Tropical Cyclones
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 August 1999
[Summaries and Track Data] [Prepared by Gary Padgett]


                              AUGUST, 1999

  (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.)


                   !!!!! IMPORTANT NOTE BY AUTHOR !!!!!

     At the outset of writing these summaries, I determined that I would
  not introduce any tropical cyclone nomenclature, names, or numbering
  systems that were not assigned by any tropical cyclone warning center.
  This month I am making a departure from this policy in a limited
  fashion.   During August there were several weather systems in the
  Northwest Pacific basin which, in the opinion of some very
  knowledgeable and trained meteorologists (and also in the author's
  opinion), were very likely tropical storms but for which no tropical
  cyclone warning agency issued any warnings.    Since these systems
  deserve further study, and since they were clustered both temporally
  and spatially; in order to reduce confusion in any possible discussions
  which this summary may engender, I have decided to assign unofficial
  names to three of them which have no other name or number.

     To label these cyclones I have decided to use the letters of the
  Greek alphabet.  The letters of the Greek alphabet are widely known
  through their frequent use as symbols in various scientific
  disciplines (e.g., mathematics, physics, astronomy) and should lead
  to no cross-cultural problems (which can happen, for instance, when
  a common English name may sound similar to a "bad" word in another
  language).  In the future, whenever some weather system occurs which:
  (1) in the opinion of some reliable professional meteorologist(s),
  was quite possibly a tropical cyclone (or else a hybrid with strong
  tropical-like features);  (2) likely contained 1-min MSW of at least
  34 kts; and (3) was not officially assigned a name or number by any
  TCWC--then I shall report on it in a monthly summary and designate it
  with the next available letter of the Greek alphabet.    I do not
  anticipate this happening very often, and with 24 letters in
  the Greek alphabet, it should be quite a few years before it would be
  necessary to start over.  (Incidentally, for those who are not aware
  of it, the names of the Greek letters are the backup names for the
  Atlantic and Eastern North Pacific basins should the regular list of
  names be exhausted in any given year.)


                         !!!!! NEW ACRONYMS !!!!!

     About a year ago I developed a glossary of frequently-used acronyms
  and abbreviations which are commonly and widely used in tropical
  cyclone-related discussions.   I have included the glossary in two or
  three monthly summaries which were shorter in length.  This month's
  summary is anything but short, so I don't want to make it longer by
  appending the glossary at the end.  But there is one very common
  acronym which is widely used in tropical cyclone discussions which I
  am going to start employing:  LLCC - which stands for low-level
  circulation center.   I am tired of having to type out that rather
  lengthy term so many times.     Another one which I have started using
  in this summary is CP - central pressure.


                           AUGUST HIGHLIGHTS

  --> Tropical Atlantic comes alive with a bang--first time since prior
      to 1944 with two Category 4 hurricanes before 1 September
  --> Catetory 4 hurricane strikes sparsely populated south Texas
  --> Large Category 2 hurricane brushes southeast Atlantic coast
  --> Long-lived hurricane travels from Eastern Pacific, through the
      Central Pacific, and across the Dateline into the Western Pacific
  --> Large, sprawling typhoon affects Philippines and southern China
  --> Spectacular outbreak of high-latitude midget tropical cyclones in
      Western North Pacific

                           ACTIVITY BY BASINS

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

  Activity for August:  1 tropical storm
                        3 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.  A special thanks to John Wallace, a tropical cyclone
  enthusiast and college student from San Antonio, for providing me with
  a log which he had kept of all Atlantic/Northeast Pacific tropical
  waves that proved to be very valuable in helping to trace the pre-
  depression history of some of the cyclones.   John also sent me much
  of the rainfall, surface observations, and damage reports on Hurricanes
  Bret and Dennis.

                       Atlantic Activity for August

     The pattern of tropical cyclone activity in the Atlantic basin
  during August, 1999, was similar to that during August, 1998.  In both
  years the tropical Atlantic basin was very quiet through most of the
  first three weeks of the month, then things became quite active with
  four named storms forming before month's end.    Also in both years,
  the portions of the basin seeing cyclogenesis were similar:  the
  tropical Atlantic between the Antilles and Cape Verde Islands and
  the western Gulf of Mexico.   And North Carolina and Texas were the
  two U. S. states affected by tropical cyclones in both years.

     One aspect in which the two years differed, however, was in the
  number and severity of major hurricanes.   August, 1998, produced one
  major (or intense) hurricane--Hurricane Bonnie--which just managed to
  reach Category 3 on the Saffir/Simpson scale whereas in August of this
  year, two hurricanes became severe Category 4 hurricanes.    Bret was
  a small, intense hurricane in the extreme western Gulf of Mexico which
  made landfall in a very sparsely populated county in south Texas while
  Cindy was a large, severe Cape Verde hurricane which reached its peak
  intensity as it recurved northeastward into the North Atlantic several
  hundred miles east of Bermuda.

     According to the Best Track file for the Atlantic basin, 1999 is
  the first year since the beginning of aerial reconnaissance in 1944
  in which two Category 4 hurricanes occurred before the beginning
  of September.    The only other year which possibly may have achieved
  this distinction was 1958.    Hurricane Cleo in mid-August appears to
  have been a solid Category 4, even allowing for the supposed upward
  bias in MSW values prior to 1970.   But Hurricane Daisy, just off the
  U. S. East Coast in late August (taking the Best Track MSW estimates
  at face value), was only a Category 3 with a peak MSW of 110 kts--yet
  the lowest measured central pressure was 935 mb--rather low for a
  MSW of only 110 kts.   Dunn and Miller, in the summary of the 1958
  hurricane season in their classic book _Atlantic Hurricanes_, report
  the highest winds (presumably sustained) in Daisy as being in excess of
  150 mph (130 kts), which admittedly seems to be a little on the high
  side, although the peak MSW and minimum central pressure for Hurricane
  Georges in September, 1998, were 135 kts and 937 mb.  Perhaps when the
  on-going revision of the Atlantic hurricane database is completed,
  Daisy's MSW will have been revised upwards a bit.

                          Hurricane Bret  (TC #3)
                               18 - 25 August

     Bret was a small but vicious hurricane which popped up in the Bay
  of Campeche, moved northward and northwestward and struck the coast
  of Texas about halfway between Brownsville and Corpus Christi as an
  intense hurricane.  (Whether or not Bret should be regarded as a
  Category 3 or 4 hurricane at landfall will be discussed later.)
  A tropical wave left the coast of Africa around 3 Aug and made its way
  across the tropical Atlantic and Caribbean Sea.  The wave apparently
  interacted with a mid-level trough in the Northwestern Caribbean
  around mid-month and the resulting area of disturbed weather moved
  across the Yucatan Peninsula on 16-17 Aug.   After moving out into the
  Bay of Campeche the disturbance began to show increasing convective
  organization.      A USAF Reserves Hurricane Hunter flight on the
  afternoon of 18 Aug found that a surface circulation had formed, so
  advisories were initiated on Tropical Depression #3 at 2100 UTC.  The
  broad, poorly-organized center was located around 275 nm east-southeast
  of Tampico, Mexico.   (The parent tropical wave continued on across
  Central America into the Eastern North Pacific where it eventually
  developed into TD-11E on 23 Aug.)

     The convective pattern was rather chaotic at first and the system
  meandered around in the Bay of Campeche for about 48 hours.  During the
  morning of the 19th convection began to increase with a 180 nm-wide
  band covering the LLCC.  An early afternoon reconnaissance flight found
  a central pressure of 1006 mb and an inner wind maximum forming with
  39-kt flight level winds only 7 nm from the center; therefore, the
  depression was upgraded to Tropical Storm Bret at 2100 UTC.  Bret was
  located roughly 225 nm east-southeast of Tampico.      (A morning
  reconnaissance flight had found 35-45 kt winds 90 nm from the center
  just offshore of Veracruz, but this is an area notorious for wind
  funneling effects.)

