually became better organized as it continued steadily northwestward on 1 July. Satellite intensity estimates had risen to 55 and 65 kts by 1200 UTC, and a SSM/I pass revealed a devel- oping banding eye feature. Both JTWC and JMA upgraded Rammasun to typhoon status as 1200 UTC with the center located roughly 550 nm east of the northern tip of Luzon. (PAGASA had upgraded the storm to typhoon status 12 hours earlier at 0000 UTC, and NMCC did so at 1800 UTC.) Typhoon Rammasun tracked steadily northwestward at around 12-14 kts all through 2 July, gradually intensifying. JTWC upped the intensity to 85 kts at 02/0000 UTC and to 90 kts at 1200 UTC. During this period satellite imagery revealed that the storm had developed concentric eyewalls. Current intensity estimates were already at T5.5 (102 kts), and at 1800 UTC the MSW was upped to 105 kts. The typhoon was then located 240 nm south of Kadena AB, Okinawa. A 02/1303 UTC SSM/I pass had revealed a well-defined eyewall with a second banding feature forming in the southwestern quadrant. Typhoon Rammasun reached its peak intensity of 110 kts at 03/0000 UTC when it was centered approximately 190 nm south-southwest of Kadena AB, or roughly 275 nm east-southeast of Taipei, Taiwan. JTWC's MSW was based on CI estimates of 102 and 115 kts. The storm remained at peak intensity for 30 hours, but even so the eyewall at times appeared to be weakening. Rammasun continued to track northwestward on the 3rd, steered by a subtropical ridge to the east, but at a slower pace, and by 1800 UTC the storm had turned more toward the north-northwest into a weakness in the ridge. The center was located about 160 nm west- southwest of Naha at 1800 UTC, and although CI estimates were still up to 115 kts, animated infrared imagery indicated that the eyewall con- tinued to weaken. (The peak 10-min avg MSW for Rammasun estimated by JMA was 85 kts with an attendant CP of 945 mb. NMCC estimated the peak intensity at 100 kts while the maximum 10-min avg MSW reported by HKO and CWBT was 90 kts. PAGASA's estimated intensity had reached 80 kts before the typhoon left its AOR.) By 0600 UTC on 4 July Rammasun was located a little more than 200 nm northwest of Okinawa, moving north-northwestward at 10 kts. The MSW was reduced to 105 kts based on CI estimates of 102 kts. Satellite animation indicated that a quasi-stationary longwave trough over the Chinese coast had begun to influence the western semicircle of the storm with convection rapidly weakening. At 1200 UTC the weakening cyclone was located 160 nm southeast of Shanghai and moving northward quickly at 17 kts. The MSW was lowered to 90 kts, and further to 80 kts at 1800 UTC, based on CI estimates of 65 and 77 kts. At 0000 UTC on the 5th Rammasun was located 150 nm east-northeast of Shanghai and was barely of typhoon intensity. JTWC, JMA and NMCC all downgraded the weakening cyclone to tropical storm status at 0600 UTC when it was located about 130 nm west-southwest of Cheju Do off the Korean coast. Water vapor imagery and upper-level wind analysis indicated strong southerly winds weakening the system's associated convection. By 05/1200 UTC Rammasun was accelerating north-northeastward in response to a major shortwave trough making its way toward the Russian coast. Enhanced infrared imagery depicted a fully-exposed LLCC decoupled well to the southwest of the rapidly weakening convection. JTWC lowered the MSW to 45 kts--JMA's and NMCC's intensity estimates were somewhat higher, as they often are for storms undergoing extratropical transition. Rammasun's center was located about 90 nm southwest of Kunsan, South Korea, at 1800 UTC, moving north-northeastward at 12 kts. The storm made landfall at 05/2330 UTC just south of the city of Sosan. JTWC estimated the MSW at landfall to be 35 kts, based on synoptic reports from nearby stations. At 0600 UTC on 6 July the center was inland east-northeast of Seoul, and by 1200 UTC had emerged into the Sea of Japan. JTWC and NMCC issued their final warnings at 0600 UTC, but JMA carried Rammasun through a couple more warning cycles, finally declaring the system a 35-kt extratropical gale in the Sea of Japan at 1800 UTC. C. Meteorological Observations ------------------------------ Typhoon Rammasun passed directly over the island of Mijako-jima (WMO 47927) shortly before 1200 UTC on 3 July. A minimum SLP of 945.5 mb was recorded at 03/1130 UTC during passage of the eye. The maximum 10-min mean wind of 49.8 kts was recorded at 03/1050 UTC, and the peak gust of 91.4 kts at 03/1049 UTC, during passage of the northern eyewall. However, Roger Edson pointed out that the southern eyewall was stronger than the northern one at the time Rammasun was near the island. Even though Okinawa was nearly 180 nm east of the typhoon's center, winds of 50 kts almost reached the island. Huang Chunliang has sent me an extensive amount of information about the typhoon's effects in China. Rainfall amounts in the Shanghai area were less than 75 mm at all stations. Several reporting stations outside the urban area reported peak wind gusts exceeding gale force: Pudong 35 kts Minhang 47 kts Jinshan 37 kts Congming 49 kts Qingpu 37 kts Nanhui 54 kts Baoshan 43 kts Changjiangkou 58 kts Fengxian 43 kts (Mouth of the Chang Jiang River) Some other miscellaneous observations Chunliang gleaned from local news reports include: (1) Dachen Dao, Putuo, Shengsi and Shipu, of Zhejiang Province, all reported wind gusts exceeding 85 kts as the storm passed by. (2) Wind gusts in Dachen Dao and Shenjiamen of Zhejiang Province reached 89 kts and 87 kts, respectively. (3) Stations Xiangshan and Xiajialing in Ningbo City, Zhejiang Province, recorded storm rainfall totals of 105 mm and 225 mm, respectively. (4) A station located on the coast of Sizapou of Cixi, Ningbo City, recorded a peak gust of 60 kts in the storm. Chunliang also sent the following rainfall totals from Taiwan (all amounts in mm): 1. Storm totals, from 02/1600 UTC through 03/0600 UTC: TA-CHI, Taoyuan County 64 2. Storm totals, from 02/1600 UTC through 03/0900 UTC: NEO-TSU, Hsinchu County 173 3. For the 23-hour period ending at 03/1500 UTC: KUANG-WU, Miaoli County 331 PAI-SHIH, Hsinchu County 246 PA LING, Taoyuan County 189 TAIPIN-S, Ilan County 212 SHEUILIN, Taichung County 133 PINGLING, Taipei County 115 PI-HU, Taipei County 119 4. Storm totals, from 02/1600 UTC through 04/0900 UTC: MATALA, Miaoli County 681 KUANG-WU, Miaoli County 582 PAI-SHIH, Hsinchu County 518 SHEUILIN, Taichung County 397 PA LING, Taoyuan County 354 CHUTZHU 292 TAIPIN-S, Ilan County 275 PINGLING, Taipei County 264 FEITSUI, Taipei County 254 TIEN-MU 210 Taipei City 125 (A very special thanks to Chunliang for sending the reports.) D. Damage and Casualties ------------------------ In the Shanghai area, strong gales and heavy rainfall battered the city, killing five people and injuring several dozen. The five killed were migrant workers taking refuge in a construction shed on Jinxiu Road in Pudong. High winds blew the shed over, killing the five and injuring 44 others. In the Nanhui District a steel workshed collapsed, and the roof of a temporary house for migrant workers in Baoshan District was blown away. The winds of Rammasun blew down approximately 270 trees along roads and streets despite the work local gardening authorities had done to strengthen supports for thousands of roadside trees. At the Pudong International Airport more than 200 flights were cancelled, and at the Hongqiao Airport 24 flights were cancelled and 47 postponed. In Chongming, just to the north of the city of Shanghai, an elderly woman was killed when a wall was blown over by a gust of wind. In the Nanhui District winds caused the collapse of the sheds of 333 hectares of fruits and vegetables. (I take this to likely mean greenhouses where crops were being grown.) Also in Nanhui winds felled 165 trees along the streets and brought down more than 50 power lines, plunging part of the district into darkness. In Zhejiang Province (along the coast immediately south of Shanghai) direct economic losses were estimated to have been approximately 700,000,000 yuan. In addition, two residents were reported missing there. (I do not have available any monetary damage estimates for the Shanghai area.) Although the center of Typhoon Rammasun passed nearly 200 nm west of Okinawa, the storm had a large wind field which caused some problems for the island. JMA reported that wind gusts to 39 kts were recorded in Okinawa, causing a loss of power for some 10,000 households. Most airline flights were cancelled, and bus services in Naha came to a complete standstill. (Report written by