Tropical Cyclones
[Index]
Main Index Home Page Stock Weather Photos Australian Severe Weather Forum Storm News and Storm Chasing Reports Tropical Cyclones / Hurricanes / Typhoons Weather Data and Links Wild Fires / Bushfires Weather Observation Techniques Weather Picture Catalogue Tornado Pictures and Reports Stock Video Footage and DVDs for sale
Monthly Global Tropical Cyclone Summary September 1998
[Summaries and Track Data] [Prepared by Gary Padgett]

                 MONTHLY GLOBAL TROPICAL CYCLONE SUMMARY

                             SEPTEMBER, 1998

  (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.)

  ***********************************************************************

                          SEPTEMBER HIGHLIGHTS

  --> Atlantic Ocean and Gulf of Mexico very active--first time this
      century with four simultaneous hurricanes
  --> Long-lived hurricane spreads death and destruction from
      eastern Caribbean Islands to central U. S. Gulf Coast
  --> Mexican west coast experiences a hurricane landfall
  --> Significant increase in Northwestern Pacific activity--three
      tropical cyclones strike Japan; Philippines, Taiwan, and China
      also affected

  ***********************************************************************

                           ACTIVITY BY BASINS

  ATLANTIC (ATL) - North Atlantic Ocean, Caribbean Sea, Gulf of Mexico

  Activity for September:  2 tropical storms
                           6 hurricanes

  NOTE:  Much of the information presented below was obtained from the
  TPC/NHC discussion bulletins issued with every regular advisory.  All
  references to sustained winds imply a 1-min averaging period unless
  otherwise noted.   Also, a special thanks to Eric Blake, a graduate
  student at Colorado State University and owner of the Atlantic Tropical
  Weather Center website, for passing along various bits and pieces of
  information.

                        Hurricane Danielle  (TC #4)
                          24 August - 5 September
                        ---------------------------

     Hurricane Danielle was a much-traveled hurricane which moved from
  its origin in the eastern Atlantic on a west-northwesterly heading to
  recurvature a few hundred miles east of Florida, and thence on a
  general northeasterly course which took it northwest of Bermuda and
  to the south of the Canadian Maritimes.  Danielle was active into the
  first few days of September, but its history was covered in its
  entirety in the Global Tropical Cyclone Summary for August.


                          Hurricane Earl  (TC #5)
                          31 August - 8 September
                          -----------------------

     The precursor of Hurricane Earl was a tropical wave which had spent
  almost two weeks traversing the tropical Atlantic and Caribbean Sea.
  By 31 Aug the associated area of disturbed weather had moved into the
  Bay of Campeche and was showing signs of increased organization.  An
  Air Force Reserve Hurricane Hunter plane investigated the large area
  during the afternoon and found a 1002-mb LOW with peak winds of 43 kts
  at the 450 m flight level.  The system was upgraded to Tropical Storm
  Earl on the first advisory which located the storm's center about
  375 nm south-southeast of Corpus Christi, Texas.   (The prior history 
  of the pre-Earl system was covered in a little more detail in the 
  August summary.)

     Earl initially was a very broad, disorganized tropical storm.
  Pinpointing an exact center was a somewhat difficult task, especially
  considering how the center seemed to "jump" across the Gulf of
  Mexico.  The primary center of Earl seemed to re-form under heavy
  convection  3 or 4 times, making it a very challenging system to
  accurately track and forecast.  On the morning of 1 Sep the center
  was relocated almost 200 nm to the north to a position about 300 nm
  southeast of Corpus Christi.  A reconnaissance flight found peak winds
  of 61 kts with a central pressure of 999 mb.  Late in the afternoon the
  center seemed to have re-formed again to the north and east of the
  1500 UTC advisory position.

     The advisory at 0300 UTC on 2 Sep indicated that Earl's center had
  apparently re-formed yet again nearer to deeper convection, although
  part of the displacement could be attributed to real motion.  That
  advisory also mentioned that Earl was being affected by westerly shear
  due to a shortwave trough to the west intruding into the Gulf and
  that the system didn't really look like a tropical storm in satellite
  imagery.   Data collected by a NOAA research aircraft reinforced the
  idea that Earl was not strictly a tropical cyclone; however, data from
  an Air Force reconnaissance showed that Earl was warm-cored at 450 m.
  Early on 2 Sep satellite fixes from SAB and TAFB were 110 nm apart, 
  and an Air Force reconnaissance reported multiple centers at the 450 m
  flight level.  Maximum winds found were 51 kts and surface MSW was 
  estimated at 45 kts based upon data from buoys and oil rigs.

     By 02/1200 UTC the center of Earl had shifted again to a position
  about 125 nm southwest of Pensacola, Florida.  During the morning a
  reconnaissance plane found 88-kt winds at flight level within a band
  southeast of the center along with a 990-mb minimum pressure.  Earl
  was therefore upgraded to a hurricane at 1500 UTC.   At 2028 UTC
  an Air Force reconnaissance plane found winds of 104 kts in a feeder
  band well east of the center with a central pressure of 986 mb, but
  with no eye or wall cloud forming.   Earl was under strong shear but 
  was able to intensify--another indication that the storm was not 
  operating as a purely tropical system.

     Hurricane Earl continued on a northeasterly course toward the
  Florida Panhandle and made landfall near Panama City around 0600 UTC
  on 3 Sep.   The storm had weakened since the previous afternoon and
  the MSW was estimated at 70 kts at the time of landfall.  Most of the
  hurricane force winds were in the southeast quadrant of the storm
  within strong rainbands.  Cape San Blas reported sustained winds of
  48 kts with gusts to 61 kts about the time of landfall.  (Cape San
  Blas is about 30 nm or so southeast of Panama City.)    A buoy with
  WMO ID of 42039 (28.8 N, 86.0 W) reported sustained winds of 45 kts 
  at 03/0100 UTC with gusts to 58 kts.  The same buoy recorded a peak
  gust of 62 kts between 2000 and 2100 UTC on the 2nd.
  
     After making landfall Earl accelerated northeastward through
  Georgia.  Early on the 3rd a reconnaissance plane found hurricane
  force winds still occuring in a small area over Apalachee Bay, well
  to the south of the center.   Earl was downgraded to a tropical storm
  at 1500 UTC and was declared extratropical by 2100 UTC.  Gusts to
  tropical storm force were reported from waters off the Georgia coast
  during the morning of 3 Sep.  The extratropical storm continued across
  coastal Georgia and the Carolinas, and then accelerated quite rapidly
  across the Atlantic, later passing over the southeastern corner of
  Newfoundland.  (Coordinates for the extratropical portion of Earl's
  track in the companion tracking file were obtained from the High Seas
  Forecasts issued by the Marine Prediction Center.)

     Hurricane Earl proved to be a most difficult tropical cyclone to
  forecast.  The constant shifting of the center made getting a handle
  on the cyclone's motion almost impossible at times, and the somewhat
  hybrid nature of the storm played havoc with many of the computer
  models.  At one point the models were suggesting tracks in just about
  every direction of the compass!

     As would be expected, overall damage from Hurricane Earl was not
  particularly severe.  There was a report of a 3.3 m storm surge along
  portions of the Northwest Florida coast which flooded some homes and
  other structures near the shoreline.   There was one death reported
  in Northwest Florida from a capsized fishing boat.  Several tornadoes
  were spawned by the hurricane in Florida, Georgia, and the Carolinas,
  and one of these was responsible for two deaths in South Carolina.
  In most areas across the region rainfall was not unusually heavy,
  although Panama City recorded 585 mm.   There was some local flooding
  in a few areas and some scattered reports of wind damage--downed trees
  and utility poles and roofs lifted from structures.

  
                       Tropical Storm Frances  (TC #6)
                               8 - 12 September
                       -------------------------------

     Tropical Storm Frances was a rare Atlantic basin example of a
  monsoon depression.   Monsoon depressions are large, sprawling
  low-pressure areas which form in the monsoon trough in many parts
  of the tropics.  To quote from the Annual Tropical Cyclone Report
  published by the JTWC:  "A monsoon depression is a tropical cyclonic
  vortex characterized by: (1) its large size--the outer-most closed
  isobar may have a diameter on the order of 600 nm (1000 km); (2) a
  loosely organized cluster of deep convective elements; (3) a low-
  level wind distribution which features a 100 nm (200 km) diameter
  light wind core which may be partially surrounded by a band of gales;
  and (4) a lack of a distinct cloud system center.  Note: most monsoon
  depressions that form in the Western North Pacific eventually acquire
  persistent central convection and accelerated core winds marking its
  transition into a conventional tropical cyclone."

     According to Dr. Mark Lander of the University of Guam,
  approximately two-thirds of Northwest Pacific basin tropical cyclones
  originate as monsoon depressions.  Also many (probably most) tropical
  cyclones in the Indian Ocean, Australian region, and Southwest Pacific
  begin in this fashion, and occasionally an Eastern North Pacific storm
  originates as a monsoon LOW (e.g., Hurricane Isis).

     The events leading to Frances' formation were in some aspects
  similar to those preceeding Earl's development.  A tropical wave moved
  across the Caribbean and into the southern Gulf of Mexico.  As was the
  case with Earl, computer models had been forecasting tropical cyclone
  development in the area for a few days.   The first reconnaissance
  flight into the area on 8 Sep found a minimum pressure of 999 mb but
  a very broad, poorly-defined center.  Enough deep convection was
  present to justify upgrading the system to a tropical depression.
  The very broad center was placed about 200 nm southeast of Corpus
  Christi, Texas, and was essentially stationary.