     Dr. Jack Beven of TPC/NHC has raised the interesting question of
  just what caused Bret's "inner spinner" to form.   A Hurricane Hunter
  flight had found a very broad center with the radius of maximum winds
  on the order of 50-80 nm on the first three fixes.  However, the final
  fix (near 1700 UTC on the 19th) showed the development of the inner
  wind maximum with a radius of maximum winds of 7 nm coincident with
  the formation of the aforementioned convective band near the center.
  Examination of the aircraft data suggests that this wind maximum
  formed in place and did not contract in from a greater distance as
  is often the case.

     The next aircraft into the storm found a 10 nm-wide eye--a somewhat
  unusual feature in a tropical storm with a central pressure of 1000 mb
  and 56-kt winds at 450 m flight level.   Once having formed, this
  very tight core remained with Bret throughout its life.  The radius
  of maximum winds in subsequent aircraft fixes was never more than 22 nm
  and usually less than 15 nm.   A reconnaissance flight early on 20 Aug
  found a peak flight-level wind of 60 kts only 4 nm from the center.

     Tropical Storm Bret continued to steadily increase in intensity on
  20 Aug as it moved slowly northward parallel to the Mexican Gulf Coast.
  By 21/0000 UTC the central pressure had dropped to 981 mb and a GPS
  dropwindsonde in the north eyewall reported surface winds of 62 kts;
  therefore, Bret was upgraded to a hurricane at 0300 UTC at a position
  about 175 nm east of Tampico.   The central pressure in Bret dropped
  from 981 mb at 0000 UTC down to 976 mb at 1800 UTC.       A GPS
  dropwindsonde reported a maximum wind of 112 kts at 900 mb in the
  eyewall so the official MSW at 2100 UTC was reported as 90 kts.

     Between 1800 UTC and 0000 UTC, 22 Aug, Bret underwent a period of
  very rapid intensification.    A reconnaissance flight during the early
  evening of 21 Aug found maximum flight level winds of 110 kts, but
  GPS dropwindsondes revealed far, far stronger winds at lower levels.
  Winds of 149 kts were measured at 888 mb.  A mean boundary layer wind
  of 141 kts was measured and 155 kts was found at 90 m above the
  surface. One dropwindsonde was reporting 135 kts shortly before falling
  into the ocean.     Bret's CP dropped from 976 mb at 21/1703 UTC to
  958 mb at 21/2115 UTC--a drop of 18 mb in four hours, which translates
  to a drop of 4.5 mb per hour.

     Bret was upgraded from a Category 1 hurricane to a Category 4
  hurricane with 115-kt MSW in a space of 12 hours.  The hurricane
  was centered about 150 nm east-southeast of Brownsville, Texas,
  at this time.      To further complicate the picture, as least from 
  a forecasting scenario, Bret's predicted turn to the northwest had not
  occurred and the hurricane temporarily accelerated a bit on a straight
  northward course, being influenced by an upper-level ridge over the
  Yucatan Peninsula and a cold LOW over Texas.

     Severe Hurricane Bret was at its estimated peak intensity of 120 kts
  from 0600 UTC on 22 Aug until the western edge of the eyewall made
  landfall on Padre Island around 1700-1800 UTC.  Hurricane force winds
  only extended out 30 nm from the center and gales covered a zone less
  than 150 nm wide.   The hurricane began to turn more to the northwest
  (as had been consistently forecast) very early on 22 Aug and by the
  time of landfall was moving just north of due west.  The small, intense
  hurricane made landfall in Kenedy County, which, according to the
  _1998 World Almanac_, had in 1996 a resident population of 438 (humans)
  while press reports indicated the county was home to over 55,000 head
  of cattle.   The landfall point of Bret was almost exactly halfway
  between Brownsville and Corpus Christi, being about 60 nm from either.

     At 22/1900 UTC Buoy 42020 off the Texas coast reported an 8-min avg
  wind of 58 kts with a peak gust of 73 kts.  Lowest pressure was 989.2
  mb and the peak wave height was 8.2 m.  (The relationship of the buoy
  to Bret's center at this time is unknown to the author.)  At 2200 UTC,
  just after the center of the hurricane had made landfall, the CMAN
  station at Port Aransas (near Corpus Christi) reported a peak 2-min
  avg wind of 41 kts with a peak gust of 50 kts.

     The hurricane was downgraded to a tropical storm at 23/1200 UTC
  and to a dissipating depression at 2300 UTC.  Bret drifted across
  Kenedy County into Brooks County where the last NHC advisory was
  written at 1800 UTC.  The remnant depression drifted northwestward
  to a point north of Laredo and then turned back to the west, crossing
  the Rio Grande and dissipating in northern Coahuila State, Mexico.

     A branch of the Brownsville Forecast Office located in Kenedy
  County reported estimated winds of 80-85 kts.   The lowest measured
  central pressure in Bret was 944 mb at 22/1200 UTC, and the pressure
  was 946 mb when the western edge of the eyewall reached Padre Island.
  During the early morning hours of 23 Aug a wind gust of 70 kts was
  reported in Falfurrias, Texas.    Doppler radar rainfall estimates
  indicated that about 760 mm of rain fell on Kenedy County and up to
  250 mm on both Corpus Christi and Laredo.  Rainfall amounts of 125 mm
  or more were common over much of southern Texas.    A storm report
  from the Hydrometeorological Prediction Center stated that NESDIS
  satellite estimates indicated that 200 mm of rain may have fallen
  in north-central Mexico, just southwest of Maverick County, Texas,
  between 1745 and 2345 UTC on 24 Aug.   By early evening of the 24th
  the Rio Grande at Laredo was at 2.57 m (flood stage being 2.44 m)
  and the Aransas River near Skidmore was almost at flood stage of
  3.96 m.

     There were four deaths near Laredo, Texas, related to a collision
  between a truck and tractor on some slick roads.   Bret's storm surge
  cut 12 new inlets through the long, narrow Padre Island.  There were
  at least four confirmed tornadoes spawned by the hurricane--perhaps

     An interesting question to ponder is:  Should Bret be regarded as
  a Category 3 or Category 4 hurricane on the Saffir/Simpson scale at
  landfall?   Patrick Hoareau, a great tropical cyclone enthusiast from
  Rennes, France, sent me several excellent radar images of Bret as it
  approached the coast.  At 22/2300 UTC the western eyewall was directly
  over Padre Island and the convection in that portion of the storm
  was more intense than it had been a couple of hours earlier.  The
  2100 UTC NHC advisory reported the MSW at 120 kts, but in an
  intermediate bulletin at 2300 UTC, the MSW had been lowered to
  110 kts.   Given the increase in the intensity of the convection in
  the western eyewall between 2100 and 2300 UTC, it seems quite plausible
  that the 120-kt winds were still present when the western eyewall
  moved over Padre Island, which, in the author's opinion, would qualify
  Bret as a landfalling Category 4 hurricane.

     In any case, overall damage from Bret was relative minor due to the
  storm's small size and the fact that it made landfall in a virtually
  unpopulated stretch of coast.     If this hurricane, at the same
  intensity, had made a direct strike on a large city, damage would
  likely have been several billion dollars.   Or to look at it from
  another perspective--Bret was comparable in size to the famous Labor
  Day Storm which devastated the Florida Keys in 1935.   If a storm
  like that one with its 892-mb central pressure and possibly 175-kt
  winds had struck Kenedy County, Texas, the overall damage would likely
  not have been a whole lot greater than that caused by Bret.

                        Hurricane Cindy  (TC #4)
                             19 - 31 August

     A very strong, well-organized tropical wave moved off the African
  coast on 18 Aug and immediately began to show signs of further
  development.  Advisories were initiated on Tropical Depression #4
  at 0000 UTC on 19 Aug when the center was located about 300 nm east-
  southeast of the Cape Verde Islands.   Ship VRVK6, located about
  150 nm south of the depression's center, reported a pressure of 1011 mb
  and southwest winds of 33 kts at 19/0000 UTC.   There were some very
  cold cloud tops to -80 C south and west of the center, and satellite
  intensity estimates by the morning of 19 Aug had reached 35 kts from
  TAFB but were only 25 kts from SAB and AFGWC.  There was a 45-60 nm
  spread in the satellite-based center fixes.  The depression was under
  easterly shear of about 25-30 kts at the time.