Gary Padgett) SUPER TYPHOON HALONG (TC-10W / TY 0207 / INDAY) 7 - 19 July ---------------------------------------------- Halong: contributed by Vietnam, is a famous picturesque place in Viet- nam, lying in the Bacbo Gulf and consisting of more than 1000 limestone isles Inday: a Filipino woman's nickname, usually applied to maids or helpers A. Storm Origins ---------------- The first tropical cyclone to actually form in the month of July, Typhoon Halong threatened to strike the island of Guam less than a week after the damaging Typhoon Chataan had passed over the island. However, the storm fortunately veered to the west as it approached Guam and passed to the south, sparing the island a direct strike. An area of convection developed just west of Eniwetok Atoll on 4 July. Multi- spectral animation depicted a LLCC developing along a monsoon trough stretching eastward from Typhoon Chataan. The disturbance remained quasi-stationary for the next day or two with new bursts of convection occasionally erupting near the LLCC. JTWC upgraded the development potential to fair at 05/2300 UTC when the system was located about 95 nm southwest of Eniwetok. A 05/1928 UTC TRMM pass revealed the development of two convective rain bands extending out from the LLCC, and the overall organization of the system was improving. By 06/0600 UTC the disturbance was located approximately 170 nm west- southwest of Eniwetok. Animated visible imagery showed a well-defined LLCC surrounded by new deep convection in the southeast quadrant. JTWC issued the first warning on TD-10W at 07/0000 UTC with the center located about 100 nm north of Pohnpei, moving west at 11 kts. A 06/2256 UTC SSM/I pass had depicted a well-defined LLCC with a convective band in the southern semicircle. (I could not locate the TCFA issued by JTWC for this system. Obviously one must have been issued between the STWO at 06/0600 UTC and the first warning at 07/0000 UTC.) The depression initially moved westward, turning more to a west-northwesterly track by 1200 UTC. At 1800 UTC JTWC upgraded the system to a 35-kt tropical storm, located approximately 200 nm northwest of Pohnpei. This upgrade was based on CI estimates of 30 and 35 kts. A low to mid-level ridge north of the cyclone was forecast to continue steering it on a west- northwestward course for the next 48 hours. B. Track and Intensity History ------------------------------ Tropical Storm 10W intensified slowly at first. At 08/0000 UTC JMA upgraded the system to Tropical Storm Halong, and NMCC upgraded it to tropical storm status six hours later. JTWC upped the MSW to 40 kts at 08/0600 UTC and to 45 kts at 1800 UTC when Halong was centered about 350 nm east-southeast of Guam. A 08/1129 UTC SSM/I pass showed that deep convection in the equatorward semicircle was wrapping into the LLCC. Tropical Storm Halong continued moving in the general direction of Guam on 9 July as it steadily intensified. By 1200 UTC CI estimates had reached 65 kts, so JTWC upgraded Halong to typhoon status when it was located about 220 nm east-southeast of Guam. A 09/1115 UTC SSM/I pass revealed a developing banding feature, and deep convection around the system was increasing in areal extent. The intensity was increased to 75 kts at 1800 UTC and to 90 kts at 10/0000 UTC when Halong was centered about 80 nm south-southeast of the southern tip of Guam. The storm had taken a westward jog which alleviated the threat of a direct strike on the island. A 09/1943 UTC QuikScat pass had revealed a compact system with the highest winds in the eastern semicircle. (Note: NMCC upgraded Halong to typhoon intensity at 10/0000 UTC, but JMA did not do so until 1800 UTC.) The eye of Halong passed about 75 nm south of the southern tip of Guam around 10/0200 UTC, and by 0600 UTC was located about 70 nm south-southwest of that point. A SSM/I pass at 10/0855 UTC revealed a developing banding eye feature so the MSW was bumped up to 95 kts at 1200 UTC. By 1800 UTC Halong was located 150 nm west-southwest of Guam, moving west-northwestward at 9 kts. Animated water vapor imagery indicated that the system was being sheared from the northwest as an upper-level trough deepened north of the typhoon and a weak upper-level ridge extending over the Philippine Sea cut off its poleward outflow. A 10/1802 UTC TRMM pass depicted the LLCC on the northern edge of the deep convection. The deepening trough resulted in a decrease in convection in the storm's northwest quadrant on the 11th and JTWC lowered the MSW to 90 kts at 1200 UTC. By 1800 UTC convection had begun to increase once more in the northwest quadrant, and a 11/1705 UTC TRMM pass depicted a band of deep convection wrapping around the LLCC. JTWC increased the MSW back to 95 kts at 12/0000 UTC based on CI estimates of 90 and 102 kts and the appearance of an eye. Typhoon Halong at this time was located approximately 640 nm south- southwest of Iwo Jima, moving west-northwestward at 10 kts. The storm intensified steadily with the MSW being raised to 105 kts at 12/0600 UTC and to 115 kts by 1800 UTC. The typhoon at 1800 UTC was located about 650 nm southeast of Okinawa and tracking in the direction of the island at about 14 kts. Satellite CI estimates were 115 and 127 kts, and infrared imagery revealed a 34-nm diameter eye. JTWC increased the MSW to 130 kts at 13/0000 UTC based on CI estimates of 115 to 140 kts, making Halong the fourth super typhoon of the season. Based on JTWC's warnings, Super Typhoon Halong reached its peak intensity at 0600 UTC on the 13th when 135-kt winds roared around a 20-nm wide eye. Gales extended outward from the center 155 nm in the northwest quadrant and up to 225 nm in the northeast sector. The radius of 50-kt winds was estimated at 115 nm in the northeast quadrant and from 70 to 95 nm elsewhere. Halong was a large, severe typhoon. (Interestingly, JMA's and PAGASA's estimated 10-min avg MSW reached their peak of 85 kts at 12/1800 UTC and remained steady through 14/0000 UTC, not reflecting the sudden increase at 13/0000 UTC reported by JTWC. Also, NMCC's maximum 10-min avg wind was estimated at 90 kts from 12/1200 through 14/0600 UTC. The minimum CP estimated by JMA was 945 mb.) The JTWC forecast issued at 13/0600 UTC was a very grave one for Oki- nawa, bringing a 135-kt super typhoon very near the island in 24 hours. The storm was then located about 430 nm southeast of the island, moving northwestward at the rather quick pace of 21 kts. Fortunately for Oki- nawa, Halong began to weaken and its forward motion slowed somewhat. The MSW was decreased to 125 kts at 1200 UTC. Intensity estimates were 127 kts, and satellite imagery indicated that the deep convection had diminished some in the northern semicircle during the past few hours. Also, the eye diameter had expanded back to around 30 nm. By 0000 UTC on 14 July the storm had weakened to 115 kts. The eye had become ragged and cloud-filled with surrounding cloud top temperatures warming 15 deg Celsius during the previous six hours while the eye temperature had cooled 8 deg Celsius. Halong was located only 145 nm south-southeast of Okinawa and trekking northwestward at 14 kts. The center of Typhoon Halong was near the southwestern tip of Okinawa at 1200 UTC on 14 July. JTWC had decreased the MSW to 90 kts by this time--JMA and NMCC were estimating 10-min avg MSW values of 70 and 80 kts, respectively. SSM/I and TRMM data revealed that the deep convection was concentrated to the south of the LLCC. Animated infrared imagery indicated that Halong was being sheared and was rapidly weak- ening. The primary steering mechanism for Halong for several days had been a large subtropical ridge to the east of the system, but by the time of the storm's encounter with Okinawa it was moving northward into a weakness in the ridge. By 1800 UTC Halong had moved past Okinawa and was located 45 nm northwest of the island. Satellite CI estimates were 77 kts, so JTWC lowered the MSW to 80 kts. On 15 July Typhoon Halong accelerated northeastward along the north- western periphery of the steering ridge to the east as it came under the influence of the mid-latitude westerlies. The storm was being sheared, and by 15/0600 UTC the LLCC was beginning to decouple from the deep convection, most of which was confined to the southeastern quadrant of the system. At 1200 UTC Halong was located approximately 180 nm south- east of Sasebo, Japan, moving east-northeastward at 19 kts. All three warning agencies downgraded the cyclone from typhoon status at this time. The storm was beginning the transition into an extratropical cyclone with the LLCC fully-exposed. JTWC issued their final warning on