     During the morning of 9 Sep a reconnaissance flight was able to
  close off a center but reported an accuracy of 45 nm.  A radar center
  fix from a coastal Doppler radar was about 100 nm to the northwest.
  Peak flight-level winds of 35-39 kts were found about 60-90 nm north-
  east of the center.  However, by later in the afternoon, a buoy (ID
  42019) had reported sustained winds of 35 kts at 3 m, a reconnaissance
  plane had found winds to 40 kts, and several oil rigs well north and
  east of the center had been reporting 40-50 kt winds at 30 m during
  the day; therefore, the depression was upgraded to Tropical Storm
  Frances at 2100 UTC.  High-resolution visible satellite images
  suggested that several smaller vortices were rotating around the
  larger circulation.

     During the evening a plane found a 994-mb center about 100 nm south
  of the previous position, or about 150 nm east-northeast of Tampico,
  Mexico.  This was considered a re-formation of the center and not a
  southward motion.  About this time an oil rig to the northeast of
  Frances reported a gust of 72 kts at an elevation of 30 m, and other
  rigs were reporting gusts to 60 kts.  The MSW was increased to 40 kts
  based on the lower pressure and surface reports.

     On the morning of 10 Sep the center of Frances was relocated back
  to the north near its initial position.  The storm was very slowly
  getting a little better organized as more heavy showers were beginning
  to form nearer the center.  During the day Frances began to exhibit
  a slow northward motion, paralleling the upper Mexican and lower Texas
  coasts.  An oil rig near 28N, 92W reported sustained winds to 68 kts
  with gusts to 77 kts at an elevation of a little more than 30 m.  The
  MSW was increased to a peak of 55 kts at 11/0300 UTC when Frances was
  centered about 100 nm east-southeast of Corpus Christi.  The higher
  winds to the north and east of the center were likely a result of
  a tight pressure gradient between Frances and a fairly strong HIGH to
  the north.  With an impressive upper-level anticyclone over the storm
  and low vertical shear, Frances was looking more like a typical 
  tropical storm at this time in satellite imagery.   A reconnaissance 
  flight during the evening found a pressure of 995 mb and flight-level 
  winds of 59 kts.
  
     Late on 10 Sep Tropical Storm Frances began to move in a more 
  northwesterly direction and made landfall around 0600 UTC on 11 Sep
  between Port Aransas and Port O'Connor, Texas.  The storm caused rather
  extensive flooding from the storm surge and heavy rainfall along the
  middle and upper Texas coast.  There were reports of tides up to 2 m
  above normal, resulting in flooding of coastal communities and
  highways.  Victoria, Texas, reported sustained winds to tropical storm
  force with higher gusts as Frances made landfall; and Sabine, Texas,
  near the Louisiana border, reported a sustained wind of 41 kts with
  a peak gust of 48 kts at 1000 UTC on the 11th.

     After moving inland the center of Frances stalled near the coast.
  As a result winds were slow to diminish and buoys, oil rigs, and ships
  continued to report winds of gale-force for another 12-15 hours.   By 
  midday on the 11th Frances appeared to be located near the end of a
  large-scale and well-defined frontal zone.   Frances was downgraded to
  a depression at 12/0300 UTC.  The remnants began to track northeastward
  along the front and eventually reached the Great Lakes region.

     Heavy rains fell in parts of Texas, but unfortunately I don't have
  specific amounts available to me at this time.  However, there is one
  area which received a very large quantity of rainfall from Frances that
  I do have some amounts for--New Orleans, Louisiana.   Downtown New
  Orleans had a storm total of 543 mm, with 277 mm falling on the 11th
  and 165 mm within two hours.    Metairie, just west of New Orleans,
  received a total of 467 mm and Terrytown, a community on the west bank
  of the Mississippi River, recorded 536 mm.  (A special thanks to Eric
  Blake for passing along those figures.)


                 Tropical Disturbance of 13-14 September
                 ---------------------------------------

     It is not my normal intent to discuss unclassified tropical
  disturbances, but a system in the eastern tropical Atlantic on 13 and
  14 Sep needs a little attention.   A tropical wave which had moved off
  the African coast developed a quite well-organized appearance in
  infrared satellite imagery on the evening of the 13th.  The convective
  cluster was near 20N, 30W and was moving northward--an unusual motion
  for a system in that location at that time of year.   Convection had
  largely died down by the morning of 14 Sep, but reappeared to some
  degree later in the day.   The weak LOW, though, was moving at an 
  increased pace northward into an area unfavorable for tropical 
  development and soon fell completely apart.

     However, when the convective signature was at its most organized
  stage, there were quite a few people who were accusing TPC/NHC of
  ignoring an "unnamed tropical storm" because it was moving north
  into a no-man's land.     Scott Hegmann, a young fellow in Queens,
  New York City, who is the owner of a nice tropical weather webpage
  (http://pages.prodigy.com/weatherlink/), wrote TPC/NHC asking about
  the system.  Dr. Richard Pasch, who was the Hurricane Specialist on 
  duty that night, was kind enough to reply to Scott, and Scott later
  forwarded the information to me.  (A special thanks to Scott for
  sending along that piece of information.)

     A ship passed essentially through the center of the convective
  cluster and the pressure dropped to 1007.9 mb with a southeast wind
  of 15 kts and very heavy rain.  Satellite Dvorak classifications from
  TAFB and SAB were T1.5 and T1.0, respectively.  The satellite signature
  was fairly impressive but there was no evidence of a surface
  circulation.   


                        Hurricane Georges  (TC #7)
                             15 - 29 September
                        --------------------------

     The year's most significant hurricane developed from a well-
  organized tropical wave which had moved off the west African coast
  around 13 Sep.  On 15 Sep a ship just north of the center reported
  30-kt easterly winds, and another ship southwest of the center reported
  northwest winds of 20 kts.     Based on these reports and satellite
  imagery, the first depression advisory was issued at 1500 UTC locating
  the center about 350 nm south of the Cape Verde Islands.     The
  depression moved westward at a fairly quick pace and became Tropical
  Storm Georges at 1200 UTC on the 16th when it was located roughly
  500 nm west-southwest of the Cape Verdes.  The cloud pattern continued
  to improve throughout the day as it changed from a CDO to a banding
  pattern.

     Georges continued to become better organized and slowly strengthened
  as it moved rather quickly on a course that was slightly north of due
  west, guided by a persistent ridge to the north.  Hurricane intensity
  was reached around 17/1800 UTC when the storm was located roughly
  halfway between the Lesser Antilles and Cape Verde Islands.   The
  hurricane at this point had all the classical features needed to become
  a major hurricane:  good upper-level outflow, inflow from the ridge to
  the north, warm SSTs and increasing along the track, and no westerlies
  ahead.

     The hurricane continued to increase in intensity during the 18th and
  19th with MSW estimated to have reached 110 kts by 19/1500 UTC.  When
  the first Air Force reconnaissance flight reached Georges on the
  afternoon of 19 Sep, it found peak flight-level winds of 146 kts and a
  pressure of 938 mb.  Surface MSW was set at 125 kts in the 2100 UTC
  advisory.     During the evening a GPS dropwindsonde released by a
  NOAA research aircraft reported a wind of 166 kts at low levels in the
  eyewall.   Based upon this the MSW was increased to 130 kts--the peak
  intensity for the hurricane's life.   A flight at 20/0613 UTC measured
  a pressure of 937 mb.  This was the minimum central pressure reported
  during Hurricane Georges' history.

     Georges weakened significantly on 20 Sep as the central pressure
  rose to 966 mb and the MSW dropped to 100 kts by 21/0000 UTC, based on
  a peak flight-level wind of 114 kts.  This weakening appeared to be
  the result of the hurricane moving from under one upper ridge (to 
  its east) to another (to its west).  The northwesterly winds on the 
  eastern side of the new ridge induced some shearing over the hurricane
  as it approached the Leeward Islands.   Maximum winds were further
  reduced to 95 kts as Georges reached the islands very early on the
  morning of 21 Sep.

     As best I can determine, the eye of Hurricane Georges passed very
  near or over Antigua; very near Nevis, St. Kitts, St. Eustatius, and
  Saba; just north of St. Croix; south of St. Thomas; very near Vieques
  Island; and thence into the eastern end of Puerto Rico sometime around
  2100 UTC on 21 Sep.  The Meteorological Office in Guadeloupe reported
  sustained winds of 57 kts with gusts to 76 kts in their 21/0100 UTC
  observation as Georges passed to the north.  There was an unofficial
  report from Antigua of sustained winds of 90 kts with a peak gust of
  101 kts, and also unofficial reports of some wind gusts to near 150 kts
  on the island of Saba at an elevation of 230 m with significant
  structural damage.   Georges' center passed about 35 to 45 nm south of
  St. Martin where gusts above 80 kts were reported--there were also
  reports of gusts to 80 kts on St. Croix after the eye had passed.

     As Hurricane Georges was approaching Puerto Rico the westerly shear
  lessened and the storm rapidly strengthened.     The Civil Defense
  office in Fajardo, Puerto Rico, reported a sustained wind of 95 kts
  with gusts to 113 kts as Georges' eye made landfall.     Based upon
  this, along with WSR-88D radar data, the MSW was increased to 100 kts,
  making Georges once more a Category 3 hurricane on the Saffir/Simpson
  scale.  Around 0300 UTC on the 22nd radar indicated winds of 100 kts
  at 915 m (the elevation of the radar site).     Satellite imagery
  indicated that cloud tops of -80 C were over eastern Puerto Rico.

     The center of Georges more or less bisected the island from east to
  west, exiting the southwestern end of Puerto Rico shortly after 0600
  UTC on 22 Sep.  Georges did not weaken much while over Puerto Rico and
  immediately began to strengthen after moving into the Mona Passage.
  Shortly before moving inland over the Dominican Republic, a
  reconnaissance plane found flight-level winds of 117 kts and a central
  pressure of 962 mb; therefore, the MSW was increased to 105 kts.  The 
  eye of Hurricane Georges passed very near Santo Domingo around 1800 
  UTC, and then directly across the island of Hispaniola.   By the time
  the storm had emerged into the Windward Passage off the west coast of
  Haiti, it was a much weakened system with winds barely of hurricane
  force.   As expected, the mountain peaks in excess of 3000 m on the 
  island had taken their toll on Georges.