     After some initial relocation of the center, the depression
  commenced on a westerly track at around 13 kts.  During the evening of
  the 19th the little available data on the depression showed strongly
  mixed signals.  Satellite and microwave imagery showed the system to be
  adversely affected by northeasterly shear with no convection close to
  the presumed center.  On the other hand, ship 4QUM reported southwest
  winds of 45 kts at 1800 UTC and 37 kts at 20/0000 UTC--the second
  report being about 150 nm south-southeast of the center.  The NHC
  specialist on duty, James Franklin, wrote that he interpreted those
  winds as reflecting the large-scale flow of the ITCZ, perhaps enhanced
  by the depression.   By the afternoon of the 20th, the depression was
  still sheared but the center had become more involved with the
  convection, so the system was upgraded to Tropical Storm Cindy at
  2100 UTC with the center located about 175 nm southwest of the Cape

     Visible imagery during the morning of 21 Aug showed a respectful-
  looking banding pattern, and Dvorak estimates nudged upwards during
  the day.  At 22/0300 UTC Cindy was upgraded to a 65-kt hurricane
  about 475 nm west-southwest of the Cape Verdes.     The easterly
  shear, however, remained strong and Cindy's forward motion slowed
  considerably, keeping the storm in a shearing environment.   As a
  result, Cindy was downgraded back to a tropical storm at 1800 UTC on
  22 Aug.   Over the next few days, Cindy's center was at times exposed
  east of the deep convection, but the storm had a large cyclonic
  envelope and was able to hold its own fairly well.  The estimated MSW
  never dropped below 50 kts and remained at that value from 23/0000 UTC
  through 24/1200 UTC.  By the morning of 25 Aug Cindy had changed little
  in intensity but the cloud pattern had become more organized.   The
  vertical shear had decreased considerably and the underlying waters
  were becoming warmer.  Cindy was reclassified as a 65-kt hurricane at
  0300 UTC on 26 Aug.   At 1800 UTC ship ELUU6 reported winds of 60 kts
  about 90 nm from the center; hence, the MSW was upped to 75 kts at
  2100 UTC.

     Hurricane Cindy had been moving on a general west-northwesterly
  track for several days, but by the time it regained hurricane intensity
  the storm was beginning to move northwestward in a broad recurve which
  would take it well east of Bermuda.   Visible images on the morning
  of 27 Aug showed a ragged eye and the official MSW was increased to
  90 kts.   By 28/0000 UTC Dvorak numbers from TAFB and SAB had reached
  T6.0, so Cindy's MSW was increased further to 100 kts, making it the
  second major hurricane of the season.    Six hours later the hurricane
  displayed a 35 nm-wide eye embedded in a CDO with cloud tops of -65
  to -75 C.  The MSW was increased further to 115 kts, making Cindy the
  second Category 4 hurricane within a week.   Cindy's winds reached
  their peak of 120 kts at 1200 UTC on the 28th and maintained that
  intensity for 18 hours.  The lowest CP in Cindy, based upon satellite
  estimates only, was 944 mb.  By 29/0600 UTC the hurricane was beginning  
  to weaken a bit as the MSW decreased to 110 kts.  Cindy was at this
  time 325 nm due east of Bermuda and was beginning to move to the north-

     At the time of peak intensity Hurricane Cindy was a very large
  storm.  Hurricane force winds covered a zone about 135 nm in diameter
  while tropical storm force (gale force) winds swept an area almost
  500 nm in diameter.   After recurvature Cindy began to weaken fairly
  quickly as SSTs decreased and vertical shear increased.  Winds were
  down to 80 kts by 1200 UTC on 30 Aug, and Cindy was downgraded to
  a 60-kt tropical storm at 31/0900 UTC.   At 1200 UTC Cindy was located
  about 425 nm southeast of Cape Race, Newfoundland, and the MSW was
  down to 45 kts.   By midday visible satellite images revealed that
  Cindy was rapidly being absorbed by a much stronger extratropical
  cyclone approaching from the northwest.

     One final note--Cindy was absorbed into a large, vigorous extra-
  tropical LOW, but the hurricane itself had been an absorber.  About
  the time that Cindy reached peak intensity, the remnants of tiny
  Tropical Storm Emily were absorbed into Cindy's much larger and
  stronger circulation.

                       Hurricane Dennis  (TC #5)
                        24 August - 7 September

     Hurricane Dennis apparently formed from the northern extent of a
  tropical wave which emerged off the African coast around 15 Aug.
  (The southern extent of the wave was responsible for the formation
  of Tropical Storm Emily.)   The wave moved across the Atlantic for
  several days without much associated convection.   Thunderstorms
  began to increase on 21 Aug when the system was located a few hundred
  miles east of the Leeward Islands.  The convection became even better
  organized on the 22nd and a Hurricane Hunter plane investigated the
  area on 23 Aug.   No surface circulation could be located by the
  reconnaissance flight, but by later in the day satellite imagery
  indicated that the convection had become more concentrated and a ship
  with call sign WZJF reported southeast winds of 30 kts about 90 nm
  northeast of the center.      Advisories were begun on Tropical
  Depression #5 at 24/0000 UTC when the center was estimated to be
  about 175 nm northwest of San Juan, Puerto Rico.

     During the morning of 24 Aug there were mixed signals regarding
  the depression's intensity.  Satellite intensity estimates from TAFB
  and SAB were 35 kts and 30 kts, respectively, and ship WZJF reported
  35-kt winds.     However, the wind direction at the ship remained
  unchanged over six hours, suggesting that the circulation was poorly
  defined or elongated, and this was supported by an observation from
  ship 3FOB5, located west of the estimated center, of east winds of
  only 5 kts.   A reconnaissance flight during the morning found only a
  very broad, ill-defined center with strong gusty winds well to the
  east of the possible center.       However, another flight in early
  afternoon found that the system had strengthened some and a special
  intermediate advisory, upgrading the depression to Tropical Storm
  Dennis, was issued at 1800 UTC with the MSW estimated at 40 kts.
  The storm's poorly-defined center was located about 400 nm northwest
  of San Juan and all the tropical storm force winds were confined to
  the eastern and northern quadrants.

     For the next three days Dennis moved on a slow northwesterly course
  parallel to the Bahama Islands.   During the 25th the center became
  more involved with the deep convection and the MSW had increased to
  55 kts by 25/1800 UTC.   Outflow was well-defined in all quadrants
  although there were some low clouds moving away from the circulation
  in the northwest quadrant.   An early evening flight by the Hurricane
  Hunters found FLW of 67 kts and a CP of 998 mb--down 10 mb in 24
  hours.  However, the next flight early on 26 Aug found FLW of 80 kts
  and pressure down to 995 mb, so Dennis was upgraded to a 65-kt
  hurricane at 0900 UTC located about 400 nm east-southeast of Miami.
  Most of the models were indicating that Dennis would intensify rather
  quickly into a strong Category 2 or Category 3 hurricane, but this
  did not materialize.   An upper-level anticyclone over central Cuba
  restricted outflow somewhat over the southwest quadrant, and the
  inner core of Dennis never got well-established.  The storm's MSW
  reached 70 kts at 26/1200 UTC and remained pegged there for a couple
  of days although the CP dropped down to 975 mb during this time.