Halong at 1800 UTC when the center was about 180 nm south-southwest of Kyoto. NMCC and JMA, however, maintained Halong as a tropical system through 16/0600 and 17/0000 UTC, respectively. The storm was located approximately 85 nm southwest of Tokyo at 16/0000 UTC, and by 0600 UTC had sped across southeastern Honshu and was over the Pacific well north- east of Tokyo. JMA declared Halong extratropical at 17/0600 UTC when the system was located over the northern Kuril Islands. The gale center subsequently turned eastward, and the final reference to the system in JMA's High Seas Bulletins was at 19/0600 UTC when it was a weakening 35-kt gale south of the southeastern Aleutians. C. Meteorological Observations ------------------------------ Mark Lander reports that winds gusted to 55-60 kts over the southern portion of Guam as Typhoon Halong passed to the south. The storm also brought waves up to 5 metres to the southeast side of the island--bigger than any produced by the earlier Typhoon Chataan. According to some information sent to me by Jose Garcia of Puerto Rico, Agana International Airport on the northwestern side of Guam recorded sustained winds of 37 kts at 10/0300 UTC with an attendant MSLP of 999 mb. Two hours later the winds at Agana were 28 kts, gusting to 41 kts, with a pressure of 998 mb. Karl Hoarau passed along some observations from Okinawa. The northern eyewall of Halong was not very active due to strong shear. Winds were stronger after the passage of the eye. At Naha the peak wind (10-min avg) of 48 kts, gusting to 62 kts, occurred at 14/1300 UTC with a minimum SLP of 959 mb in the eye at 1100 UTC. Kadena AB recorded a peak MSW (1-min avg) of 85 kts with a peak gust of 99 kts at 1500 UTC while the station was in the convectively active eastern eyewall. Kadena's minimum SLP of 961 mb also occurred in the eye at 1100 UTC. The JTWC warning at 14/1200 UTC estimated the MSW to be 90 kts--the 85 kts recorded at Kadena is an excellent verification of the estimated intensity. As Halong sped over southeastern Honshu it passed near Tokyo around 16/0300 UTC. The Narita/Tokyo International Airport (RJAA) recorded wind gusts to 34 kts and a pressure of 979 mb. At 16/0319 UTC, RJAA reported sustained winds of 33 kts, gusting to 43 kts, with an attendant SLP of 980 mb. Around 16/0900 UTC a QuikScat pass showed a wide area of 20 to 40-kt winds southeast of Honshu. (This information was sent to me by Dickson Fu of Hong Kong.) (A very special thanks to Mark, Karl and Dickson for the information they provided.) D. Damage and Casualties ------------------------ Typhoon Halong apparently did not cause significant damage to Okinawa. Press reports stated that three persons were injured but none with life- threatening injuries. About 30,000 homes were without power for several hours. The storm also brought high winds and rains to Honshu, necessi- tating the evacuation of several thousand persons from their homes. Several persons were injured--the total injured in Japan (Okinawa and Honshu) was placed at nine according to several press reports. Well-known storm and volcano chaser, photographer, and videographer Geoff Mackley was in Naha when Typhoon Halong rumbled through. Several nice photos and a write-up of Geoff's experiences with Halong can be found at the following URL: (Report written by Gary Padgett) TROPICAL STORM NAKRI (TC-11W/ TS 0208 / HAMBALOS) 8 - 13 July ------------------------------------------------ Nakri: submitted by Cambodia, is the name of a flower Hambalos: Filipino name meaning 'a flogging with a cudgel or club' A. Storm Origins ---------------- The disturbance that became Tropical Storm Nakri was first mentioned in the JTWC's STWO issued at 07/1100 UTC when a broad circulation became evident in animated visible satellite imagery. Deep convective cells were located equatorward of the LLCC in an area of convergence associated with a southwesterly wind burst. JTWC upgraded the development potential to fair at 1700 UTC on 7 July when animated water vapor imagery depicted favourable outflow channels, although deep convection was still scattered. The system was then located about 150 nm southwest of Taiwan. Synoptic reports confirmed the presence of a closed circulation with the MSLP estimated at 994 mb. Satellite imagery at this time showed that the disturbance was organizing and associated deep convection was increasing. Based on this, JTWC issued a TCFA at 2300 UTC. The first JTWC warning on TD-11W was issued at 1200 UTC on 8 July with the system centred 130 nm southwest of Taiwan and moving northeastward at 3 kts. (HKO classified this system as a tropical depression at 0600 UTC, locating the centre 200 nm east-southeast of Hong Kong). A 08/1135 UTC TRMM pass depicted a well-defined LLCC with clearly defined rain bands in the western semicircle. B. Track and Intensity History ------------------------------ Tracking northeastward in a low to mid-level southwesterly monsoonal flow, TD-11W was upgraded to tropical storm status at 08/1800 UTC on the basis of satellite intensity estimates of 35 kts. JTWC, NMCC, and JMA all upgraded the system to tropical storm status with JMA assigning the name Nakri. (HKO upgraded Nakri to tropical storm status three hours later.) Visible animation and synoptic data showed Nakri to be a relatively small (approximately 90 nm in diameter) symmetric system. A 08/2153 UTC QuikScat pass indicated that the stronger winds were confined to the Taiwan Strait. Surface winds over the southern quadrant were 30 kts or less. A TUTT over Southeast Asia appeared to be temporarily enhancing the outflow over the system. (PAGASA issued three warnings on Nakri, beginning at 09/0000 UTC, dubbing the system Hambalos as it moved through their AOR.) Tropical Storm Nakri continued to move north- eastward at 9 kts and was just north of the tip of Taiwan at 0000 UTC on 10 July. Taiwanese radar imagery showed that the convection had sep- arated from the LLCC, but the centre later moved under the convection again. Due to interaction with land, Nakri weakened to a depression at 0600 UTC but recovered back to tropical storm strength by 1200 UTC as it began to move away from Taiwan. Nakri had changed to a more easterly course at a faster speed of 11 kt by 0000 UTC on 11 July as the MSW reached a maximum of 40 kt. Animated infrared satellite imagery showed deep convection rotating around the LLCC as it crept closer to Okinawa at a slower speed of 7 kts. (The system was still interacting with and being guided by the southwest monsoon flow on 12 July.) Animated visible imagery indicated that the system was becoming better organized at 1200 UTC on 12 July. The tropical storm had made its closest approach to Okinawa at 12/0900 UTC when the centre was 45 nm west-southwest of the island. A low to mid- level ridge strengthened to the east of Nakri, forcing the storm on a northward heading at 0000 UTC on 13 July. The weakening cyclone began to undergo extratropical transition at this time. JTWC issued the final warning on Tropical Storm Nakri at 13/0600 UTC 13 as the extra- tropical cyclone accelerated northeastward towards Japan. C. Meteorological Observations ------------------------------ Three Fuzhou stations: Changle, Fuqing, and Pingtan reported rainfall amounts exceeding 50 mm for the 24-hour period ending at 10/0000 UTC. (Pingtan reported 138.7 mm.) A non-Fuzhou station, Putian, also recorded more than 50 mm during the same period. The 3-day (07/0000-10/0000 UTC) totals recorded in the Fujian stations of Pingtan, Xianyou, Changtai, Xiamen, Zhangzhou, Zhangpu, Yunxiao and Dongshan all exceeded 100 mm. Among these, Pingtan County reported the highest rainfall accumulation with 162.3 mm being recorded. Chunliang also sent the following rainfall totals from Taiwan (all amounts in mm): 1. Storm totals, from 08/1600 UTC through 09/0600 UTC: KU HSIA, Pingtung County 73 TAI-AN, Hualian County 62 KUTINGKENG, Kaohsiung County 56 2. For the 23-hour period ending at 09/1500 UTC: DA-CHO-C, Ilan County 275 Ilan City 228 SU DU, Taipei County 258 Penghu 141 MA-CHIA, Pingtung County 136 ANPU 188 PU-LO-WAN, Hualian County 66 3. Storm totals, from 08/1600 UTC through 10/0000 UTC: MA-CHIA, Pingtung County 198 4. Storm totals, from 08/1600 UTC through 10/0300 UTC: HSIN WU, Taoyuan County 135 5. Storm totals, from 08/1600 UTC through 10/0600 UTC: Hsinchu City 125 6. Storm totals, from 08/1600 UTC through 10/1400 UTC: PENGJIA YU 647 ANPU 375 SU DU, Taipei County 297 DA-CHO-C, Ilan County 294 SHANGDER, Pingtung County 277 Ilan City 237 Penghu 222 FENCHIHU, Chiayi County 226 WUCHIHSHAN, Taipei County 206 HOU LUNG, Miaoli County 198 O