     The hurricane did not have much time to reorganize before making
  landfall in eastern Cuba.  Georges passed near Guantanamo, Cuba, around
  24/0000 UTC.  During the morning of the 24th the center was emerging
  into the Atlantic along the north coast of eastern Cuba and an eyewall
  appeared to be forming.  A plane found a pressure of 989 mb so the MSW
  was increased to 70 kts in the 1500 UTC advisory.  The eye of Georges
  continued on a west-northwestward heading, slowly pulling away from
  the Cuban coast and arriving in the vicinity of the Florida Keys early
  on 25 Sep.

     The center of Georges was about 50 nm southeast of Key West at
  1200 UTC and about 25 nm west or west-northwest of the city six hours
  later, so it appears that the center passed over or just to the
  south of Key West during the morning.   The MSW were estimated to be
  around 85-90 kts during this time.  The automated reporting site on
  Sombrero Key, located about 30 nm east of Key West, reported sustained
  hurricane force winds for over three hours, including a 2-minute wind
  of 82 kts with a gust to 98 kts.  Winds well above tropical storm force
  were felt over the southern portion of the Florida Peninsula.  During
  the afternoon of the 25th, Marco Island, on the southwest side of the
  Peninsula, measured a gust of 83 kts.  Earlier, the same location had
  reported sustained winds of 50 kts.  After passing Key West the eye of
  Georges passed over or very near Dry Tortugas island where a pressure
  of 976.3 mb was measured.

     After leaving the Florida Keys Hurricane Georges gradually turned
  to a more northwesterly course toward the central U. S. Gulf Coast.
  Central pressure very gradually dropped and the MSW increased to 95 kts
  by 26/1800 UTC when the storm was centered about 315 nm southeast of
  New Orleans.   Although Georges was moving over the very warm waters
  of the Gulf of Mexico, it was unable to strengthen into a Category 3
  hurricane again.   Some of the inhibitive factors likely were: some
  light westerly shear, disruption of the inner core by the mountains
  of the Greater Antilles, and some drier and stable air from the
  northern Gulf being pulled into the circulation.  The effects of the
  shear were seen on 26 Sep when a reconnaissance flight could find no
  hurricane force winds in the western semi-circle.

     The discussion issued with the 26/2100 UTC advisory illustrates the 
  uncertainty in determining the maximum wind speed.  Dvorak T-numbers
  had reached 5.5--implying 102 kts.  Hurricane hunters found a central
  pressure of 968 mb and flight-level winds of 92 kts.  The latest wind
  field analysis from HRD showed around 70 kts, but a GPS dropwindsonde
  measured 95 kts at the surface.    In this case the GPS data was
  considered a reliable measurement and the MSW was increased to 95 kts.

     Georges began to curve more to the north on 27 Sep and passed about
  50 nm east of the mouth of the Mississippi River around 0000 UTC on
  the 28th.  A reconnaissance flight at 27/1935 UTC reported a closed
  eyewall and a pressure of 961 mb along with 85-kt flight-level winds.
  However, a GPS dropwindsonde recorded a 100-kt wind at the 952 mb
  level, which agreed with Doppler winds from the New Orleans radar.
  Georges' MSW was decreased slightly to 90 kts in the 28/0300 UTC
  advisory.
  
     The eye of Hurricane Georges reached the Mississippi coast very
  near Gulfport around daybreak on 28 Sep and then essentially
  stalled.   A peak gust of 110 kts was recorded at Biloxi.  This agrees
  well with a 90-kt MSW, taking into account the accepted value for the
  overwater 1-min MSW to peak gust ratio.  To the east Dauphin Island,
  Alabama, recorded a peak gust of 71 kts while to the west, New Orleans
  International Airport reported a gust to 48 kts.   Once inland Georges
  began to steadily weaken.  The center drifted very slowly northward,
  meandered around a little, and then began a slow drift to the east.
  Georges was downgraded to a tropical storm at 2100 UTC on 28 Sep.
  Reports of sustained 39-kt winds from Hurlburt Field in Northwest
  Florida and 34 kts from Dauphin Island late on the 28th and very early
  on the 29th were the basis for keeping Georges at tropical storm 
  intensity through the morning of 29 Sep.  The final TPC/NHC advisory 
  downgraded Georges to a depression at 29/1500 UTC about 20 nm north
  of Mobile, Alabama.   The remnants of Georges moved eastward across
  extreme southern Alabama and Georgia and eventually into the Atlantic,
  but soon thereafter merged with a frontal zone.

     Hurricane Georges was a very destructive and deadly tropical cyclone
  in many of the areas it visited during its long life.  I will give a
  brief synopsis of some of the information I've gleaned from various
  press reports and websites, arranged by islands/areas affected.
  (Some of the information from Puerto Rico comes from an e-mail posted
  by Dr. Matthew C. Larsen, a hydrologist with the U. S. Geological
  Survey at Guaynabo, Puerto Rico; and some information given from Cuba
  is taken from an e-mail posted by Dr. Jose Rubiera of the Instituto
  de Meteorologia of Cuba.)

  Antigua:  2 deaths; 3800 homeless (Antigua and Barbuda)

  St. Kitts/Nevis:  5 deaths; 3000 homeless; 85% of homes damaged;
  damage estimated at $402 million

  U. S. Virgin Islands:  widespread damage to homes and crops; 55 boats
  sunk

  Puerto Rico:  5 deaths; damage estimated $2 billion; most all residents
  lost power; central mountains received 250-400 mm of rain with Jayuya 
  area receiving 500-635 mm; maximum storm surge of 1.5-2.0 m on east 
  coast near Ceiba; in Rio Saliente near Jayuya the average discharge
  of water during flooding was calculated to have been 333 cubic meters
  per second; almost all flowering plants were defoliated--as a result,
  bees invaded homes and businesses all over the island looking for food.

  Dominican Republic:  283 deaths, 64 missing; 100,000 homeless; 90% of
  banana and other plantations destroyed; severe flooding in many parts 
  of the nation; damage estimated at $1.2 billion.
                       
  Haiti:  87 deaths, possibly many more will be discovered later; over
  500 mm of rain in Artibonite Valley; agricultural losses severe--90%  
  or more of many crops destroyed.

  Cuba:  6 deaths indirectly related to storm; rainfall amounts in excess
  of 200 mm common over eastern Cuba--Santiago de Cuba recorded 471 mm in
  24 hours; Limonar recorded 620 mm while Georges was intensifying in the
  Florida Straits; almost 3500 homes destroyed;  agricultural losses 
  heavy, especially to banana plantations; unofficial estimate of insured
  losses placed at $15 million U. S. dollars; one beneficial effect--
  rainfall helped to replenish water accumulations in reservoirs depleted
  by severe drought conditions which had prevailed during much of the
  preceeding year.
         
  Florida Keys:  There was fairly severe wind and water damage reported
  in the lower Keys, but at the time I do not have any specific figures 
  available.

  U. S. Gulf Coast:  There was widespread scattered wind-related damage
  reported, but the major impact of Georges in this region was flooding
  due to storm surge and incredibly heavy rainfall.  Onshore southerly
  winds pushed the waters of Mobile Bay into downtown Mobile, Alabama.
  At one point water was reported to be up to 1.5 m deep in some of the
  main streets.  The heaviest rainfall amounts were reported over North-
  west Florida and south-central Alabama.  Amounts in excess of 500 mm
  were common.  An observer near the small city of Opp, Alabama, recorded
  750 mm for a storm total.       Many of the streams in the region
  experienced some of the worst flooding seen in many, many years.
  Fortunately,  no major cities were flooded  (except for the
  aforementioned tidal flooding in Mobile).  Many agricultural crops in
  the area, including cotton, soybean, peanut, and pecan, were heavily
  damaged.

     Near the end of the official Atlantic Hurricane Season (30 Nov)
  there are normally several articles in the press, as well as one
  prepared by TPC/NHC, summarizing the tropical cyclones of the year.
  Hopefully they will contain some more complete figures regarding
  casualties and damage estimates.  If so, I will pass this information
  along in a future summary.


                     Tropical Storm Hermine  (TC #8)
                           17 - 20 September
                     -------------------------------

     A tropical wave moving through the Caribbean Sea during the second
  week of September began to show signs of getting better organized by
  the morning of 15 Sep.  An area of low pressure developed over the
  Northwest Caribbean with numerous showers and squally weather affecting
  adjacent coastal areas.  For the third week in a row computer models
  had been forecasting some sort of tropical development in this region
  (the other cases being Earl and Frances).  The broad LOW moved up into
  the eastern Gulf of Mexico on 16 Sep where it was investigated by a
  reconnaissance plane, but no well-defined circulation was found.
  Development was somewhat impeded by an upper-level LOW over the Gulf
  to the west.

     On the 17th, however, a reconnaissance flight found a very small
  center of circulation with 30 to 35-kt winds close to the center and
  a pressure of 1001 mb; therefore, advisories were initiated on the
  system at 2100 UTC.  The depression center was located south of the
  central Louisiana coast about 200 nm south-southwest of New Orleans.
  The system was undergoing shear at the time and the satellite cloud
  signature did not look like a tropical cyclone at all, although there
  were some clusters of deep convection to the south of the center.
  During the 17th and 18th the depression drifted to the southwest a
  little and then just meandered around in the same area.  Environmental
  conditions were still hostile for further development, and planes
  investigating the system found flight-level winds at 450 m to be
  around 37-38 kts--not quite enough to justify upgrading the depression
  to a tropical storm, although a pressure of 999 mb was found on the
  morning of 18 Sep.  The depression was still in the grip of the large
  upper LOW located to its southeast and apparently was slowly orbiting
  around the periphery of that system.