     During the night of 27-28 Aug Hurricane Dennis turned to more of a
  north-northwesterly heading which carried it just east of Great Abaco
  Island in the Bahamas.  (Less than three weeks later a much, much
  stronger Hurricane Floyd would follow a very similar path through the
  same area.)   By 28/0000 UTC ham radio operators at Little Harbor
  on Great Abaco reported wind gusts to 58 kts.  Radar images relayed
  from the NOAA P-3 reconnaissance aircraft showed a fully-formed eye
  35-40 nm in diameter.  The center was embedded within very cold cloud
  tops of -75 C and the CP was down to 980 mb.  During the evening
  Man-of-War Cay reported gusts to 68 kts.   The center of Dennis passed
  very close to Green Turtle Cay on the morning of the 28th where ham
  radio operators reported that wind gusts reached 100 kts with some
  storm surge flooding.   There were reports of significant damage to
  boats and crops in the Abacos where winds were still gusting to 60 kts
  at 1500 UTC.

     Around 0600 UTC on 29 Aug the hurricane passed about 30 nm east of
  NOAA buoy 41010 (28.9N, 78.5W).   The buoy reported a peak 8-min avg  
  wind of 54 kts at 0500 UTC; a peak gust of 72 kts at 0800 UTC; a 
  minimum pressure of 980.2 mb at 0800 UTC; and a peak wave height of 
  7.6 m at 2100 UTC on 28 Aug.  (Thanks to John Wallace for collecting 
  and sending me this information.)

     Dennis began to strengthen on 28 Aug, reaching its estimated peak
  intensity of 90 kts with a CP of 969 mb by 1800 UTC.  At this time the
  hurricane was centered less than 200 nm off the central Florida coast
  and about 325 nm south-southeast of Charleston, South Carolina.  A GPS
  dropwindsonde on the morning of the 29th measured a peak wind of 109
  kts.  The large-scale circulation pattern favored intensification and
  ocean temperatures were quite warm, but the lack of a well-defined 
  inner core prevented Dennis from reaching its full potential intensity-
  wise.  On the afternoon of the 29th the storm sported a ragged banding-
  type eye 45-50 nm in diameter.   The hurricane by this time was moving
  due northward off the Georgia coast, and by 0000 UTC on 30 Aug had
  begun to turn to the north-northeast.   Dennis' CP fell to its lowest
  value in the storm's history at 30/0000 UTC when a reconnaissance
  flight measured 963 mb.

     On 30 Aug the hurricane turned to a northeasterly heading which
  was parallel to and about 100 nm off the Carolina coasts.  Dennis'
  forward motion accelerated to about 18 kts for a few hours as it
  sped by Cape Hatteras.   Sustained winds of 44 kts were reported from
  Oak Island near Cape Fear as Dennis raced by.    The automated station
  at Frying Pan Shoals (33.5N, 77.5W)  reported sustained winds of
  hurricane force from 0800 UTC through 1300 UTC as the storm passed
  off the North Carolina coast, reaching a maximum of 80 kts with a
  peak gust to 93 kts at 1000 UTC.   Dennis' CP remained below 970 mb
  throughout the 30th and a USAF reconnaissance flight during the
  afternoon found 98-kt winds in the southeast quadrant at flight level.
  GPS dropwindsonde data over several days showed that Dennis was not
  particularly effective at bringing high winds down to the surface.
  (In contrast Hurricane Bret was very effective at doing this, which
  just goes to underscore the extreme individuality of hurricanes.)

     Dennis passed about 120 nm north of NOAA buoy 41002 (32.3N, 75.2W)
  around 1600 UTC on the 30th.  At 1200 UTC that buoy was reporting 8-min
  avg winds of 41 kts (which continued through 1600 UTC), a peak gust
  to 58 kts, and a peak wave height of 11.3 m.  The minimum pressure of
  997.6 mb was reported at 1100 UTC.   The by-now weakening hurricane
  was located about 30 nm north of NOAA buoy 41001 (34.7N, 72.6W) at
  31/1200 UTC.    However, most of the extremes from this buoy were
  reported near 30/2200 UTC when the center was still more than 50 nm
  to the west-northwest.   The 8-min avg wind was 47 kts, peak gust
  was 60 kts, and the peak wave height was 10.3 m at 2200 UTC while the
  minimum pressure of 976 mb was reported at 31/0400 UTC.  As Dennis
  swept by the North Carolina coast on 30 Aug it was a rather large
  hurricane: the diameter of hurricane force winds was around 130 nm
  and the area experiencing gales was approximately 300 nm wide. (Thanks
  to John Wallace and Eric Blake for sending the buoy information to me.)
     Dennis' forward motion slowed considerably and the storm came to
  a crashing halt approximately 100 nm east of Cape Hatteras by early
  on 31 Aug and remained quasi-stationary in the same general spot for
  nearly three days before beginning a slow drift to the south.  During
  this time the convection associated with Dennis all but disappeared
  completely, due initially to vertical shear and the intrusion of cool,
  drier air.    The hurricane was downgraded to a 60-kt tropical storm
  at 0000 UTC on 1 Sep.   The NHC Discussion Bulletins on 31 Aug and
  1 Sep consistently remarked that Dennis looked more like an extra-
  tropical LOW than a tropical cyclone--there was little convection and
  the strongest winds were not close to the center.  Only the sporadic
  re-development of heavy showers near the center gave any hint of
  tropical characteristics.  At this stage Dennis barely met Dvorak
  classification criteria.

     However, the mid-latitude westerlies shifted to the north, reducing
  the shear, and Dennis remained near the Gulf Stream, which of course
  is a vast oceanic heat source.     Visible satellite images on the
  morning of 2 Sep showed a well-organized system with denser clouds
  near and over the center, although the tops were not very cold.  During
  the afternoon a vessel with call sign ELFV3 reported west-southwest
  winds of 45 kts about 50 nm southwest of the center.   During the
  evening hours there was some limited deep convection noted east of the
  center with tops of about -45 C.   Dennis at this stage still showed
  some subtropical characteristics as it appeared to be embedded within
  a larger-scale cyclonic circulation.

     Tropical Storm Dennis drifted generally southward to a point about
  150 nm southeast of Cape Hatteras by 03/1800 UTC.  With a high-pressure
  ridge building to the northeast, the storm commenced on a slow, steady
  northwesterly track toward the North Carolina coast.  The MSW remained
  near 50 kts for about three days during this time with the CP running
  from about 985-990 mb.   On the morning of the 4th WSR-88D data from
  Morehead City, North Carolina, indicated that Dennis was becoming
  better organized.  Reflectivity data showed 65-kt winds aloft, and an
  earlier reconnaissance flight had found FLW of 66 kts.  A GPS drop-
  windsonde reported surface winds of 63 kts and a CP of 984 mb as Dennis
  was making landfall on the North Carolina coast around 2100 UTC just
  northeast of Cape Lookout and a short distance east of the Morehead
  City/Beaufort area.  An automated weather station at Cape Hatteras
  reported a peak gust of 53 kts about the time of landfall.  

     Dennis continued moving northwestward across North Carolina as it
  weakened, being downgraded to a depression at 05/0600 UTC.   By the
  evening of the 5th winds were down to 10-15 kts at lower elevations
  near the center of Dennis while winds in excess of 25 kts were still
  being reported at higher elevations of southwestern Virginia and
  western North Carolina and also in coastal waters off North Carolina
  and Virginia.   The weakening system passed near Raleigh-Durham,
  North Carolina and entered Virginia near Danville.  It then turned
  more to the north, passing between Roanoke and Lynchburg, across
  northeastern West Virginia, extreme western Maryland (passing near
  Cumberland), western Pennsylvania (passing near Johnstown), and on
  into western New York.    The final position supplied by the Hydro-
  meteorological Prediction center placed the weak LOW over extreme
  southern Lake Ontario, just northwest of Rochester, New York.

     By Sunday evening, 5 Sep, up to 150 mm of rain had fallen over
  north-central North Carolina, with totals of 125 to 200 mm common
  over upslope areas in the Blue Ridge Mountains.   A storm total of
  214 mm was reported at Mills Creek Dam in Augusta County, Virginia;
  and 209 mm at North Mountain in Rockbridge County, also in Virginia.