MEI, Hsinchu County 165 YANG MEI, Taoyuan County 154 WU-TU, Keelung City 155 (A special thanks to Huang Chunliang of Fuzhou City, Fujian Province, China, for providing the above information.) D. Damage and Casualties ------------------------ Tropical Storm Nakri affected Guangdong, Fujian, and Zhejiang Prov- inces of mainland China and the Taiwan area. The storm was responsible for two deaths in Taiwan with eleven injured and one missing. No casualties were reported on the mainland, nor was any significant damage reported. Media reports indicate that 61 people died in the Philippines as a result of Tropical Storm Nakri/Hambalos. (However, fatalities in the Philippines which the press attributed to the tropical cyclones were a result of heavy monsoonal rains, which no doubt were enhanced by the storms, but it seems somewhat misleading to specify a particular number of deaths related to a given cyclone--editor's note.) (Report written by Kevin Boyle) SUPER TYPHOON FENGSHEN (TC-12W - TY 0209) 14 - 28 July ------------------------------------------ Fengshen: contributed by China, means 'God of Wind' in Chinese A. Storm Origins ---------------- Super Typhoon Fengshen was a long-lived intense typhoon which traveled from its birthplace near the eastern Marshall Islands to a final landfall as an innocuous depression on China's Shandong Peninsula. Based on JTWC's analysis the storm maintained a MSW of 130 kts for a period of 120 hours--six hours longer than Joan of 1997, the previous record holder for the longest continuous period of winds greater than or equal to 130 kts. Fengshen developed quickly--increasing from a tropical depression to typhoon intensity in 12 hours per both JTWC's and JMA's warnings. At 14/0000 UTC JMA classified a weak LOW located approximately 225 nm north-northwest of Majuro Atoll as a tropical depression. A TCFA issued by JTWC at 0400 UTC noted that an area of convection had persisted for 12 hours with several satellite-based sensors indicating that deep convection had rapidly organized over a distinct LLCC. Upper- level analysis indicated good divergence aloft with weak vertical shear. (It is very unusual for JTWC to issue a TCFA for a system which has not been previously mentioned in STWOs.) JTWC issued the first warning on TD-12W at 14/0600 UTC, just two hours after the issuance of the TCFA. The depression's center was located 275 nm north-northwest of Majuro, moving northwestward at 6 kts and with peak sustained winds estimated at 25 kts. Winds were upped to 30 kts at 1200 UTC--a TRMM pass at 14/1000 UTC had depicted a well-defined LLCC with a curving rainband in the southern semicircle. At 1800 UTC TD-12W was upgraded to a tropical storm with 45-kt winds, based on CI estimates of 35 to 55 kts. The cyclone's center was located about 325 nm north-northwest of Majuro, or about 260 nm north-northeast of Kwajalein. Animated satellite imagery and microwave data suggested a rapidly developing midget tropical cyclone. Enhanced infrared imagery depicted a warm center to the south of a tight banding feature in the northeast quadrant. B. Track and Intensity History ------------------------------ On the very next warning (15/0000 UTC) JTWC upgraded TS-12W to an 80-kt typhoon based on CI estimates of 65 to 90 kts. Visible imagery depicted a 7-nm ragged eye embedded 50 nm within the deep convection. The storm was located approximately 350 nm north-northwest of Majuro, moving north-northeastward at 7 kts. The 34-kt and 50-kt gale radii were estimated at 60 nm and 20 nm, respectively. Also at 15/0000 UTC, JMA upgraded the system to a 40-kt tropical storm, naming it Fengshen. However, six hours later, JMA upgraded Fengshen to a 70-kt typhoon. By 1200 UTC the storm was tracking slowly northward as it steadily intensified--winds were up to 95 kts at 1800 UTC (per JTWC's warning). The small eye appeared to be centered within the CDO. As the subtropical ridge to the north strengthened, Typhoon Fengshen slowly turned to the west. By 16/0600 UTC the storm was moving westward at 4 kts. Fengshen remained compact--a 16/1234 UTC TRMM pass depicted a rainband extending out 90 nm in the southeastern quadrant. JTWC upped the MSW to 105 kts at 0000 UTC, and by 1800 UTC winds had increased to 125 kts. (JMA's 10-min avg MSW estimate had reached 90 kts by this time, and NMCC raised theirs to 90 kts at 17/0000 UTC.) Typhoon Fengshen passed about 325 nm north of Kwajalein around 2100 UTC. On 17 July the storm continued moving pretty much due westward, guided by the ridge to the north and northwest. Around 1200 UTC Fengshen was located about 275 nm south of Wake Island. The typhoon's intensity held steady at 125 kts through the 17th--CI estimates ranged from 115 to 127 kts for most of the day. TRMM and SSM/I passes early in the day indicated that the system had undergone an eyewall replacement cycle during the previous few hours. Likely this was the reason for the halt to the intensifi- cation process. (JMA's reported intensity remained at 90 kts while NMCC upped the 10-min avg MSW to 100 kts at 0600 UTC.) By 0000 UTC on 18 July CI estimates were ranging from 127 to 140 kts, so JTWC increased the MSW to 130 kts, making Fengshen the fifth super typhoon of the year. The storm was centered at that time approximately 300 nm south-southwest of Wake Island. The intense typhoon continued moving westward throughout the 18th--by 1800 UTC it was located about 900 nm east of Saipan and was shifting to more of a west-northwesterly track. Satellite intensity estimates remained at T6.5 and T7.0. JTWC had increased the MSW to 135 kts at 0600 UTC where it remained for the remainder of the 18th. The storm had grown in areal extent--at 1800 UTC gales had expanded to cover a region about 250 nm in diameter. The eye remained fairly small, however. A SSM/I pass at 18/1042 UTC revealed at 15-nm diameter symmetrical eye. Super Typhoon Fengshen continued to track west-northwestward on the 19th as the subtropical ridge to its north re-oriented itself and a peripheral ridge developed. JTWC upped the MSW to an initial peak of 140 kts at 19/0000 UTC. During the day satellite imagery indicated that deep convection had expanded in areal coverage. (JMA's and NMCC's 10-min avg MSW estimates also reached initial peaks of 100 kts and 120 kts, respectively, on 19 July.) Dvorak intensity estimates remained a solid 140 kts throughout the day. By 1800 UTC Fengshen's center was located roughly 750 nm east of Saipan, still moving west-northwestward. The storm appeared to weaken very slightly on 20 July. An upper-level LOW located to the north-northwest of the typhoon had been enhancing outflow. This LOW tracked westward away from Fengshen, thereby decreasing the poleward outflow. However, a second upper-level LOW was forecast to track westward with the storm and improve ventilation. JTWC decreased the MSW to 135 kts for one warning cycle (at 20/1200 UTC), but raised it back to 140 kts at 1800 UTC. (JMA and NMCC decreased their 10-min avg MSW estimates to 95 kts and 110 kts, respectively, on the 20th.) As time progressed, Fengshen's direction of motion slowly took on more of a northwesterly component. The eye diameter, which had held steady around 15 nm for several days, had increased to 24 nm by 1800 UTC. Fengshen was then centered approximately 665 nm east-northeast of Saipan, moving northwestward at 8 kts. The storm tracked steadily northwestward at around 10-12 kts all through 21 July. The 21/0000 UTC warning noted that the eye diameter had shrunk back to 16 nm. Poleward outflow increased slightly as an upper-level LOW northwest of Fengshen propagated westward, and equatorward outflow was enhanced as another upper-level LOW approached from the east. JTWC increased the MSW to the peak value of 145 kts at 1200 UTC, based on CI estimates of 140 and 155 kts. (JMA and NMCC had also upped the 10-min avg MSW estimates back their respective peaks of 100 and 120 kts by this time.) Fengshen was located about 750 nm east- southeast of Iwo Jima, or about 525 nm east of the northernmost Marianas Islands. Storm-force winds were estimated to extend out 80 nm from the eye in all directions, and gales extended outward 145 nm in the northeast quadrant and 100 nm in the southwest sector. JTWC lowered the MSW back to 140 kts at 1800 UTC, even though at least one CI estimate was still at 155 kts. (The remarks in the 1800 UTC warning are practically identical to those in the 1200 UTC warning--no explanation was given as to why the intensity was dropped slightly.) Fengshen continued northwestward on the 22nd, but by 1800 UTC the track was beginning to take on more of a westerly component. The