     Very early on 19 Sep the depression was still quite severely sheared
  and it was beginning to look doubtful if it would ever strengthen at
  all.   A reconnaissance flight found a very weak vortex with only light
  winds near the center.  However, there were bursts of deep convection
  about 120 nm northeast of the center, and a plane found winds to 40 kts
  near this convection.  A new center appeared to be forming under this
  convection, and a reconnaissance plane later in the morning found winds
  to 56 kts east of the new center.   Ship XCMG reported winds of 33 kts
  to the north of the LOW center, and a platform (S58) reported sustained
  winds of 40 kts at an elevation of 36 m; hence, the depression was
  upgraded to Tropical Storm Hermine at 1500 UTC with 40-kt MSW.  The
  center of Hermine was then located about 160 nm south-southwest of
  New Orleans.

     The tropical storm moved north toward the Louisiana coast and made
  landfall around 0000 UTC on 20 Sep in Terrebonne Parish about 70 nm
  southwest of New Orleans.  After its initial act of intensification,
  Hermine did not strengthen any further.  As the system made landfall,
  aircraft and Doppler radar showed only a few patches of winds to
  tropical storm force, mainly east and northeast of the center. Hermine
  seemed to stall at the coast for a few hours, then commenced a slow
  northward and later northeastward drift as it weakened into a
  depression.

     Hermine brought some locally heavy rains to some spots, but nothing
  like the deluge of a week earlier associated with Frances.  Some areas
  of Mississippi and Alabama experienced some rather heavy rainfall over
  the next couple of days as the remnants of Hermine drifted slowly
  across those states (the author's home in southern Alabama being one
  of those places--109 mm on 20-21 Sep).


                         Hurricane Ivan  (TC #9)
                            20 - 28 September
                         -----------------------

     A tropical wave which left the African coast around the middle of
  September had developed sufficiently that depression advisories were
  begun at 1500 UTC on 20 Sep, locating the tropical depression's center
  about 500 nm west of the Cape Verde Islands.   The system began to
  move on a northwesterly heading and was upgraded to Tropical Storm
  Ivan twelve hours later.   Ivan intensified very slowly due to the
  shearing effects of a trough located to the west, which had also
  induced the more northerly motion.  Ivan moved due northward early on
  22 Sep, and then, as the upper LOW to the west cut off from the
  westerlies, returned to a north-northwesterly heading.  The effects
  of the shear had diminished by this time and Ivan's winds reached
  55 kts, based on satellite imagery.

     By late on the 22nd, however, deep convection waned and Ivan
  appeared to weaken some.  The TPC/NHC discussion at 23/0300 UTC also
  noted that Ivan exhibited some subtropical-like features.  This
  weakening trend was reversed the next day (23 Sep) as deep convection
  returned, and an eye was first noted around 1400 UTC.  This feature
  persisted and Ivan was upgraded to a 65-kt hurricane at 2100 UTC when
  centered about 800 nm southwest of the Azores.   Ivan slowly turned
  to a more northerly course, and then began to accelerate to the east-
  northeast on 25 Sep as it got caught up in westerly flow.  Estimated
  peak intensity of 80 kts was reached around 26/0000 UTC and was
  maintained for about twelve hours.   Ivan displayed a distinct eye
  that was about 18-20 nm in diameter.  Convective tops were not very
  cold, but that is not unusual for high-latitude hurricanes.

     Ivan began to weaken and slowly lost its tropical characteristics
  after 1200 UTC on 26 Sep.    Shortly before 1800 UTC the hurricane
  passed about 75 nm north of Flores Island, the northwesternmost
  island in the Azores group.   Ivan was downgraded to an extratropical
  storm of less than hurricane force at 27/0300 UTC, and the remnants
  continued to speed eastward, being located off the coast of Portugal
  by 28/0000 UTC.   Since Ivan passed only about 75 nm north of the
  island of Flores and 34-kt winds were estimated to extend outward to
  120 nm south of the center, it is likely that tropical storm force
  winds were experienced on the island.


                       Hurricane Jeanne  (TC #10)
                        21 September - 1 October
                       --------------------------

     An exceptionally well-organized tropical wave moved off the west
  African coast around 19 Sep and moved very slowly westward for a couple
  of days.  By the afternoon of 21 Sep banding features and good outflow
  had become prominent, and Dvorak T-numbers and an observation from a
  ship northeast of the center implied 35-kt winds, so advisories were
  initiated on Tropical Storm Jeanne at 2100 UTC.   The initial position
  of Jeanne was about 350 nm southeast of the Cape Verde Islands.  The
  TPC/NHC discussion mentioned that Jeanne was only the third tropical
  storm on record to develop so far east.

     The two easternmost developing tropical storms in the Atlantic I 
  could find were Tropical Storm Ginger of October, 1967 (18.1 W) and
  an unnamed tropical storm in September, 1988 (18.5 W).  However, there
  were two hurricanes (in 1947 and 1948) which are listed in the Best 
  Track file as having winds of 45 and 50 kts by the time they crossed 
  the 20th meridian.

     A high pressure ridge to the north of Jeanne, warm SSTs, and low
  vertical shear helped to provide a favorable environment for further
  intensification, and Jeanne was upgraded to a hurricane at 22/2100
  UTC when located about 125 nm southwest of the southernmost Cape
  Verdes.   Hurricane Jeanne continued on a steady west-northwesterly
  track for several days and gradually intensified to its peak estimated
  MSW of 90 kts.   By the 25th Jeanne was beginning to feel the effects
  of vertical shear from a trough to the west.   The storm gradually
  weakened as it turned to a more northerly course.      MSW were
  estimated at 65-70 kts from early on the 26th until the morning of
  28 Sep.

     Visible satellite pictures on 28 Sep revealed that Jeanne had
  become stronger, and winds were increased to 75 kts.   By 2100 UTC
  Jeanne's center was still imbedded in deep convection with cloud
  tops of -75 C, and Dvorak estimates from TAFB and SAB were T5.0 and
  T4.5, respectively; therefore, the MSW was increased to a secondary
  maximum of 80 kts.  Jeanne was by this time moving to the northeast,
  and it was felt at one point that Jeanne might accelerate toward the
  Azores and move through the island group as a hurricane.  However,
  by the 29th Jeanne was beginning to weaken rapidly as increased
  westerly shear and cooler SSTs began to take their toll; and the storm,
  while moving by now on an east-northeasterly heading straight toward
  the Azores, did not accelerate, and had weakened into a depression by
  the time the center moved through the central Azores late on 30 Sep.

     Early in Jeanne's history, as it passed south of the Cape Verde
  Islands, it is possible that tropical storm force winds were felt on
  the southernmost islands in this group.   Very few Atlantic tropical
  cyclones bring tropical storm conditions to these islands, and fewer
  still bring hurricane force winds.   The last storm that could have
  possibly caused tropical storm force winds in the Cape Verde group
  was Tropical Storm Felix in August, 1989.


                       Hurricane Karl  (TC #11)
                           23 - 29 September
                       ------------------------

     The origin of Hurricane Karl is perhaps a little bit of a mystery.
  The author remembers noting a small convective cloud pattern on the
  morning of 22 Sep about halfway between Cape Hatteras and Bermuda that
  was showing increased organization.  At 1200 UTC on 23 Sep, TPC/NHC
  issued a Special Tropical Disturbance Statement which reported that
  a well-defined low pressure system was moving eastward just to the
  north of Bermuda, and that Bermuda had experienced wind gusts to 34
  kts.   The statement indicated that the system seemed to be acquiring
  tropical characteristics and that if this trend continued, advisories
  would be initiated.  (More on the origin of Karl following the main
  discussion of the storm's history.)

     The first depression advisory was issued at 23/2100 UTC, locating
  the center about 115 nm east-northeast of Bermuda.   Tropical Storm
  Karl was christened six hours later.   Karl was initially forecast to
  move rather quickly eastward ahead of an approaching frontal trough,
  but the storm begin to exhibit an east-southeastward motion and was
  not picked up by the trough.  Karl gradually strengthened and was
  upgraded to the season's seventh hurricane at 1500 UTC on 25 Sep when
  it was centered about 575 nm east-southeast of Bermuda.  The NOGAPS
  model was indicating a later possible Fujiwhara interaction with
  Hurricane Jeanne, but this failed to materialize.   About the time
  Karl reached hurricane intensity, the southeastward motion came to
  a halt and the storm began to move slowly eastward, then turned more
  to the east-northeast and began to accelerate ahead of the next 
  approaching trough.

     Hurricane Karl's MSW reached an estimated peak of 90 kts at 27/0000
  UTC (when Dvorak estimates from TAFB and SAB were both T5.0) and
  maintained this intensity for twelve hours.  Karl was already involved
  with a frontal zone at this time and the storm soon began to weaken
  quickly as it accelerated rapidly across the cooler North Atlantic
  waters.  The last TPC/NHC advisory, issued at 28/0300 UTC, downgraded
  Karl to a tropical storm which was rapidly becoming extratropical.
  The storm was at that time located about 150 nm west-northwest of the
  island of Flores in the Azores.  Twenty-four hours later the extra-
  tropical remnants were a few hundred miles southwest of Ireland,
  moving northeastward.  Based on the gale-force wind radii from the
  last advisory, it is doubtful if any of the Azores were affected by
  strong winds from Karl.

     When Karl was upgraded to a hurricane on 25 Sep, Hurricanes Jeanne
  and Ivan were concurrently operating in the Atlantic, and Hurricane
  Georges was in the Florida Straits battering the lower Keys.   This
  represents the first known instance during the 20th century that four
  hurricanes have co-existed in the Atlantic basin.  Four simultaneous
  named storms have occurred on a few more occasions (the last such
  occurrence being in late Aug/early Sep, 1995), but the last time the
  Best Track database indicates that four hurricanes existed together 
  was in 1893.    However, given that there is a lot of uncertainty 
  regarding the intensity of tropical cyclones before the advent of 
  aerial reconnaissance and satellite imagery, it is certainly possible
  that such an event has happened more frequently.