     There was significant overwash near Hatteras, and some homes near
  Nags Head (that were built right on the water) got washed away by
  sand erosion and waves.    In the Outer Banks there were some power
  outages, minor flooding, loss of some shingles and siding, etc.;
  however, overall damage from Dennis was minor.    (The information
  in this paragraph was gleaned from a posting to the WX-TALK discussion
  list by Mike Mogil from Rockville, Maryland.)

                      Tropical Storm Emily  (TC #6)
                             24 - 28 August

     Tropical Storm Emily apparently formed from the southern extent of
  a tropical wave which emerged off the African coast around 15 Aug.
  (The northern extent of the wave was responsible for the formation
  of Hurricane Dennis.)   The wave moved across the Atlantic for
  several days without much associated convection.   Thunderstorms
  began to increase on 21 Aug when the system was located a thousand
  miles or so east of the Windward Islands.    The system was mentioned
  in the six-hourly Tropical Weather Outlooks issued by TPC/NHC over the
  next few days as being somewhat organized and having some potential
  for development.   By the morning of 24 Aug satellite images revealed
  that the disturbance was quite well-organized with indications of a
  circulation although there was some unfavorable shear in the area.

     A Hurricane Hunter reconnaissance flight was dispatched to
  investigate the area around midday.   Surprisingly, they found a very
  small, tightly-wound tropical storm with a CP of 1004 mb and MSW of
  55 kts!    The new storm was named Emily, and this was only about two
  hours or less after Dennis had been named.   Tropical Storm Emily was
  a small storm with gales extending out from the center about 50 nm.
  Winds of 50 kts were confined to the northeastern quadrant within about
  25 nm of the center.  At 24/1800 UTC Emily was centered about 340 nm
  east-southeast of Barbados.   Emily was moving very slowly northward,
  apparently due to the larger Tropical Storm Cindy to the east which was
  disrupting the tradewind flow.

     During the night of 24-25 Aug northerly shear increased and Emily's
  center became exposed with only limited deep convection south of the
  center.  The MSW was decreased to 40 kts on the 25/0300 UTC advisory.
  The next afternoon a reconnaissance flight found FLW of 45 kts and a
  CP of 1005 mb, so it appeared that Emily was holding its own against
  the shear.  A French buoy with ID 41101, near 14.6N, 56.2W, reported
  a pressure of 1009.2 mb and east-southeast winds of 27 kts (apparently
  at 26/0600 UTC).     This observation helped to verify the synoptic
  position and motion.   Tropical Storm Emily moved slowly northwestward
  to a position about 260 nm east of Guadeloupe by 1800 UTC on the 26th.
  A mid-level circulation had been left behind 150 nm to the south of the
  LLCC, so Emily was downgraded to a tropical depression at 2100 UTC.

     At this juncture Emily began to move at a slightly faster pace on
  an almost due northward track.     During the night of 26-27 Aug
  convection began to increase around Emily's LLCC and outflow over the
  system began to improve somewhat.   A reconnaissance flight early on
  27 Aug made two center fixes that were almost 50 nm apart.  The second
  fix had the better defined center with a CP of 1009 mb and a FLW of
  44 kts to the east of the center.  Since the fixes were so far apart
  and the convection was still poorly organized, Emily remained 
  classified as a depression for the time being.

     A flight into the depression in the early afternoon of the 27th
  found a CP of 1007 mb and a peak FLW of 49 kts at 450 m; therefore,
  Emily was upgraded once more to a tropical storm.   Emily was a very
  tiny storm--tropical storm force winds extended out only 10 nm to the
  northeast and 25 nm to the southeast.  Six hours later the gale radii
  in these quadrants had increased to 25 nm and 30 nm, respectively.
  The circulation of Emily would have just about fit inside the eye of
  Hurricane Cindy.

     Emily's revival as a tropical storm was short-lived--on the morning
  of 28 Aug the distance between tiny Emily and the very large, severe
  Hurricane Cindy continued to decrease.      The 0900 UTC Discussion
  Bulletin indicated that low-level cloud elements to the west of Emily
  were moving from west to east around the large circulation of Cindy,
  indicating that Emily was getting entrained into Cindy.    The final
  advisory at 1500 UTC indicated that Emily was dissipating as it was
  being absorbed into Hurricane Cindy.     The 28/1200 UTC position
  placed the center of the dissipating Emily about 375 nm south-southeast
  of Cindy's center.

     It is interesting to note that for the first two days of Emily's
  life, even with the much larger Cindy approaching from the east and
  the outflow from Dennis impinging on the storm from the northwest,
  and in spite of the persisent shear over the small storm, some of the
  numerical models were still predicting that Emily would reach hurricane
  force.    This possibility was reflected in official TPC/NHC forecasts
  through 26/1500 UTC, which was the advisory before Emily was first
  downgraded to a depression.


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

  Activity for August:  2 tropical depressions
                        1 tropical storm
                        2 hurricanes

  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.  

     A special thanks to John Wallace, a tropical cyclone enthusiast
  and college student from San Antonio, for providing me with a log
  which he had kept of all Atlantic/Northeast Pacific tropical waves
  that proved to be very valuable in helping to trace the pre-depression
  history of the cyclones.  

                 Northeast Pacific Activity for August

     Two hurricanes came to life in Northeastern Pacific waters during
  August, both of which moved across 140W into the Honolulu AOR (Central
  North Pacific).   One of these, Hurricane Dora, continued on across
  the Central Pacific and crossed the International Dateline into the
  Northwest Pacific Basin.   Dora was the first tropical cyclone to
  originate east of 140W and emerge into the Western Pacific since
  Hurricane John of Aug-Sep, 1994.    In addition to Hurricanes Dora
  and Eugene, Tropical Storm Fernanda and two tropical depressions
  formed during the month.  Both of the non-developing depressions were
  quite short-lived.  The first, TD-09E, formed on 13 Aug about 650 nm
  southwest of Cabo San Lucas, Mexico.  This depression likely formed
  from a tropical wave which had first been identified in the tropical
  Atlantic east of the Lesser Antilles on 1 Aug and had crossed Central
  America into the Northeast Pacific basin on 9 Aug.  The depression had
  a broad center which remained exposed east of an area of deep
  convection.   The system moved slowly west-northwestward and had
  dissipated by 15/0000 UTC less than 200 nm from its point of origin.

     TD-11E formed around 1800 UTC on 23 Aug about 125 nm southwest of
  Cabo San Lucas.    This depression formed from a tropical wave which
  left the African coast on 3 Aug and entered the Northeast Pacific on
  16 Aug.  (This was the same wave which appeared to be instrumental
  in the formation of Hurricane Bret in the Gulf of Mexico.)  Even as
  the first advisory was issued on this system it was nearing cooler
  SSTs and convection was on the wane.  The depression initially moved
  northwestward, then was relocated to the southwest at 24/1200 UTC.
  The final advisory at 1800 UTC placed the dissipating center about
  200 nm west-southwest of Cabo San Lucas.

                   Hurricane Dora  (TC-07E / TS 9911)
                             6 - 23 August

     A tropical wave left the coast of western Africa on 23 Jul and made
  its way westward across the tropical Atlantic.    The main wave axis
  had entered the eastern Caribbean by 29 Jul and had slipped across
  Central America and into the Eastern Pacific by 4 Aug.  Late on 5 Aug
  the cloud pattern associated with the disturbance had become better
  organized with a curved band forming to the west of the deepest
  convection.   Also, a ship (KSBG) reported a 30-kt easterly wind from
  near the center and a pressure of 1003 mb.  Advisories were initiated
  on Tropical Depression 07E at 06/0000 UTC with the center estimated
  to be about 300 nm south-southwest of Acapulco.