MSW was maintained at 140 kts through the 0600 UTC warning. The remarks in that warning noted that outflow was still good in all quadrants as convection wrapped into the upper-level LOWs located to the northwest and east. The typhoon at that time was approximately 550 nm east of Iwo Jima. The intensity was lowered to 130 kts at 1200 UTC--a 22/1127 UTC SSM/I pass indicated that the eye had become irregular, and at least one CI estimate had dropped to 127 kts. By 1800 UTC Fengshen was located approximately 400 nm east of Iwo Jima and was moving west-northwestward at 14 kts. JTWC was still estimating the MSW at 130 kts, based on CI estimates of 127 kts, but JMA and NMCC had lowered their 10-min avg MSW values to 90 and 110 kts, respectively. Weakening began in earnest on 23 July. The MSW was reduced to 120 kts at 23/0000 UTC. Dvorak estimates had dropped to 115 kts, and SSM/I data indicated decreased convection in the southern semicircle and an eroding eyewall. The intensity continued to decline and was lowered to 95 kts at 1200 UTC when Fengshen was centered approximately 135 nm east of Chichi Jima. Satellite intensity estimates were 90 and 102 kts, and a 23/1113 UTC SSM/I pass revealed that the southern quadrant was virtually convection-free due to dry air intrusion. Also, the TUTT situated earlier to the northwest of Fengshen had drifted southwestward and was positioned due west of the cyclone, leading to a significant diminution in the outflow over the north quadrant. The storm's west-northwesterly motion had gradually increased and by 1800 UTC Fengshen was zipping along at 19 kts. This was due to a strengthening mid-level ridge over central Japan and also possibly due to interaction with Typhoon Fung-wong, which at 1800 UTC was located approximately 765 nm west-southwest of Fengshen. At 0000 UTC on 24 July Fengshen's center was located approximately 225 nm north of Iwo Jima, moving west-northwestward at 16 kts. JTWC reduced the MSW to 80 kts (JMA and NMCC were reporting 70 kts), and the intensity continued to drop throughout the day. By 1800 UTC Fengshen was a minimal typhoon located about 400 nm southeast of Sasebo, Japan, and sailing westward at 18 kts as it was guided by a mid-level ridge extending from the Korean Peninsula to a position east of Japan. A 24/0326 UTC TRMM 37-GHz image depicted a large irregular eye, and a SSM/I pass at 24/1114 UTC indicated that the eye had opened up. Satellite intensity estimates exhibited a large spread, ranging from 55 to 90 kts at 1200 UTC and from 45 to 65 kts at 1800 UTC. At the latter hour the center of by-now Tropical Storm Fung-wong was located about 530 nm south-southwest of Fengshen. NMCC downgraded Fengshen to tropical storm status at 24/1800 UTC, and JMA followed suit at 25/0000 UTC. JTWC maintained Fengshen as a 65-kt typhoon in the 0000 UTC warning, but downgraded it at 0600 UTC. The storm was then centered approximately 230 nm southeast of Sasebo. The ridge to the north had weakened some, allowing Fengshen to track on a more west-northwesterly course. Even though CI estimates were decreasing, satellite imagery indicated an increase in deep convection over the western semicircle, and a 25/1226 UTC SSM/I pass revealed tightly-curved rainbands wrapping into the LLCC. Interestingly, while JTWC and JMA slowly decreased Fengshen's intensity, NMCC re-upgraded the storm to minimal typhoon status for a 12-hour period beginning at 25/0000 UTC, likely due to the rejuvenated convection noted above. At 1800 UTC Fengshen was located about 125 nm south of Sasebo, and the MSW was estimated at 55 kts, 55 kts and 60 kts by JTWC, JMA and NMCC, respectively. Tropical Storm Fung-wong had swung around to the south and at 1800 UTC was located 475 nm southeast of Fengshen's center. By 0600 UTC on the 26th the once powerful super typhoon was located about 60 nm south-southeast of Cheju Do. Winds were down to 45 kts (although JMA and NMCC were reporting higher 10-min avg MSW values of 50 and 60 kts, respectively, at that time). Visible satellite imagery indicated an exposed LLCC on the southern edge of deep convection. The cyclone continued moving generally west-northwestward across the southern Yellow Sea and gradually weakened with the LLCC becoming decoupled from the mid-level circulation. Convection continued to diminish and by 1800 UTC Fengshen was a minimal tropical storm located about 85 nm west-northwest of Cheju Do. The also-weakening Fung-wong was located approximately 460 nm east-southeast of Fengshen at 1800 UTC. JTWC downgraded Fengshen to a depression at 27/0000 UTC, based on CI estimates of 30 and 35 kts, and noted that the LLCC was completely- exposed and almost devoid of deep convection. The final JTWC warning was issued at 0600 UTC on the 27th, placing the center approximately 300 nm southeast of Beijing. The system was still tracking west- northwestward along the low to mid-level ridge to its north. JMA maintained Fengshen as a 35-kt tropical storm through 27/0600 UTC, and NMCC did so through 1200 UTC when the cyclone was approaching the coast. Fengshen made landfall on the southern Shandong Peninsula around 1700 UTC near Jiaonan City with the MSW (10-min avg) estimated at 30 kts by NMCC. The system quickly weakened to a depression and drifted northward back over water, the final NMCC position placing it northeast of the mouth of the mighty Huang He (Yellow River). The NMCC track for Fengshen ends at 28/1200 UTC, but the remnant LOW traveled north, then northeastward, bringing heavy rain to northeastern China. C. Meteorological Observations ------------------------------ Fengshen was the first tropical cyclone to have significant effects in the Beijing area since Rita of 1972. The main Beijing station (WMO 54511) recorded 35.3 mm of rain in the 24-hour period ending at 28/0000 UTC and 41.4 mm in the 48 hours ending at 29/0000 UTC. Leting (WMO 54539), in Hebei Province north of Beijing, recorded a 48-hour total of 41.1 mm, while Funing and Qinglong (both in Hebei Province) measured totals of 85 mm and 66 mm, respectively, in the 24 hours ending at 29/0000 UTC. A weather station at the Beijing Undergraduate Gym reported an accumulation of 57 mm in the 24 hours ending at 28/0000 UTC--the highest in the Beijing area. Heavier rains fell in the two northeasternmost Chinese provinces of Jilin and Heilongjiang as the remnants of Fengshen moved through that region. Some rainfall totals for the 24-hour period ending at 0000 UTC on 30 July include: Province Station Total Rainfall ----------------------------------------------- Jilin Songyuan 76.5 mm Fuyu 69.2 mm Dehui 62.3 mm Jiutai 62.2 mm Shulan 82.4 mm Antu 58.4 mm Jingyu 51.2 mm Yushu 104.9 mm Sanchahe 69.0 mm Heilongjiang Shuangcheng 74.0 mm Wuchang 61.0 mm (A special thanks to Huang Chunliang for sending me all the above rainfall information.) D. Damage and Casualties ------------------------ No reports of damage or casualties resulting from Typhoon Fengshen have been received. (Report written by Gary Padgett) TROPICAL STORM (TC-13W / JUAN) 18 - 23 July ----------------------------------- JTWC issued a STWO at 1200 UTC on 16 July describing an area of convection located approximately 130 nm northwest of the Republic of Palau. The development potential was considered poor. Animated satellite imagery had depicted unorganized deep convective activity several hours prior to the Outlook. A 16/0901 UTC QuikScat satellite pass showed a weak LLCC located just east of the deep convection, and a 200-mb analysis depicted good upper-level divergence with moderate vertical wind shear. The LLCC and its associated cycling scattered convection was moving slowly westward. JTWC upgraded the development potential to fair in the 17/1800 UTC STWO. Increasing organization of convection associated with the LLCC prompted JTWC to issue a TCFA at 2230 UTC. The first warning on TD-13W was issued at 1200 UT on 18 July by both JTWC and PAGASA, the latter agency assigning the name Juan. The system was now located approximately 50 nm east of Tandag in the Philippines (on Mindanao). Moving west- northwestward at 6 kts, the depression accelerated to 12 kts and turned north-northwestward towards the central Philippines. Tropical Depression Juan crossed Samar between 0000-0600 UTC on 19 July with the MSW esti- mated at 30 kts. A 19/1014 UTC SSM/I pass revealed that the depression was becoming more disorganized and that associated convection was decreasing. The system continued moving across the central Philippines at forward speeds varying from 6-11 kts. Animated satellite imagery revealed that new convection was developing over the depression at 0000 UTC