     Was Hurricane Karl a reincarnation of Tropical Storm Hermine?  The
  first TPC/NHC discussion mentioned this as a possibility.    Scott
  Bachmeier of the University of Wisconsin posted an analysis based on
  a study of satellite imagery during the period 17-22 Sep which was
  very interesting.   A small, low-level vortex had been seen off the
  Carolina coasts on 17-18 Sep.  This vortex was not evident on 19 Sep,
  but a vortex could be seen off the coast of the Carolinas on 20 Sep
  with some convection beginning to develop around the LOW.

     However, at 1500 UTC on 20 Sep the weakening Hermine was about 75
  miles north of New Orleans, moving north-northeast at 13 kts.  By
  21 Sep Hermine's remnants were moving slowly across northern Alabama.
  It seems very unlikely that the remnants of Hermine could have reached
  the location of the offshore vortex seen on that date, and since the
  20 Sep vortex was likely the precursor of Karl, it follows that Karl
  was not a re-development of the earlier tropical storm.  This was the
  conclusion reached by Scott,  and based on what I observed, I am
  inclined to agree.      It is unclear whether or not there was a
  connection with the small vortex seen during the previous week.

  ***********************************************************************

  NORTHEAST PACIFIC (NEP) - North Pacific Ocean East of Longitude 180

  Activity for September:  1 tropical storm
                           1 hurricane

  NOTE:  Much of the information presented below was obtained from the
  TPC/NHC discussion bulletins issued with every regular advisory (CPHC
  for locations west of 140W.)  All references to sustained winds imply
  a 1-min averaging period unless otherwise noted.


                        Hurricane Isis  (TC-10E)
                            1 - 3 September
                        ------------------------

     A tropical disturbance off the southwest coast of Mexico in late
  August had some monsoon depression-like features.  On 1 Sep the system
  appeared to be getting organized, and some ship reports of 40-kt winds
  and a pressure of 997 mb led to the system being upgraded directly to
  Tropical Storm Isis at 2100 UTC.   Isis' center was estimated to be
  located about 300 nm west of Manzanillo.  On the next advisory (02/0300
  UTC) the center was relocated about 120 nm farther to the north based
  on a ship report of Beaufort force 10 winds (48-55 kts) and a pressure
  of 993 mb.  The MSW was increased to 60 kts based on this report.  By
  0900 UTC on the 2nd a very large symmetrical area of deep convection
  had developed (tops around -80 C), and a prominent banding feature
  was evident along the coast south and east of the center extending to
  near Puerta Vallarta.

     By the afternoon of 2 Sep radar observations from Guasave, Mexico,
  and satellite imagery revealed an eye forming over the southern Gulf
  of California.   Isis was upgraded to a hurricane with 65-kt MSW at
  2100 UTC.   The storm was centered at this time about 125 nm west-
  northwest of Mazatlan, after having earlier passed only about 30 nm 
  east of Cabo San Lucas on the tip of the Baja California peninsula.

     Some of the computer models forecast Isis to march on a north-
  westerly heading right up the Gulf of California.  Such a trajectory
  would be very unusual but not unprecedented.    In 1967 Hurricane
  Katrina moved up the entire length of the Gulf and inland at the
  very northern extremety of the Gulf.  Katrina caused severe damage
  in some localities and brought tropical storm force winds to Yuma,
  Arizona.

     Isis, however, accelerated on an almost due northerly heading and
  made landfall near Los Mochis in northwestern Sinaloa state shortly
  after 0000 UTC on 3 Sep.  Isis continued northward over the Sierra
  Madre Occidental range and steadily weakened.  The last advisory, an
  Intermediate Advisory at 03/1800 UTC,  placed the dissipating
  circulation over the mountains along the border between Sonora and
  Chihuahua states in northwestern Mexico.

     After Isis had moved inland there was a burst of very deep
  convection with tops to -75 C.  This led to some very heavy rainfall
  which caused widespread flooding in Sinaloa and Sonora states. Ten
  deaths were reported, and over 300 homes were destroyed in the Los
  Mochis area.  At Empalme (near Guaymas in Sonora state) about 200 mm
  of rainfall was recorded.  Normally arid San Bernardino, near the
  U. S. border, reported 56 mm of rain.


                     Tropical Storm Javier  (TC-11E)
                             7 - 12 September
                     -------------------------------

     Tropical Storm Javier developed in and remained in an environment
  of easterly shear throughout its life.   The first depression advisory
  by TPC/NHC at 1500 UTC on 7 Sep located the system's center about 300
  nm west of Manzanillo.    The center was relocated a little to the
  north and west, then began to move on a slow westerly course.  By
  08/0000 UTC the depression was centered about 175 nm southwest of Cabo
  San Lucas.   Both TAFB and SAB were reporting Dvorak T-numbers of
  2.5 and Socorro Island, located more than 100 nm to the south, was
  reporting a west wind of 28 kts and a pressure of 1000.6; therefore,
  the depression was upgraded to Tropical Storm Javier at 0300 UTC.

     Javier continued on a slow westerly track and reached its peak
  intensity of 45 kts at 08/1200 UTC.  At this point the shear had
  relaxed enough to allow the low-level center to get underneath some of
  the deep convection, and the outflow also had improved.  However,
  shortly thereafter the shear increased again and the center became
  exposed, so the MSW was decreased to 40 kts.  Javier also reversed its
  heading and began to move slowly back eastward.  By 1800 UTC on 9 Sep
  the weakening storm was downgraded to a depression about 150 nm
  southwest of Cabo San Lucas at 2100 UTC.

     Over the next two days the depression, which was only a swirl of
  low clouds with some occasional sporadic convection, drifted slowly
  east-southeasterly back to near its point of origin.     The final
  advisory at 11/2100 UTC placed the weak center about 275 nm west of
  Manzanillo.   However, shortly afterward some convection began to
  regenerate, and ship reports indicated that Javier briefly regained
  tropical storm strength to 45 kts on 12 Sep with the center very near
  Manzanillo.   This eleventh-hour resurgence was very short-lived and
  Javier dissipated quickly later on the 12th.

     The information about Javier's brief re-intensification on 12 Sep
  was taken from the monthly tropical activity report prepared by the
  staff of TPC/NHC.   To my knowledge no advisories were written during
  this phase of the storm; consequently, this portion of the storm's 
  history is not reflected in the accompanying tracking file.

  ***********************************************************************

  NORTHWEST PACIFIC (NWP) - North Pacific Ocean West of Longitude 180

  Activity for September:  3 tropical depressions
                           1 tropical storm
                           5 typhoons **

  ** - This based upon JTWC's classification--JMA did not classify
       Stella as a typhoon.

  NOTE: Most of the information on each cyclone's history presented in 
  the narrative will be based upon JTWC's advisories, and references to
  winds should be understood as a 1-min avg MSW unless otherwise noted.
  However, in the accompanying tracking document I have made comparisons
  of coordinates with JMA (Japan) and the Philippines (PAGASA) when their
  positions differed from JTWC's by usually 40-50 nm or more.  Also, as
  announced earlier in a separate posting, a column of 10-min avg MSW
  is included--the values being obtained from either PAGASA's or JMA's
  advisories.  A special thanks to Michael V. Padua, owner of the Typhoon
  '98 webpage, for sending me the PAGASA tracks.  Also some information,
  primarily on the pre-depression stages of the various cyclones, was
  taken from the Monthly Report of the RSMC, Darwin, Northern Territory,
  Australia.  A special thanks to Sam Cleland for forwarding that report
  to me.

     In the title line for each storm I plan to reference all the cyclone
  names/numbers I have available:  JTWC's depression number and name (if
  any), JMA's tropical storm numeric designator, and PAGASA's name for
  systems forming in or passing through their area of responsibility.

     All of the information regarding damage and casualties--with two
  exceptions (which will be noted)--were taken from press releases
  located on various websites.


                  Typhoon Rex/Deling  (TC-06W / TY 9804)
                          24 August - 7 September
                  --------------------------------------

     Typhoon Rex formed in the Philippine Sea in late August and moved
  on a general but somewhat erratic northeastward course during the last
  week of that month.  The August portion of Rex's history was covered
  in the summary for August, but as promised, I'll try to fill in a few
  more details on the atmospheric circulation features which helped to
  steer Typhoon Rex.  Initially Rex was steered to the north-northeast
  by a ridge to the southeast.  On 26 Aug Rex had moved to the north
  side of the ridge and turned more to the east.  Also, an upper-level
  LOW helped to change the orientation of the ridge to the south of Rex
  and as a result, the by-now typhoon turned to the east-southeast.

     By 27 Aug a ridge to the southeast of Rex influenced the storm to
  begin moving more to the east-northeast.  Later on the 27th Rex turned
  more to the northeast and then north due to the steering influence of
  a ridge to the northeast.  The subtropical ridge to Rex's east steered
  the typhoon generally to the north until 30 Aug when its forward motion
  slowed down as it tried to break through a 200-mb ridge axis to its
  north.   Typhoon Rex was still moving east-northeastward on 31 Aug but
  was coming under the increasing influence of a TUTT cell to the north-
  east.   Due to the interaction with this TUTT cell, as well as with a
  subtropical ridge to the storm's southwest, Rex began moving on an 
  unusual east-southeasterly course which continued until early on 2 Sep.
  At 01/0000 UTC Rex was centered about 500 nm southeast of Tokyo with a
  100-kt MSW.  This east-southeasterly motion continued until very early
  on 2 Sep when the storm was located about 700 nm southeast of Tokyo, or
  more than 500 nm northeast of Iwo Jima.