     The depression maintained deep convection with good banding during
  the day and at 1800 UTC was upgraded to Tropical Storm Dora about
  350 nm south-southwest of Manzanillo.   Dora continued to steadily
  increase in strength throughout 7 Aug and at 0600 UTC on 8 Aug was
  upgraded to a 65-kt hurricane based on the presence of a symmetric
  and cold CDO and a Dvorak T-number of 4.5 from TAFB.   Dora at this
  time was centered approximately 450 nm south-southwest of Cabo San
  Lucas.   The MSW had increased to 90 kts by 09/0000 UTC and SSM/I
  imagery at 0100 UTC depicted an eye that was open on the west side.
  Throughout the day Dora continued to display a small, well-defined
  eye 5-10 nm in diameter, and the MSW had increased to 100 kts by
  10/0000 UTC, making Dora a Category 3 hurricane on the Saffir/Simpson
  scale.   The eye of Hurricane Dora was located about 675 nm southwest
  of Cabo San Lucas at this time.

     Guided by a strong and persistent subtropical ridge, the constancy
  of Dora's westward track for thousands of miles across the tropical 
  Pacific was amazing.  From the point at which Dora was upgraded to
  a tropical storm on 6 Aug to the point at which it crossed the Dateline
  on the 19th, the storm moved westward almost 4500 nm while gaining only
  about 250 nm in latitude.     Dora became a Category 4 hurricane at
  1800 UTC on 10 Aug and maintained that status through 1200 UTC on
  the 13th, reaching a peak MSW of 120 kts and a minimum CP of 943 mb
  at 0000 UTC on 12 Aug when Dvorak estimates from SAB, TAFB, and AFGWC
  reached 127 kts, 115 kts, and 115 kts, respectively.     During the
  period when Dora was at peak intensity, the diameter of the eye was 
  about 20 nm and surrounded by a ring of solid convection with cloud
  tops generally between -65 and -75 C.

     By the afternoon of 13 Aug some stable trade wind clouds were
  being advected into the hurricane from the north and west and the
  cloud tops were warming.   Dvorak numbers were decreasing and Dora's
  MSW was lowered to 110 kts at 1800 UTC.    Hurricane Dora crossed
  140W and entered the CPHC's area of responsibility around 14/0600 UTC
  at a point about 800 nm east-southeast of Hilo, Hawaii.   Dora was
  moving into cooler waters and the MSW had dropped to 85 kts by this
  time.  The peak winds were further reduced to 70 kts six hours later.
  However, on 15 Aug Dora appeared to re-intensify.   The Hurricane
  Hunters flew some reconnaissance missions into Dora since it posed
  a potential threat to Hawaii.   CPHC increased the MSW to 90 kts
  at 15/0600 UTC and to 100 kts 24 hours later as Dora passed about
  200 nm south of South Point on the Big Island.  During a reconnaissance
  mission on the 16th a GPS dropwindsonde measured a maximum wind of
  120 kts at the 982-mb level with a mean boundary layer wind of 114 kts.
  Satellite imagery at this time was certainly supportive of a Category
  3 hurricane on the Saffir/Simpson scale.  Although quite intense as
  it passed south of Hawaii, the only effects of the storm were enhanced
  trades and some high surf along the Big Island's southeast-facing

     After passing Hawaii Dora began to steadily weaken as its forward
  motion accelerated.  The hurricane passed about 65 nm south of tiny
  Johnston Island around 0600 UTC on 18 Aug as a minimal hurricane.  The
  strongest winds measured on the island were 35 to 40 kts in gusts with
  a minimum pressure of 1011 mb.   By comparison, Hurricane John passed
  nearer the island on 25 Aug 1994 as a more intense storm and produced
  serious damage that was then estimated at $15 million.    After passing
  by Johnston Island Dora continued to sail westward toward the Dateline.
  The storm crossed into the Northwest Pacific basin around 2100 UTC
  on 19 Aug at a point about 650 nm west of Johnston Island.  The first
  warning from JTWC downgraded Dora to a 60-kt tropical storm.

     Shortly before reaching the 180th meridian Dora was beginning to
  turn to a slightly more west-northwesterly track.  After crossing the
  Dateline the weakening storm turned more to the northwest.   As Dora
  traversed the easternmost reaches of the Northwest Pacific basin, it
  began to run into increasing southwesterly shear.  SSM/I data from
  around 21/0000 UTC indicated only an isolated area of convection
  about 40 nm northeast of the LLCC.   The MSW was dropped to 40 kts
  at 0600 UTC based on satellite intensity estimates of 25 kts and
  45 kts.   JTWC downgraded Dora to a tropical depression at 22/0000 UTC
  when the system was centered about 300 nm northeast of Wake Island.
  The LLCC was fully exposed about 75 nm west of the deepest convection.
  (JMA was of course issuing warnings on Dora and maintained the system
  as a tropical storm through 23/0000 UTC.)   Convection persisted for
  a couple of days in spite of the shear due to an area of diffluence
  over the region.     The final position of Dora, supplied by JMA, 
  located the dissipating system about 475 nm north-northeast of Wake
  Island at 1200 UTC on 23 Aug.

     The factors responsible for the rather sudden and unexpected 
  re-intensification of Dora to major hurricane status over marginally 
  warm waters are somewhat obscure.   Rich Henning, a Major in the
  USAF Reserves and a meteorologist with the 46th Weather Squadron at
  Eglin AFB--and a member of the 53rd Weather Reconnaissance Squadron
  (The Hurricane Hunters)--suggests that Dora's resurgence was an
  example of the effects of mesoscale and convective scale inner core
  processes when they are undisturbed by significant shear.   Early
  on 16 Aug there was a burst of strong convection in the northeast
  quadrant just as the eyewall drop mentioned above was released.
  Drops into the eye the previous day and night reported eye surface
  air temperatures around 24 C, but the lack of any shear had allowed
  the eyewall to consolidate very nicely.   However, Dr. Mark Lander
  of the Water and Environmental Research Institute at the University
  of Guam questions whether or not Dora actually re-intensified all
  that significantly.  Mark's contention is that the hurricane had not
  previously weakened as much as the operational advisories had

                        Hurricane Eugene  (TC-08E)
                              6 - 15 August

     Hurricane Eugene formed from the tropical wave which left the coast
  of Africa just before the one from which Dora developed. The pre-Eugene
  system started its march across the Atlantic around 19 Jul, was near
  the Lesser Antilles on 25 Jul, and had entered the Northeast Pacific
  basin by 25 Jul--about four days ahead of the pre-Dora wave.   First
  visible satellite imagery on 6 Aug indicated that an area of disturbed
  weather about 900 nm southwest of Cabo San Lucas had rapidly increased
  in organization and that a tropical depression had formed. By afternoon
  the small system had remained well-organized and was developing some
  good banding features; therefore, it was upgraded to Tropical Storm
  Eugene at 1800 UTC.   Eugene seemed to intensify in discrete steps.
  After reaching tropical storm intensity, the MSW held at 35 kts for
  about 24 hours.  At 07/1800 UTC a cloud band was observed wrapping
  around the center giving the appearance of the formation of a banding-
  type eye; therefore, the MSW was increased from 35 to 50 kts.

     The intensity inched upward to 55 kts on the next advisory but
  remained there until 08/1800 UTC when a small eye 10 nm in diameter
  had made an appearance in satellite images.   At that point Eugene
  was upgraded to a 70-kt hurricane at a location about 1150 nm
  southwest of Cabo San Lucas.       By 1200 UTC on 9 Aug Eugene had
  reached its estimated peak intensity of 95 kts which it maintained
  for 24 hours.  The estimated minimum CP was 965 mb.    These peak
  measurements were based on Dvorak estimates of a strong T5.0 from
  both TAFB and SAB.   Eugene weakened some on the 10th as the eye
  disappeared and the CDO slipped southeastward from the storm's center.
  The MSW was decreased to 85 kts at 1200 UTC but held there for a
  day or so.     During the early evening hours the CDO had become
  somewhat irregular but had cold cloud tops to -80 C.