on 20 July, and 12 hours later, Juan/13W was upgraded to a tropical storm by JTWC, located approximately 65 nm south of Manila. (JMA did not upgrade this system to tropical storm status, hence, there is no international name.) Tropical storm intensity lasted only 6 hours, and on the 1200 UTC warning, Juan/13W was downgraded to a depression. Visible satellite imagery on 21 July was showing an exposed LLCC east of Luzon with the associated cloud shield of Juan being sheared west into the South China Sea. At 21/0000 UTC synoptic reports depicted a complex system with three separate LLCCs: one southwest of Manila, one east of Luzon, and a third west of Baguio in the Lingayen Gulf. Tropical Depression Juan passed over Manila and Tarlac at 0000 UTC and 0600 UTC on 21 July, respectively, on a northwesterly to northerly heading with the MSW estimated at 30 kts. Convection continued to be sheared westward away from the LLCC, and the weakening vortex was guided northward at 7 kts west of the Cordillera mountain range (peaks to 2000 m) throughout 21 July. The system managed to move into the South China Sea approximately 240 nm north-northwest of Manila at 0000 UTC on 22 July, but convection was dissipating as well as the LLCC. JTWC issued the final warning at this time. PAGASA continued to issue advisories until 23/1200 UTC on the remnant LOW which remained quasi-stationary off the northwestern coast of Luzon. The remnants of Juan were finally mopped up by Typhoon Fung-wong on 24 July. According to press reports taken from the Manila Bulletin, Tropical Storm Juan caused 18 deaths, three of which were attributed to land- slides and tornadoes spawned by the tropical cyclone. Heavy rains caused flooding, forcing the evacuation of over 2400 people. Quezon, Southern Tagalog, and central Luzon (including the Manila area) were the hardest hit regions. Damage was estimated at 12.1 million pesos, equivalent to $0.24 million. (Report written by Kevin Boyle) TYPHOON FUNG-WONG (TC-14W / TY 0211 / KAKA) 19 - 27 July --------------------------------------------- Fung-wong: contributed by Hong Kong, means 'Phoenix', and also is the name of a peak in Hong Kong Kaka: PAGASA name, is a Filipino feminine nickname A. Storm Origins ---------------- Typhoon Fung-wong ran its complete course within the lifetime of Super Typhoon Fengshen, and during its latter stages underwent a Fujiwhara interaction with the larger and stronger tropical cyclone. Fung-wong described a large counterclockwise loop between the Marianas and Ryukyu Islands. (PAGASA applied the name Kaka to the cyclone during the time it was located within the Philippines' AOR--west of 135E and south of 25N.) JTWC issued an interim STWO at 2200 UTC on 18 July for an area of convection located a few hundred miles northeast of the Marianas Islands. Animated infrared imagery depicted scattered deep convection associated with a small LLCC. A 200-mb analysis indicated an area of diffluence aloft associated with the southwestern quadrant of an upper-level ridge. The development potential was assessed as fair. JMA classified the disturbance as a 30-kt tropical depression at 0000 UTC on the 19th. The system was centered roughly 325 nm east-northeast of the northern Marianas and was moving slowly in a general westward direction. Twenty-four hours later JMA's estimated intensity was still 30 kts, and the depression had reached a location approximately 250 nm east-southeast of Iwo Jima. A new area of convection had developed in the region, and animated water vapor imagery indicated an upper-level LOW enhancing outflow over the disturbance. A 19/2033 UTC QuikScat pass showed broad troughing extending northeastward from TS-13W/Juan, while a 200-mb analysis depicted moderate equatorward diffluence and weak vertical shear. Yet, JTWC downgraded the development potential to poor at 20/0000 UTC. However, a TCFA was issued at 20/1030 UTC for the LLCC then located about 80 nm southeast of Iwo Jima. Convection was improving in organization and a 20/0822 UTC SSM/I pass had depicted a compact area of deep convection near the LLCC. JTWC issued the first warning on TD-14W at 1200 UTC, placing the center approximately 40 nm south- southwest of Iwo Jima and moving westward at 12 kts. The initial warning intensity was set to 25 kts, based on CI estimates of 25 and 30 kts. Convection continued to increase in coverage and organization, and at 1800 UTC TD-14W was upgraded to tropical storm status. The new tropical storm was located about 70 nm west-southwest of Iwo Jima, moving west at 10 kts with the MSW estimated at 35 kts. B. Track and Intensity History ------------------------------ Both NMCC and JMA upgraded the system to tropical storm status at 0000 UTC on 21 July with JMA assigning the name Fung-wong. The fledgling tropical storm was then located approximately 160 nm west-southwest of Iwo Jima and moving west-southwestward at 13 kts. This motion was forecast to continue for a day or so as Fung-wong was guided by a low- level ridge to the north. Within 24 to 48 hours the storm was expected to slow down and begin to move southwestward as Super Typhoon Fengshen approached from the east and began to interact with Fung-wong. During the 21st organization improved and outflow increased--JMA quickly raised the intensity (10-min avg) to 50 kts, but JTWC's warnings reflected only a modest increase in intensity to 40 kts. (NMCC's 10-min avg MSW remained at 35 kts on 21 July.) A general westerly motion ensued on the 22nd--at 0000 UTC Fung-wong's center was about 415 nm east-southeast of Kadena AB on Okinawa, and by 1800 UTC had reached a point 325 nm southeast of Kadena. JTWC increased the MSW to 50 kts at 0000 UTC, but the intensity then remained fairly steady, fluctuating between 50 and 55 kts for about 24 hours. There was some evidence of northwesterly shearing during the day. (JMA's 10-min avg MSW remained at 50 kts while NMCC increased their estimate to 45 kts.) As early as 22/0000 UTC JTWC's forecast hinted at the anticipated turn to the south and then east as the intense Fengshen approached. At 1800 UTC Fung-wong's center lay about 900 nm almost due west of Super Typhoon Fengshen. (Fung-wong had entered PAGASA's AOR at 22/0000 UTC and was known locally in the Philippines as Tropical Storm Kaka.) A 22/2207 UTC SSM/I pass revealed a developing banding eye feature, which turned out to be transient. However, an eye later began to reappear, and with CI estimates having reached 65 kts by 23/0600 UTC, JTWC upgraded Fung-wong to typhoon status. The storm was still tracking slowly westward from a position about 280 nm southeast of Kadena, but by 1800 UTC Fung-wong was tracking south-southwestward, apparently beginning to feel the influence of Fengshen which was then located 765 nm to the east-northeast. A very small 8-nm diameter eye was revealed by a SSM/I pass at 23/1253 UTC. By 1800 UTC NMCC and JMA had increased their respective 10-min avg MSW estimates to 70 and 75 kts, but JTWC's 1-min avg MSW never rose above minimal typhoon intensity of 65 kts--definitely an unusual situation. (More discussion on this below.) JTWC downgraded Fung-wong to a 60-kt (1-min avg) tropical storm at 0000 UTC on 24 July. The storm was then located approximately 300 nm south-southeast of Kadena, moving south-southwestward at 5 kts. (JMA and NMCC, however, were still maintaining Fung-wong as a 70-kt typhoon.) A 23/2357 UTC SSM/I pass revealed that the LLCC was partially-exposed with the deep convection concentrated in the southern semicircle. Satellite intensity estimates remained at 55 kts, but JTWC lowered the MSW to 50 kts at 1200 UTC based on QuikScat data which revealed a very small, isolated area of 50-kt winds beneath deep convection decoupled to the west of the center. (NMCC had by this time downgraded Fung-wong to tropical storm status and JMA did so at 1800 UTC. PAGASA, however, which had also classified Fung-wong/Kaka as a typhoon, did not downgrade until 25/0000 UTC.) During the 24th the cyclone's track rather sharply curved to the south and then east-southeast, and by 1800 UTC Fung-wong was located about 410 nm southeast of Kadena, moving eastward at 8 kts. Fengshen, by then a minimal typhoon, had approached to within 530 nm to the north-northeast of Fung-wong. Tropical Storm Fung-wong's intensity remained pegged at 50 kts through 25 July per JTWC's warnings. (NMCC's 10-min avg MSW estimate remained at 60 kts while JMA's had come down to 55 kts by 25/1200 UTC.) Satellite CI estimates were also plateaued at 55 kts through the day, and even though a 24/2345 UTC SSM/I pass had shown some indications of the development of an eye, this