     On 2 Sep Rex began once more to move to the east-northeast as it
  passed south of the upper LOW and came increasingly under the influence
  of a subtropical ridge to the northeast.  Over the next couple of days
  Rex moved slowly in a general east-northeasterly to easterly direction,
  due in part to the influence of a ridge to its south.  It also began to
  slowly weaken as it began to experience increased vertical wind shear.
  As the weakening Rex moved closer to a strong ridge to its east on the
  4th, it turned rather sharply to the north and began to accelerate to
  the north-northeast on 5 Sep ahead of an approaching mid-latitude
  trough.  As a 55-kt tropical storm Rex passed over 900 nm east of Tokyo
  very early on 6 Sep.   By 06/1800 UTC Rex's low-level center had become
  exposed with the nearest convection 45 nm to the north.

     The storm continued to race to the north-northeast as it began to
  transition into an extratropical storm.  The last advisories by both
  JMA and JTWC early on 7 Sep located Rex several hundred miles to the
  southwest of the westernmost Aleutian Islands.

     Roger Edson of JTWC posted an interesting discussion to one list
  about the unusually challenging forecast situation presented by Typhoon
  Rex and the resulting poor JTWC forecasts.    Practically all the
  numerical models were taking Rex northward and recurving it into the
  westerlies.  This was an especially convincing solution with a large
  blocking ridge anchored over the Pacific east of Japan and a very
  active frontal band over and just east of central Japan.  The only
  models not forecasting this were the Japanese Spectral and Typhoon
  models--both of which produced some pretty spectacular results for
  Typhoon Rex.  The main player leading to the bad forecasts was likely
  the TUTT (mentioned above), which was the primary factor responsible
  for Rex's unusual southeastward motion on 31 Aug and 1 Sep.  Roger's
  comments raise the question of whether or not some direct interaction
  (i.e., a Fujiwhara-type effect) happened between the TUTT and Typhoon
  Rex--a scenario which seems plausible considering that Rex's cyclonic
  circulation extended well above 200 mb, and this particularly strong
  TUTT extended well below 200 mb.    No doubt Typhoon Rex will serve
  as a catalyst for some debate and discussion amongst forecasters and
  atmospheric modelers.

     I located one press article which indicated that during the time
  Typhoon Rex was nearest to Japan in late August, heavy rainfall
  associated with the storm led to some significant flooding in parts 
  of Honshu with a loss of 16 lives.
  

                    Tropical Depression  (TC-07W)
                           1 - 4 September
                    -----------------------------

     The first tropical depression of September developed from a
  vorticity maximum in a trough located off the east coast of China.
  The system was steered throughout its existence by a mid-level ridge
  to the south, and was unable to intensify into a tropical storm due to
  persistent vertical shear.  JMA was the first meteorological agency
  to issue advisories on this depression, carrying it in their high
  seas Warning and Summary for about 24 hours before JTWC began issuing
  warnings.   The depression was first located (by JMA) about 100 nm
  west-southwest of Okinawa at 0600 UTC on 1 Sep.  Other than a (likely)
  readjustment of the center to the southeast, the depression moved on
  a rather straight east-northeasterly course throughout its life.

     The center of the depression passed to the south of Okinawa, and by
  02/0600 UTC (when JTWC commenced warnings) the system was located 200
  nm east-northeast of the island.  Continuing on its east-northeasterly
  track the depression passed about 350 nm south of Tokyo around 1200
  UTC on 3 Sep.  By 04/0000 UTC multi-spectral satellite imagery showed
  a tightly-wound low-level center indicative of 30-kt winds, but with
  the main convection sheared 65 nm to the east.  Thereafter, the system
  began to weaken and the last JTWC warning was issued at 1200 UTC on
  the 4th, placing the dissipating center about 400 nm east-southeast
  of Tokyo.


                   Typhoon Stella  (TC-08W / STS 9805)
                           12 - 18 September
                   -----------------------------------

     Stella was the first tropical cyclone of 1998 to make landfall in
  Japan, and was followed by two more within a week.  A high-pressure 
  ridge was evident in the Northwestern Pacific between 10N and 15N early
  in the month as Typhoon Rex moved well to the north.    This ridge 
  weakened and slipped slowly to the south as a weak LOW formed near
  20N, 165E on 7 Sep.   The circulation around this system developed 
  monsoonal trough characteristics and extended to the Philippine 
  Islands on the 8th where it linked up with the Southeast Asian monsoon
  by 10 Sep.  JTWC issued a Formation Alert at 12/0000 UTC; at the same 
  time JMA initiated depression advisories with the broad center located 
  roughly 150 nm north-northwest of Saipan.  (JTWC began issuing warnings
  six hours later.)  The depression was steered on a northwestward course
  by a subtropical ridge located to the northeast.

     JTWC upgraded the system to Tropical Storm Stella at 0600 UTC on
  13 Sep when it was centered about 175 nm south of Iwo Jima.  Stella
  was at this time a developing monsoon depression with gales covering
  an area 450 nm in diameter.  Winds were increased to 55 kts at 14/0000
  UTC based on synoptic observations from Iwo Jima.   (Stella passed
  about 70 nm west-southwest of the island around that time.)  The
  primary convective features were in the storm's southern hemisphere,
  with the largest area of convection located about 200 nm south of the
  center; however, during the 14th convection began to develop closer
  to the center and became better organized.

     JTWC upgraded Stella to a minimal typhoon at 15/0000 UTC when the
  storm was centered about 475 nm south-southwest of Tokyo.  Stella's
  course became more northerly and then north-northeasterly as it
  accelerated ahead of an approaching mid-latitude trough.  Even as a
  minimal typhoon, Stella exhibited a partially exposed low-level
  circulation center--its intensity was enhanced somewhat by the high
  translational speed.   Stella made landfall on the southern coast of
  Honshu near Shizuoka (about 125 nm southwest of Tokyo) shortly after
  1800 UTC on 15 Sep.  The storm raced north-northeastward across eastern
  Honshu at a forward speed reaching almost 50 kts!     Stella passed
  practically over Tokyo around 16/0000 UTC, and then parallel to and
  just inland from the east coast of Honshu, exiting east of Aomori 
  (which recorded a pressure of 975 mb) around 0600 UTC.  By 1200 UTC 
  the storm was poleward of the upper-level jet and rapidly losing 
  tropical characteristics.  The last advisories by both JTWC and JMA 
  were issued at this time with Stella east of Hokkaido and racing 
  northeastward.    The extratropical remnants continued rapidly
  northeastward and by the 18th were moving through the Aleutian Island
  chain.
  
     JMA did not classify Stella as a typhoon but as a severe tropical
  storm with a 60-kt MSW (10-min), whereas JTWC's highest estimated MSW
  (1-min) was 65 kts.  It should be pointed out that, considering the
  most commonly used conversion factors for converting from 1-min to
  10-min avg winds, the JMA-assessed intensity implies a slightly
  stronger storm.  A 60-kt 10-min MSW implies a 70- to 75-kt 1-min MSW,
  depending on the specific value used as a conversion factor.

     Press reports indicated that Stella was responsible for four deaths
  in Japan with 25 injuries reported.      Waves to 7 m battered the
  southern coast, and 356 mm of rain fell on an area which had received
  heavy rainfall in late August.  Over 3000 homes were flooded and the
  heavy rains were responsible for triggering 180 landslides.


                     Tropical Depression  (TC-09W)
                           13 - 14 September
                     -----------------------------

     A broad monsoon depression began developing in the northern South
  China Sea during the second week of September.  At 0600 UTC on the
  13th JTWC issued a warning locating a depression about 200 nm south-
  west of Hong Kong, or just east of the northern tip of Hainan Island.
  The stronger winds were located along the periphery of the system in
  the eastern semicircle.       The circulation associated with the
  depression covered an area over 400 nm in diameter.

     Steered by a subtropical ridge to the north, the depression tracked
  westward across the northern end of Hainan and into the Gulf of Tonkin.
  JTWC wrote its last warning at 1800 UTC on 13 Sep, but JMA and the
  Royal Observatory of Hong Kong continued to follow the system across
  the Gulf and inland near Vietnam's Red River Delta region around 0600
  UTC on the 14th.


                  Typhoon Todd/Emang  (TC-10W / TY 9806)
                            16 - 20 September
                  --------------------------------------

     A strong high pressure system (1040 mb) in the south Indian Ocean
  during the second week of September enhanced cross-equatorial flow,
  which in turn assisted a monsoon surge during the middle of the month.
  JTWC issued a Formation Alert at 0900 UTC on 15 Sep for a developing
  LOW in the monsoon trough.    The first warning, issued at 16/0000 UTC
  (about the same time that Stella was near Tokyo), directly upgraded the
  LOW to Tropical Storm Todd (called Emang by PAGASA).  Todd's center
  at this time was located about 350 nm south-southeast of Okinawa.
  Todd was a small system which developed rapidly--it was upgraded to a
  typhoon with 75-kt winds only 18 hrs after being named.  Gales at this
  time covered an area only 150 nm in diameter.  Todd initially moved to
  the east and east-northeast, but as a ridge to the east began to
  strengthen and build to the west, Todd turned to a northerly course.

     As Typhoon Todd turned to the north it continued to rapidly
  intensify.  At 17/0000 UTC manual satellite estimates of its intensity
  were 90 kts, but the relatively new digital Dvorak technique yielded
  125 kts--a MSW of 115 kts was set as a compromise.   Six hours later 
  Todd reached its peak estimated intensity with a MSW of 120 kts. (This
  was JTWC's maximum 1-min avg estimate--JMA's 10-min avg MSW never got
  higher than 75 kts.)  At peak intensity Todd passed northward about
  350 nm west of Iwo Jima.  The radius of 100-kt winds was estimated
  to be only 15 nm, but gales covered an area by this time about 360 nm
  in diameter.