     Eugene had moved on a course slightly north of due west from its
  formation until it reached hurricane intensity.   From that point the
  storm moved west-southwestward for a day or so, then took off on an
  almost due-west track which carried it into the Central North Pacific
  sub-region.  Eugene entered Honolulu's AOR around 1200 UTC on 11 Aug
  at a point about 1000 nm east-southeast of Hilo.   Eugene by this time
  had weakened some with the MSW estimated at 75 kts.    The hurricane's
  intensity remained in a steady state at 75 kts for about 36 hours
  before more rapid weakening set in.

     The Hurricane Hunters made a flight into the storm on 12 Aug and
  found the center farther south than anticipated based on the previous
  track.  Shearing had increased slightly and the LLCC was exposed.  The
  MSW was decreased to 65 kts at 1800 UTC, and Eugene was downgraded to
  a 55-kt tropical storm six hours later at a position about 575 nm
  east-southeast of Hilo.   Shearing over the storm continued and Eugene
  slowly weakened as it moved westward.  The system was downgraded to a
  tropical depression at 15/0000 UTC when it was centered about 350 nm
  south-southwest of South Point on the Big Island.  From this point the
  depression drifted west-southwestward for another 24 hours.  SSTs were
  still warm (27 C) and vertical shear decreased significantly in this
  region but Eugene was not able to regenerate.  By 1800 UTC on 15 Aug
  only isolated convective cells remained south of the LLCC, so CPHC
  issued the last advisory at that time with the weak center located
  about 550 nm east-southeast of Johnston Island.

                   Tropical Storm Fernanda  (TC-10E)
                            17 - 22 August

     A tropical wave left the coast of western Africa on 1 Aug but
  apparently weakened on 4 Aug near 39W.    Another wave was introduced
  near 52W on 5 Aug in the Tropical Weather Discussions issued four times
  daily by TPC/NHC.   (John Wallace, who supplied me with tropical wave
  information, thought that this second wave was possibly a
  re-development of the earlier wave.)   The new (or else rejuvenated)
  wave reached the eastern Caribbean on 7 Aug and had emerged into the
  Eastern Pacific by 11 Aug.   The disturbance continued to move to the
  west and by 17 Aug was located several hundred miles south-southwest
  of Baja California.   A scatterometer overpass at 0600 UTC showed a
  well-defined circulation with 25-kt winds about 75 nm from the center.
  With abundant convection over the western semicircle and Dvorak
  intensity estimates up to 30 kts from SAB, the first advisory on
  Tropical Depression 10E was issued at 1500 UTC.  The poorly-organized
  center was estimated to be about 650 nm south-southwest of Cabo San

     The new depression was undergoing significant easterly shear which
  kept the strong convection pushed to the west of the LLCC.  Dvorak
  intensity estimates from TAFB, SAB, and AFGWC had all reached 35 kts
  by 18/0000 UTC but the system was still carried as a tropical
  depression due to the poorly-defined center.  By 0600 UTC multi-channel
  satellite imagery indicated that the LLCC was just on the northeastern
  edge of the deep convection so the depression was upgraded to
  Tropical Storm Fernanda on the 0900 UTC advisory.     The system
  had been moving on a west-northwesterly track, and this motion
  continued through 1200 UTC on 19 Aug when Fernanda reached a point
  about 750 nm west-southwest of Cabo San Lucas.  The storm reached its
  peak intensity of 55 kts and minimum estimated CP of 994 mb at this
  time.  Fernanda remained in a shearing environment but the shear had
  decreased enough that some convection had wrapped around the eastern
  side of the center.

     After reaching peak intensity Fernanda began moving on a slow
  west-southwesterly track and slowly weakened.  The easterly shear
  continued to affect the cyclone and SSTs were slowly cooling off.
  Fernanda had weakened to a minimal 35-kt tropical storm by 1200 UTC
  on the 20th, and would likely have been downgraded to a depression
  at 1800 UTC had it not been for some SSM/I wind speed estimates of
  30 kts which suggested that higher winds might be found underneath
  the convective area.   However, by the 21/0300 UTC advisory Fernanda
  had become a tight swirl of low clouds with a cluster of deep
  convection to the southwest of the LLCC; hence, it was downgraded to
  a tropical depression at that time.  Occasional bursts of convection
  continued for a day or so but the system continued to slowly spin
  down.   The final advisory placed the dissipating center almost
  1300 nm west-southwest of Cabo San Lucas at 1200 UTC on 22 Aug.


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

  Activity for August:  3 tropical depressions (a)
                        7 tropical storms (b)
                        4 typhoons (c)

  (a) - it is unknown to the author if JMA classified any of these
        systems as depressions; one of these was possibly a midget
        tropical storm but was not treated as such by JTWC

  (b) - one of these was a visitor from east of the Dateline; three
        of these were not warned on as tropical cyclones by any
        TCWCs in the Northwest Pacific basin

  (c) - three of these were considered typhoons only by JTWC, the
        other being Typhoon Olga which was active at the beginning
        of the month and was briefly classified as a minimal typhoon
        by JMA

  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 '99 webpage, for sending me the PAGASA and JMA tracks.

     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.

                  Northwest Pacific Activity for August

     "It was a month of interesting discrepancies between the various
  warning centers in matters of tropical cyclone classification."  This
  statement opened the discussion of Northwest Pacific tropical activity
  in the July summary.   But the discrepancies alluded to in the July
  summary were nothing compared to what happened in August.  The monsoon
  trough extended to fairly high subtropical latitudes during the month,
  and there was an outbreak of several small, midget tropical cyclones.
  Three of these were not warned on by either JMA or JTWC, while amongst
  the other three, there were very significant differences in intensity

     For information on the three unnamed/unnumbered systems, I am
  indebted to Dr. Mark Lander of the University of Guam for providing
  me with much information and some satellite imagery of these systems.
  Dr. Lander is a professor at the Water and Environmental Institute of
  the Pacific at the University of Guam, and for 10 years worked under 
  a grant from the Office of Naval Research to conduct research on 
  tropical cyclones.  Dr. Lander worked closely during those years with
  forecasters and satellite analysts at JTWC.  A very special thanks to
  Mark for cluing me in on these systems and sending me all the

     There were three non-developing tropical depressions during August,
  but one of these, TD-18W, is covered separately below since Mark Lander
  felt it was quite likely a tropical storm.   Tropical Depression 14W
  emerged from the cloud band of a monsoon gyre southeast of Japan.  An
  exposed LLCC was noted as early as 7 Aug to the south of Iwo Jima.
  The weak system drifted northward and by 1800 UTC on 8 Aug had become
  sufficiently organized that JTWC initiated depression warnings.  The
  center was located about 130 nm north-northwest of Iwo Jima at that
  time.  The depression drifted generally north-northwestward throughout
  its life, steered by a ridge to the northeast.  A TUTT located to the
  west and an anticyclone to the east created southerly shear which
  prevented the depression from strengthening significantly.   Mark
  Lander sent me an observation from Chichijima at 09/0000 UTC of
  southerly winds of 25 kts with a peak gust to 41 kts.  This was after
  the depression had made its closest approach to the island.  On the
  basis of this observation and the system's satellite signature, Mark
  was of the opinion that this depression could have been a 35-45 kt
  tropical storm at the time.    The system weakened as it approached
  Japan and was only a weak depression when it limped ashore southeast
  of Osaka.

     Tropical Depression 15W may have been traceable to a disturbance
  noted east of Luzon on 8 Aug which drifted slowly northward.  This
  disturbance appeared to weaken on the 14th northeast of Okinawa.
  JTWC initiated warnings on the system at 0600 UTC on 16 Aug when an
  exposed LLCC was located less than 100 nm west of southern Kyushu.
  The depression was possibly a hybrid-type system, having both tropical
  and extratropical features.  The system drifted east-northeastwards 
  towards southwestern Kyushu, then turned northward across the island 
  and dissipated in the Korea Strait to the north.  The depression was in
  a vertical shear environment which prevented it from strengthening--
  MSW were never estimated higher than 25 kts.   A TUTT over southeast
  Korea created diffluence which enhanced convection, but the persistent
  westerly shear displaced most of the convection to the east, resulting
  in the heaviest rains falling on Shikoku and southern Honshu.