apparently did not materialize. Animated infrared imagery revealed that the LLCC had become completely-exposed by 1200 UTC. Although still trekking eastward at 0000 UTC, by 0600 UTC the cyclone had turned abruptly to the north-northeast, apparently due to the influence of Fengshen and strengthening southwest monsoonal flow. Sometime around 1800 UTC Fung-wong crossed its track of about four days earlier, completing a fairly large cyclonic loop. This motion swung the storm around the southern side of Fengshen's circulation, and at 1800 UTC Fung-wong was located approximately 475 nm southeast of the weakening Fengshen, which by then was located southwest of Kyushu. After completing the loop, Fung-wong tracked north-northwestward in the general direction of Kyushu as it was guided by a low to mid-level subtropical ridge situated over Japan. At 0000 UTC on 26 July the storm was centered approximately 340 nm east of Kadena AB. JTWC upped the MSW to 55 kts based on CI estimates of 45 to 65 kts--the LLCC had become partially-exposed on the northeastern edge of the deep convection. However, six hours later the center was once more fully-exposed and the intensity was reduced. Weakening continued through the 26th--by 1800 UTC the MSW was down to 35 kts (JMA and NMCC both were reporting 40 kts). Satellite imagery indicated that the deep convection was limited to a small area west of the LLCC. Fung-wong was then located roughly 250 nm south-southeast of Sasebo, moving northwestward at 17 kts. Fengshen, also a minimal tropical storm, was located 460 nm to the west-northwest. JTWC downgraded Fung-wong to a 30-kt depression at 27/0000 UTC, based on CI estimates of 25 and 35 kts, and the final warning from that agency was issued at 0600 UTC, placing the center about 110 nm south of Sasebo. The system had weakened considerably in unfavorable horizontal and vertical shear conditions. A 27/0352 UTC TRMM pass had revealed that the LLCC was free of deep convection, and synoptic reports indicated the central pressure had risen to near 1000 mb and that winds associated with the center were 20 to 25 kts. JMA also downgraded Fung-wong to a depression at 0600 UTC and issued their final warning. NMCC maintained the cyclone as a minimal tropical storm through 1200 UTC, but issued no more warnings, so presumably it dissipated shortly thereafter. The final NMCC position placed Fung-wong about 75 nm southeast of Cheju Do off the Korean coast, and about 390 nm east-southeast of the weakening Fengshen which was about to make landfall in China. C. Intensity Comparisons ------------------------ Although I've decided to omit the section I've been including for more than a year comparing intensity estimates between various warning centers, the unusual situation with Fung-wong warrants a little more discussion. Anyone who has been monitoring Western Pacific tropical cyclones for very many years and comparing the intensities estimated by the various warning agencies knows that for storms near typhoon intensity and higher, JTWC's estimated MSW is almost always the highest after conversion to a common time averaging period--especially higher than JMA's intensity. The reverse was true for Fung-wong. Four TCWCs issued warnings on Fung-wong during its most intense phase. The respective peak intensity estimates, after converting to an equivalent 1-min avg MSW are: JTWC - 65 kts (23/0600 - 24/0000 UTC) PAGASA - 75 kts (24/0000 - 25/0000 UTC) NMCC - 80 kts (23/1200 - 24/0600 UTC) JMA - 85 kts (23/1800 UTC) & 80 kts (23/1200 - 24/1200 UTC) The higher intensities from NMCC and JMA have some support from SAB and AFWA. At 23/1243 and 23/1508 UTC, AFWA reported the intensity at T4.5/4.5 -- 77 kts. By 23/1828 the T-number had dropped to 65 kts while the CI was still 4.5. SAB was T5.0/5.0 -- 90 kts -- at 0832 and 1432 UTC on the 23rd. At 24/0832 UTC the T-number had dropped to 65 kts while the CI number was still 5.0. Normally the Remarks section in the JTWC warnings give the range of CI estimates. The warning at 23/1200 UTC noted that the warning intensity was based on CI estimates of 65 kts--very puzzling since at that time CIs from AFWA and SAB were 77 and 90 kts, respectively. Similarly the warning at 24/0600 UTC remarked that the MSW of 55 kts was based on CI estimates of 55 kts, but SAB was still reporting T4.0/5.0. (Thanks to Paul McCrone for providing me with the AFWA bulletins.) Karl Hoarau also commented on the Fung-wong situation. At 0732 UTC on 23 July the Objective T-number was 5.5, suggesting that Fung-wong was beginning to undergo a period of rapid intensification. Karl noted that based on visual analysis, AFWA's T4.0 and 0600 UTC and T4.5 at 1200 UTC (on 23 July) seemed appropriate. But taking into account enhanced infrared data, T5.0 was a better estimate. Karl is of the opinion that the 75 kts (10-min avg) from JMA, equivalent to a 1-min avg MSW of 85 kts, was a good compromise. Karl also thought it worth mentioning that visible Dvorak analysis from the DMSP satellites at 1200 UTC on the 23rd and 24th was possible because of moonlight. (That hour is about 2100-2200 local time in the Western Pacific.) The moon reached its full phase at 24/0907 UTC, providing enough illumination for visual analysis from the lower polar-orbiting satellites. The purpose of this little analysis is not to attempt to establish which TCWC did a better job or a poorer job of reporting Fung-wong's intensity--I personally am not qualified to do that--but just to document that there was support from SAB and AFWA for a more intense typhoon than reported by JTWC. D. Damage and Casualties ------------------------ No reports of damage or casualties resulting from Typhoon Fung-wong have been received. (Report written by Gary Padgett) TROPICAL STORM KALMAEGI (TC-15W / TS 0210) 20 - 21 July ------------------------------------------- Kalmaegi: contributed by North Korea, means 'sea gull' Tropical Storm Kalmaegi was an ephemeral tropical storm which briefly flourished just west of the Dateline. JTWC did not upgrade this system to tropical storm status, but both JMA and NMCC classified it as a tropical storm. A STWO issued by JTWC at 0600 UTC on 18 July noted that an area of convection previously located near 10N, 178W, was located farther east about 500 nm west-southwest of Johnston Atoll. (The STWO at 17/1800 UTC did not mention such a disturbance, so I can only conclude that an interim STWO was issued sometime between 17/1800 and 08/0600 UTC which I missed.) Deep convection was associated with convergence south and east of the LLCC, and a 200-mb analysis indicated that an upper-level LOW to the northeast was enhancing outflow. The development potential was upgraded to fair at 2200 UTC based on increasing organization of deep convection. The system was then located about 480 nm west-southwest of Johnston Atoll. JTWC issued a TCFA for the disturbance at 19/0100 UTC followed by a second one 24 hours later. At 20/0100 UTC the LLCC was located about 530 nm west-southwest of Johnston Atoll--convection was continuing to increase in organization near the LLCC. JMA classified the system as a tropical depression at 20/0600 UTC, placing the center approximately 600 nm west of Johnston Atoll and right on the Dateline at latitude 16N. Six hours later JMA upgraded the depression to Tropical Storm Kalmaegi, located about 650 nm west of Johnston Atoll. At 20/1800 UTC JTWC issued the first warning on TD-15W. The system was located roughly 675 nm east of Wake Island, moving northwest at 10 kts. The MSW was estimated at 30 kts, based on CI estimates of 25 and 35 kts. Convection had continued to increase and become better organized during the preceding six hours. Also at 1800 UTC, NMCC upgraded the system to tropical storm status and JMA raised the estimated 10-min avg MSW to 40 kts. However, by 21/0000 UTC animated water vapor imagery indicated that convection associated with Kalmaegi had decreased. JTWC issued their final warning on Kalmaegi at 0600 UTC. Visible animation and an earlier TRMM pass depicted a fully-exposed LLCC. Winds were estimated at no higher than 25 kts, and an upper-level analysis indicated that the moderate to strong convergent southerly winds associated with a TUTT to the northwest of the system had had a destructive effect on Kalmaegi. NMCC issued their final warning also at 0600 UTC, although with winds still estimated at 35 kts. At 0900 UTC JMA downgraded Kalmaegi to a depression and issued their final bulletin, placing the dissipating center about 600 nm south-southwest of Midway. (Report written by Gary Padgett) TROPICAL DEPRESSION 29 - 30 July --------------------------------------- A