     By 1800 UTC on 17 Sep Todd was embedded in a frontal system
  associated with a mid-latitude trough and was being steered by the 
  subtropical ridge to its east.   The typhoon was moving northward at
  17 kts and weakening.  As Todd approached the 30th parallel lower and
  mid-level ridging forced the storm to a westerly track.  Increasing
  vertical shear took its toll on Todd and the storm had weakened into
  a minimal typhoon by the time it passed westward through the Ryukyu
  Island chain.  At 18/0600 UTC reports of 50-kt sustained winds (1-min
  avg) and gusts to 63 kts from Tanegashima and Amamioshima Islands
  (relationship to storm's center unknown to author) indicated that the
  potential for stronger winds existed near the center of Todd.

     Todd passed about 200 nm north of Okinawa around 1500 UTC on 18 Sep
  as it moved westward into the East China Sea.  Cooler SSTs and vertical
  shear continued to cause Todd to weaken, and the storm made landfall
  in eastern China, about 85 nm south-southeast of Shanghai, as a
  tropical storm with an estimated MSW of 55 kts.  Once inland the
  storm quickly began to dissipate.  Todd's translational speed had
  been quite rapid as it curved northwestward toward the northern
  Ryukyus--between 25 and 30 kts--but slowed as it straightened out on
  its westward trek across the East China Sea.   By the time of landfall
  in China Todd was moving slightly south of due west at less than 10
  kts.

     Heavy rains from Todd in Japan (presumably on Kyushu) led to four
  fatalities from flooding with three more missing due to mudslides.
  I have no information regarding any adverse effects of this storm in
  the region where it made landfall in China.

                                                            
                  Typhoon Vicki/Gading  (TC-11W / TY 9807)
                              17 - 25 September
                  ----------------------------------------

     Vicki was the first tropical storm in September to form south of
  latitude 20N.  The system developed in the South China Sea from a LOW
  in the active monsoon trough.  A depression formed just west of
  northern Luzon, about 60 nm northwest of Baguio, on 17 Sep.  Initially
  the system was thought to be moving northeastward, but as its satellite
  signature improved, a drift to the southwest became apparent.

     JTWC upgraded the system to Tropical Storm Vicki at 17/1200 UTC.
  PAGASA had named the storm Gading, and since it significantly affected
  the Philippines early in its life, I shall refer to the storm by that
  name when discussing its history in the vicinity of that region.
  Gading drifted slowly southward and then southeastward before moving
  eastward into Luzon in the Lingayen Gulf region about 40 nm west-
  southwest of Baguio.   The storm was a small cyclone in the South
  China Sea--the radius of 50-kt winds was 25 nm while gales covered an
  area 120 nm in diameter.

     According to JTWC's analysis of satellite imagery Gading reached 
  typhoon intensity with an 85-kt MSW before making landfall in Luzon.
  JMA's 10-min avg MSW was given as 60 kts, which would imply a 1-min avg
  MSW of about 70-75 kts.  The author has some satellite imagery sent by
  Michael Padua which clearly shows a rather distinct eye.  PAGASA was
  reporting a MSW (10-min) of only 55 kts near the time of landfall;
  however, based upon the satellite imagery, JTWC's and JMA's analyses,
  and the magnitude of the damage reported (see below), it seems likely
  that Gading was a stronger tropical cyclone than this.

     While inland over Luzon Gading began to move on a northeastward
  course due the combined influences of a ridge to its southeast and
  a trough to its northwest.    The cyclone weakened into a minimal
  tropical storm while over Luzon, but began to re-intensify after
  moving out into the Philippine Sea south of Cabo Engano.  Vicki/Gading
  began to accelerate to the northeast and later north-northeast.  The
  storm regained typhoon intensity (as per both JTWC and JMA) at 0000
  UTC on 21 Sep when located about 125 nm southeast of Okinawa.

     Tropical Storm Waldo was located to the northwest of Vicki and the
  two cyclones perhaps interacted to some degree, but the deep trough
  to the northwest proved to be a very dominant steering influence and
  propelled both systems rapidly northward into Honshu.   Typhoon Vicki
  reached its peak intensity of 90 kts (as per JTWC's operational
  warnings) at 21/1800 UTC.  Moving north-northeastward at more than 30
  kts, Vicki made landfall in Japan over the peninsula south of Osaka.
  The typhoon passed just east of Osaka and Kyoto and west of Nagoya,
  barely reached the Sea of Japan coast in central Honshu, then curved a
  little more to the northeast, moving across northern Honshu and out
  to sea at almost the exact spot where Stella had exited Honshu. Vicki's
  forward translational speed reached 42 kts as it roared across Honshu.
  The storm weakened considerably due to the effects of vertical shear 
  and from being over land, and had merged with a frontal zone and become
  extratropical east of Hokkaido by 23/0000 UTC.  The resulting gale
  center continued northeastward, crossing the southwestern Aleutian
  Islands and the Bering Sea, and eventually moving into southwestern
  Alaska.

     At the time of landfall Vicki's MSW was being reported as 90 kts
  (1-min avg) and 75 kts (10-min avg) by JTWC and JMA, respectively.
  However, there was a synoptic observation of a 10-min avg sustained
  wind of 85 kts near the point of landfall south of Osaka.  This would
  imply a 1-min avg MSW of 100-110 kts, depending on the conversion
  factor utilized.   Vicki apparently continued to intensify and reached
  its peak intensity right at the time of landfall.

     One report was located by the author which indicated that heavy
  flooding due to Vicki/Gading's rains on Luzon contributed to 27 deaths.
  Also, Michael Padua sent some information regarding the storm's effects
  in Pangasinan province where it initially made landfall.  Damage was
  quite severe with about 90% of electrical/telephone poles down and
  many trees uprooted (especially mango).    Ten people were reported
  killed and thousands were left homeless in the province, especially in
  the city of Dagupan.  (This information was reported in the Philippine
  press--thanks to Michael for passing it along.)

     The author also discovered a report that the ship M/V Princess of
  the Orient sank on 18 Sep due to Typhoon Vicki with the loss of at
  least 39 lives, possibly more.  No information was given as to where
  the ship was located, but presumably was in the South China Sea.

     Regarding the effects of Typhoon Vicki in Japan--unfortunately I
  do not have any information, but given that the storm struck southern
  Honshu as a fairly strong typhoon, damage would likely have been
  quite extensive.    If anyone has any information describing the
  impact of Vicki on Japan, I would be grateful if they could e-mail it
  to me for inclusion in a future summary.


                      Tropical Depression  (TC-12W)
                            18 - 19 September
                      -----------------------------

     The JTWC on Guam issued a Formation Alert at 1100 UTC on 17 Sep
  for an area of convection in the South China Sea located just off the
  Vietnamese coast.  At 18/0600 UTC the first depression advisory was
  issued, placing the ill-defined center about 250 nm south-southeast of
  Hanoi, Vietnam.   This weak depression was a "coast hugger" which
  crept northwestward along the coast and eventually moved inland without
  further strengthening.  It did become slightly better organized on
  19 Sep as more convection formed closer to the center, but this did
  not persist.   By 19/1800 UTC the center was inland about 125 nm south
  of Hanoi and dissipating.

     JTWC was the only agency (that I'm aware of) that classified this
  system as a tropical depression and issued formal warnings.  In the
  summary portion of their high seas Warning and Summary bulletins,
  JMA did mention this system as a "low pressure area".


                 Tropical Storm Waldo  (TC-13W / TS 9808)
                             19 -22 September
                 ----------------------------------------

     Tropical Storm Waldo formed from another LOW in the monsoon trough
  to the northeast of Vicki.   JMA classified this system as a tropical
  depression at 0600 UTC on 19 Sep when it was located about 275 nm
  west-southwest of Iwo Jima.   The new depression initially drifted
  perhaps a little to the northeast before taking off on a north-
  northwesterly course toward Japan.   JMA upgraded the depression to
  Tropical Storm 9808 at 0600 UTC on 20 Sep when it was centered about
  250 nm west of Iwo Jima.  JTWC did not name the system until 1800 UTC,
  and it is interesting to note that at one point (1200 UTC), JMA was
  reporting a 45-kt MSW (10-min) while JTWC still had the system as a
  30-kt depression (1-min MSW).   JTWC did name the depression Tropical
  Storm Waldo at 20/1800 UTC but never reported the MSW to be higher
  than 40 kts during the storm's history.

     There was perhaps some interaction between Waldo and the rapidly
  approaching Tropical Storm (later Typhoon) Vicki, but the forecast
  merger between the two cyclones never happened.  The strong southerly
  flow ahead of a major trough to the west proved to be a very dominant
  steering influence and both storms moved rapidly northward toward
  Japan.      By 21/0000 UTC Waldo was moving rapidly northward at
  25 kts and the forecast no longer called for a merger between Waldo
  and Vicki, which had just been upgraded to a typhoon.

     Tropical Storm Waldo made landfall in southern Honshu on the
  peninsula south of Osaka, a short distance to the east of where a
  much-stronger Typhoon Vicki would make landfall less than 24 hours
  later.  Waldo passed just west of Nagoya, then moved rapidly across 
  the island and into the Sea of Japan in the Toyama Bay region.  By
  0000 UTC on the 22nd Waldo was rapidly weakening and becoming
  extratropical just southwest of southern Hokkaido.   There were some
  reports of widespread minor wind damage due to Waldo but apparently
  nothing verious serious.


                 Typhoon Yanni/Heling  (TC-14W / TY 9809)
                         25 September - 2 October
                 ----------------------------------------

     A broad monsoon LOW was located in the Philippine Sea several
  hundred miles east of southern Luzon in late September.  Late on 24
  Sep some ships were reporting winds to 30 kts so JTWC decided to
  initiate advisories at 25/0000 UTC.   The depression embarked on a
  slow movement to the northwest due to increased ridging to the east
  and northeast.   On 26 Sep the center seemed to re-develop farther to
  the west underneath an upper LOW in the Philippine Sea east-northeast
  of northern Luzon.     It was at this time that JMA and PAGASA began 
  writing advisories on the system (PAGASA assigning the name Heling).
  By 0000 UTC on the 27th an exposed low-level circulation center was 
  visible near the northern end of a curved cloud band, but proved to
  be a transient feature.