tropical depression formed near 19N, 110E, or over Hainan Dao at 0000 UTC on 29 July. Moving northwestward, the depression moved into the Gulf of Tonkin and from there into northern Vietnam on 30 July. The system then drifted northward and dissipated inland over southern China. MSW (10-min avg) were estimated 30 kts by JMA. All agencies handled the depression differently. This system was treated as a tropical depression primarily by JMA only--and briefly by the Thai Meteorological Department. NMCC did not issue any bulletins for this disturbance (although they usually do so for depressions close to the Chinese coast), but it was mentioned as a depression on a morning weather forecast. JTWC gave the system a fair development potential on 29 July but downgraded it to poor on the 30th. (CPHC also ranked the disturbance as a depression and mentioned it in their satellite discussion bulletins.) The depression caused heavy rainfall over parts of China, notably Hainan Province and the Guangxi Zhuang Autonomous Region. Some of the more notable rainfall amounts are listed below. Hainan 0000-0000 UTC 28-29 July 2002 Ledong 85.0 mm Sanya 57.7 mm Xisha Dao 56.3 mm Shanhu Dao 51.0 mm Hainan 0000-0000 UTC 29-30 July 2002 Ledong 96.0 mm Guangxi 0000-0000 UTC 30-31 July 2002 Dongxing 258 mm Pingxiang 95 mm Qinzhou 87 mm Guangdong 0000-0000 UTC 29-30 July 2002 (only amounts greater than 50 mm are listed) Cities Stations Rainfall -------------------------------------------- Shaoguan Shixing 50.1 mm Yunfu Xinxing 59.8 mm Yangjiang Yangjiang 73.5 mm Yangjiang Yangchun 54.4 mm Jieyang Jiexi 68.9 mm Jiangmen Enping 56.6 mm (Report written by Kevin Boyle, based on information sent by Huang Chunliang. A special thanks to Chunliang for sending the report.) ************************************************************************* NORTH INDIAN OCEAN (NIO) - Bay of Bengal and Arabian Sea Activity for July: No tropical cyclones ************************************************************************* SOUTHWEST INDIAN OCEAN (SWI) - South Indian Ocean West of Longitude 90E Activity for July: No tropical cyclones ************************************************************************* NORTHWEST AUSTRALIA/SOUTHEAST INDIAN OCEAN (AUW) - From 90E to 135E Activity for July: No tropical cyclones ************************************************************************* NORTHEAST AUSTRALIA/CORAL SEA (AUE) - From 135E to 160E Activity for July: No tropical cyclones ************************************************************************* SOUTH PACIFIC (SPA) - South Pacific Ocean East of Longitude 160E Activity for July: 1 tropical depression South Pacific Tropical Activity for July ---------------------------------------- As the westerly wind burst which was instrumental in the formation of Typhoon Chataan in the Northwest Pacific bumped into the trades of the South Pacific, a "twin" circulation began to form just south of the equator. By 2100 UTC on 3 July the system had gained enough in deep convection and organization that the Nadi TCWC classified it as a tropical depression, numbered TD-17F. (Normally, 1 July is considered the starting date for the new tropical cyclone "year" in the Southern Hemisphere, so it is a little puzzling why this depression was numbered with the next number from the 2001-2002 season rather than as TD-01F. Perhaps it was because the parent disturbance was taking shape during the final days of June.) At 03/2100 UTC the diffuse center of the depression was located very deep in the tropics at 3S, 164E, or about 450 nm northeast of Guadalcanal. The system remained quasi-stationary, perhaps drifting westward very slowly. The final Nadi bulletin at 05/0600 UTC relocated the center to a point about 325 nm due north of Guadalcanal. According to an e-mail from Jeff Callaghan, before the system was classified as a tropical depression (around 1920 UTC on 2 July), QuikScat data indicated some gales near convection that was well southwest of the center. QuikScat data from 03/2100 UTC showed 20-kt winds in the south semicircle and only 15-kt winds to the north. However, there were some 55-kt rain- contaminated vectors in the south quadrant of the system. TD-17F spent its life west of an upper-level anticyclone in an environment of strong diffluent flow and moderate shear which kept the convection for the most part confined to the southern semicircle. In the cyclone tracks file which I prepared, the 10-min avg MSW was given as 25 kts. This was based on information from Roger Edson, who stated that based upon QuikScat data which he had analyzed, he didn't think winds were much higher than 25 kts on the south (higher pressure gradient) side of the circulation. ************************************************************************* EXTRA FEATURE In order to shorten the amount of typing in preparing the narrative material, I have been in the habit of freely using abbreviations and acronyms. I have tried to define most of these with the first usage in a given summary, but I may have missed one now and then. Most of these are probably understood by a majority of readers but perhaps a few aren't clear to some. To remedy this I developed a Glossary of Abbreviations and Acronyms which I first included in the July, 1998 summary. I don't normally include the Glossary in most months in order to help keep them from being too long. If anyone would like to receive a copy of the Glossary, please e-mail me and I'll be happy to send them a copy. ************************************************************************* AUTHOR'S NOTE: This summary should be considered a very preliminary overview of the tropical cyclones that occur in each month. The cyclone tracks (provided separately) will generally be based upon operational warnings issued by the various tropical cyclone warning centers. The information contained therein may differ somewhat from the tracking and intensity information obtained from a "best-track" file which is based on a detailed post-seasonal analysis of all available data. Information on where to find official "best-track" files from the various warning centers will be passed along from time to time. The track files are not being sent via e-mail. They can be retrieved in the following manner: (a) FTP to: hrd-type42.nhc.noaa.gov [140.90.176.206] (b) Login as: anonymous (c) For a password use your e-mail address (d) Go to "data" subdirectory (Type: cd data) (e) Set file type to ASCII (Type: ascii) (f) Transfer file (Type: get remote_file_name local_file_name ) (The files will be named with an obvious nomenclature--using July as an example: jul02.tracks) (g) To exit FTP, type: quit Both the summaries and the track files are standard text files created in DOS editor. Download to disk and use a viewer such as Notepad or DOS editor to view the files. The first summary in this series covered the month of October, 1997. If anyone wishes to retrieve any of the previous summaries, they may be downloaded from the aforementioned FTP site at HRD. The summary files are catalogued with the nomenclature: jul02.sum, for example. Back issues can also be obtained from the following websites (courtesy of Michael Bath, Michael V. Padua and Michael Pitt): OR Another website where much information about tropical cyclones may be found is the website for the UK Meteorological Office. Their site contains a lot of statistical information about tropical cyclones globally on a monthly basis. The URL is: TROPICAL CYCLONE REPORTS AVAILABLE JTWC now has available on its website the complete Annual Tropical Cyclone Report (ATCR) for 2001 (2000-2001 season for the Southern Hemisphere). ATCRs for earlier years are available also. The URL is: Also, TPC/NHC has available on its webpage nice "technicolor" tracking charts for the 2001 Atlantic and Eastern North Pacific tropical cyclones; also, preliminary storm reports for all the 2001 Atlantic and Eastern North Pacific cyclones are now available, as well as track charts and reports on storms from earlier years. The URL is: A special thanks to Michael Bath of Wollongbar, New South Wales, Australia, for assisting me with proofreading the summaries. PREPARED BY Gary Padgett E-mail: garyp@alaweb.com Phone: 334-222-5327 John Wallace (Northeast Pacific, North Indian Ocean, Western Gulf of Mexico) E-mail: dosidicus@aol.com Kevin Boyle (Eastern Atlantic, Western Northwest Pacific) E-mail: newchapelobservatory@btinternet.com ************************************************************************* ************************************************************************* >> Uploaded: 09-27-02 / Typhoon2000.com / Typhoon2000.ph