     Heling continued moving generally on a northwestward course toward
  Taiwan.  By 1200 UTC on 27 Sep JTWC estimated the 1-min MSW to have
  reached 35 kts so the system was named Tropical Storm Yanni at a
  location about 200 nm southeast of Taipei.   As Yanni approached the
  east coast of Taiwan early on 28 Sep it turned abruptly to the north
  and skirted the east coast of the island a short distance offshore.
  Yanni at this time exhibited features of both a monsoon depression and
  a more conventional tropical cyclone.    Gales were occurring up to
  320 nm to the northeast of the center along a cloud band on the
  periphery of the circulation, with a zone of weaker winds between this
  cloud band and the central convection.     Gale-force winds were
  estimated to extend out only 40 nm west of the center.

     Yanni was upgraded to a typhoon by JTWC at 0600 UTC on 28 Sep when
  the storm was centered about 85 nm south-southeast of Taipei.  This
  action was taken primarily based upon radar observations of an eye
  from the Taipei instrument plus Dvorak classifications of T4.0 based
  on visible imagery.  Shortly afterward, the Taipei radar showed the
  eye to be contracting, leading to a forecast for a 120-kt typhoon
  by 48 hours.  This forecast for rapid intensification, however, did not
  materialize.

     Yanni was moving northward into a weakness in the subtropical ridge
  and was being influenced by a trough approaching from the west.  By
  29/0600 UTC Yanni reached its peak MSW of 80 kts when the center was
  located about 200 nm west-northwest of Okinawa.  The storm was at this
  time moving northeastward through the East China Sea at 18 kts.  By
  1800 UTC Yanni's forward motion had increased to 23 kts and the typhoon
  was experiencing vertical shear from the trough to the west.   Typhoon
  Yanni was beginning to weaken rapidly and was downgraded to a tropical
  storm at 0600 UTC on 30 Sep.  (It should be noted that JMA classified
  Yanni as a minimal typhoon for only a 6-hour period on 30 Sep.)

     Yanni passed directly over Cheju Island south of Korea around 0000
  UTC on 30 Sep.  The rapidly weakening tropical storm reached the South
  Korean coast later on the 30th and then, as a building anticyclone
  over China extended eastward, began to move slowly back to the south-
  southeast.  Most of the convection, however, was sheared off and shot
  northeastward over parts of southern Korea and western Kyushu where
  it caused heavy rains.     The exposed low-level center that remained
  continued drifting to the southeast and by 0000 UTC on 2 Oct was
  located between Kyushu and Okinawa with only 20-kt winds.

     In a private e-mail Roger Edson of JTWC mentioned that heavy rains
  in South Korea reportedly led to 24 deaths with many homes and rice
  fields damaged or destroyed.  (Thanks to Roger for passing along that
  piece of information.)

     If anyone has any further information regarding the effects of
  Yanni (or any of the Northwestern Pacific tropical cyclones for that
  matter)--if they will pass it along to me I'll be happy to include it
  in a future summary.

  ***********************************************************************

  NORTH INDIAN OCEAN (NIO) - Bay of Bengal and Arabian Sea

  Activity for September:  1 tropical cyclone of gale intensity


                       Tropical Cyclone  (TC-04A)
                        30 September - 1 October
                       --------------------------

     The third tropical cyclone of the year to form in the Arabian Sea
  was a very weak, short-lived tropical cyclone.  The first JTWC warning
  located the system about 400 nm west of Bombay at 0000 UTC on 30 Sep.
  Although the cyclone developed in a sheared environment, the shear
  was forecast to relax and the tropical cyclone was forecast to
  strengthen.  However, as the system moved slowly westward the shear
  increased and the cyclone began to weaken.  The last JTWC advisory
  placed the dissipating center about 550 nm west of Bombay at 0000 UTC
  on 1 Oct.

  ***********************************************************************

  SOUTH INDIAN OCEAN (SIO) - South Indian Ocean West of Longitude 90E

  Activity for September:  2 tropical disturbances **

  ** - This based upon the La Reunion TCWC's assessment.  One of these
  systems (A2) was carried by JTWC as a minimal tropical storm.

  NOTE:  Normally I do not cover tropical disturbances which are not
  classified as tropical depressions by at least one TC warning center.
  However, as pointed out in the Global Tropical Cyclone Summary for
  July, 1998, La Reunion utilizes a more restrictive definition for
  the tropical depression category, requiring that a system have 10-min
  avg winds of Beaufort force 7 (28 kts).  But advisories are issued
  for some weaker tropical disturbances which may cause winds (10-min
  avg) of up to 25 kts.      Since at least some of these tropical
  disturbances would likely be carried as tropical depressions by other
  TCWCs, I am reporting on them in these summaries.    Also, for an
  explanation of La Reunion's unique way of designating weaker unnamed
  systems, see the South Indian Ocean section of the July summary.


                         Tropical Disturbance "A1"
                              3 - 6 September
                         -------------------------

     In a Tropical Weather Outlook for the Indian Ocean, issued at 1800
  UTC on 3 Sep, JTWC mentioned an area of disturbed weather near 10S,
  85E which had some development potential.  A Formation Alert was issued
  at 04/0630 UTC but was cancelled 24 hours later.   The only source of
  tracking information for this system are the advisories issued by the
  RSMC La Reunion TCWC.   The first advisory located the disturbance's
  center about 775 nm east of Diego Garcia at 1200 UTC on 3 Sep.  The
  system moved to the west-southwest until around 1200 UTC on 4 Sep,
  then commenced to move on a westerly track.   The last advisory placed
  the weakening disturbance about 200 nm southwest of Diego Garcia at
  0600 UTC on 6 Sep.

     The MSW reported was only 25 kts, but the advisories cautioned
  that winds to 30 kts might be experienced up to 200 nm south of
  the center due to a tight gradient with a HIGH to the south.


                     Tropical Disturbance "A2"  (TC-02S)
                           29 September - 2 October
                     -----------------------------------

     JTWC issued a Formation Alert for a disturbed area in the South
  Indian Ocean at 29/0600 UTC.   La Reunion issued the first advisory
  at this time, placing the center of a disturbance about 400 nm east-
  southeast of Diego Garcia.  JTWC followed with a warning at 0900 UTC.
  This system had formed from a vorticity maximum in the monsoon trough 
  to the southeast of Diego Garcia.   La Reunion reported the MSW as 
  25 kts (although near gales were forecast to be occurring well to the 
  south of the center due to a tight pressure gradient) while JTWC
  carried the system as a minimal tropical storm.    The system was
  initially expected to intensify as the inhibiting vertical shear was
  forecast to weaken, but instead the shear remained strong or even
  increased.   La Reunion wrote only one more advisory (at 1200 UTC)
  and then dropped the system.    By 1800 UTC the low-level center was
  100 nm east of the nearest convection.    The system drifted westward
  and was dropped by JTWC at 30/0600 UTC when the center was estimated
  to be about 275 nm east-southeast of Diego Garcia.

     However, based on ship reports and scatterometer winds to 30 kts,
  warnings were resumed by JTWC at 01/1200 UTC with the MSW estimated at
  40 kts.   The revived system was located well to the south and west
  of its last reported position--about 375 nm south of Diego Garcia.
  The system was moving more or less to the south-southwest and shear
  soon began to increase once more.  The final warning at 1200 UTC on
  2 Oct placed the weakening depression about 425 nm northeast of
  Rodrigues Island.  (It should be noted that La Reunion did not issue
  any advisories on this system during 1 and 2 October.)

  ***********************************************************************

  AUSTRALIAN REGION (AUG) - From Longitude 90E Eastward to Longitude 160E

  Activity for September:  No tropical cyclones

  ***********************************************************************

  SOUTHWEST PACIFIC (SWP) - South Pacific Ocean East of Longitude 160E

  Activity for September:  No tropical cyclones

  ***********************************************************************

                              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 included in the July summary.
  I will include this glossary from time to time, primarily in "lean"
  months without a lot of tropical cyclone activity to cover.  But if
  anyone missed receiving it and wishes to obtain a copy, send me an
  e-mail privately and I'll forward 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) The files will be named with an obvious nomenclature--using
           August as an example:   aug98.tracks

    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 September summary is the twelfth cyclone summary in this series;
  the first one covering the month of October, 1997.  If anyone did
  not receive any of the previous summaries, they may be downloaded
  from the aforementioned FTP site at HRD.   The summary files are
  catalogued with the nomenclature:  sep98.sum, for example.

    Back issues can also be obtained from the following websites
  (courtesy of Michael Bath and Michael V. Padua):

    http://australiansevereweather.com/cyclones/>
    http://www.geocities.com/capecanaveral/6825> (since January only)

    The preliminary storm reports for all the 1997 Atlantic and Eastern
  North Pacific tropical cyclones are available on the Tropical
  Prediction Center's website:  http://www.nhc.noaa.gov> .  These
  reports include the analyzed best-track for each cyclone.  The staff
  of JTWC is also working on an on-line version of their Annual Tropical
  Cyclone Report for 1997.  It is still under construction, but the
  best-track files are already available for 1997 Northwestern Pacific
  and North Indian Ocean cyclones.  The URL is:

       http://www.npmocw.navy.mil/npmocw/prods/jtwc.html> 

  Prepared by: Gary Padgett
  E-mail:  [email protected]
  Phone:  334-222-5327

  ***********************************************************************
  ***********************************************************************

Document: summ9809.htm
Updated: 18th March 2008

[Australian Severe Weather index] [Copyright Notice] [Email Contacts] [Search This Site]