                 Typhoon Olga/Ising  (TC-11W / TY 9907)
                            28 July - 3 August

     Typhoon Olga was active into the early days of August, but since the
  storm began in July, it was covered in the Global Tropical Cyclone
  Summary for July, 1999.   Please refer to that summary for details on

                 Tropical Storm Paul  (TC-12W / TS 9908)
                               3 - 8 August

     A broad LLCC developed southwest of Guam on 1 Aug with convection
  in all quadrants but in an environment of fairly strong vertical
  shear.  The LOW moved north-northwestward and by the next day was
  located west of the Marianas.  The system had the appearance of a
  typical monsoon depression with strongest winds on the periphery of
  the circulation.   JTWC issued a Formation Alert on 2 Aug and another
  one on the 3rd.  At this stage the system was still poorly organized
  and exhibited multiple small LLCCs within the larger circulation.
  JTWC initiated tropical depression warnings at 1200 UTC on 3 Aug when
  the broad center was estimated to be about 325 nm southwest of Iwo

     On 4 Aug the depression was still poorly organized with several
  smaller LLCCs, but there were ship reports of winds to 30 kts.  Early
  on the 4th the center appeared to re-organize farther to the north as
  two cloud bands developed north of the system.     JMA upgraded the
  depression to TS 9908 at 04/1200 UTC when the center was located
  approximately 335 nm west of Iwo Jima.   Based upon ship reports of
  35-kt winds, JTWC upgraded the system to Tropical Storm Paul at
  1800 UTC.    Paul basically tracked northwestward along the periphery
  of a large monsoon gyre that was centered to the southeast of Okinawa.
  Paul was initially undergoing shear and the LLCC was exposed, but on
  the 5th some deep convection wrapped more tightly around the center.
  Based on this and some ship reports of winds to 45 kts, JTWC estimated
  the peak intensity of Paul to have been 50 kts at 0600 UTC while the
  storm was centered about 250 nm east of Okinawa.   (JMA's 10-min MSW
  estimate at this time was 45 kts, which is in close agreement.)

     However, later on 5 Aug Paul began to weaken as apparently some
  dry air entrainment caused the deepest convection to collapse.  JTWC
  downgraded the storm back to a depression at 06/0000 UTC.    Paul
  merged with the gyre east of Okinawa and began to resemble a large
  monsoon depression once more with strongest winds along the periphery.
  The depression continued to move on a northwesterly track which
  carried it across southern Kyushu and into the Yellow Sea.  Convection
  increased some as it crossed Kyushu but was mostly sheared to the
  north.  (JMA maintained Paul as a tropical storm until it had crossed
  Kyushu.)  The center of the depression passed just south of Cheju
  Island at 07/0800 UTC where Mosulpo reported a minimum pressure of
  992 mb and a MSW of 25 kts (presumably a 10-min avg).  The final JTWC
  warning at 0000 UTC on 8 Aug placed the weakening center in the Yellow
  Sea just off the coast of China.  The remnants were forecast to move
  inland and dissipate near Qingdao.

                  Tropical Storm Rachel  (TC-13W / TS 9909)
                                5 - 9 August

     An area of convection formed in the South China Sea on 4 Aug and
  synoptic observations indicated that a broad circulation might be
  forming in the monsoon trough.  JMA classified the system as a tropical
  depression at 1800 UTC on 5 Aug when it was centered about 100 nm
  south-southeast of Hong Kong.     On 6 Aug the convection began to
  rapidly increase in organization and the upper-level environment was
  favorable for further intensification; therefore, JTWC initiated
  warnings at 0600 UTC.  Six hours later, based primarily upon remote
  wind measurements from scatterometer and microwave data, the system
  was upgraded to Tropical Storm Rachel about 250 nm east of Hong Kong.
  The storm was sheared somewhat as it was steered to the east-northeast
  by a mid-level ridge to the southeast.  At 07/0000 UTC the satellite
  current intensity estimate was only 25 kts, but there were ship
  reports of winds to 35 kts along the periphery.  

     Rachel moved rather quickly in an east-northeasterly direction and
  crossed Taiwan between 0600 and 1200 UTC on 7 Aug.  The storm weakened
  as it crossed the Chungyang Mountains and JTWC temporarily dropped
  warnings.  JMA, which had not yet classified the system as a tropical
  storm, continued issuing depression bulletins.  After crossing Taiwan,
  Rachel continued moving rapidly east-northeastward and began to
  recover from the effects of moving over the island.     JTWC
  re-instated warnings on the system at 1800 UTC on 7 Aug when the
  system was centered about 150 nm east of Taipei while JMA upgraded
  Rachel to TS 9909 at this time.   Based on synoptic ship reports of
  winds to 35 kts JTWC upgraded Rachel back to tropical storm intensity
  at 08/0600 UTC when the storm was just northwest of Okinawa.  However,
  the storm was sheared with the deepest convection located about 100 nm
  north-northeast of the center of circulation.

     Rachel turned to the north around 08/1200 UTC and later to the
  north-northwest as a subtropical ridge built over Japan.  The storm
  also began to weaken and was downgraded to a depression at 1800 UTC.
  Some sporadic convection continued, but by 09/1800 UTC scatterometer
  and microwave data indicated no winds greater than 20 kts and virtually
  no convection, so the final warning was issued at that time with the
  dissipating center between Kyushu and Shanghai.

                 !!!!!  YET ANOTHER SPECIAL NOTE  !!!!!

     I am now about to embark on something I've not done since I began
  writing these monthly summaries--cover systems for which warnings were
  not issued operationally by an established TCWC, but rather for which
  the information came from an unofficial, private (but very 
  knowledgeable and well-qualified) source.  I've already explained my
  use of the Greek alphabet names and revealed the source of my
  information on these storm systems.  The purpose of this note is to 
  underscore the fact that by reporting on these systems, I am not 
  trying to criticize nor "point a finger" at any TCWCs.     I know 
  full well how subjective and often difficult tropical cyclone 
  intensity estimation from satellite imagery is, and I also understand 
  the operational pressures and commitments which forecasters are often 
  under.    My hope is that by reporting on these interesting systems, 
  some qualified person(s), whether at one of the warning centers or 
  elsewhere, will undertake to study them more closely and perhaps come
  up with some results which will enable forecasters in the future to
  have a better handle on the character and intensity of such storm 

     For starters, it is my opinion (and also of some other qualified
  meteorologists) that part of the problem is that the Dvorak method
  of satellite intensity analysis does not handle such midget storm
  systems well.   A recent case in point---on 24 Aug a Hurricane Hunter
  reconnaissance flight investigated a tropical wave east of the
  Windward Islands.  There was a small area of convection with some hints
  of a possible circulation, but nothing that really suggested the
  presence of a tropical depression or storm.  Yet the Hunters found
  a very small, well-developed tropical storm (which was named Emily)
  only 10 kts shy of hurricane force!

     Another example: at one point in dealing with the early stages of
  Typhoon Tanya, forecasters had simultaneous Dvorak intensity estimates
  of 25 kts and 65 kts!   In general, JMA followed the more conservative
  estimates while JTWC tended to follow the higher values or else "split
  the middle."   Mark Lander's MSW estimates tend to be the most liberal.
  In addition to sending me information on the systems I have dubbed
  "Alpha", "Beta", and "Gamma", Mark provided me with information on and
  his version of Best Track files for Tanya, TD-18W, and Virgil.

     One final point--part of the reluctance of some warning centers to
  issue warnings on such systems is that their formation at higher
  latitudes may raise the question of whether or not they really should
  be considered tropical cyclones.      I have rarely attempted any
  editorializing in these summaries, but to put in my "two cents worth":
  I personally feel that, if a given disturbance has the basic
  requirements of a complete surface circulation and a 1-min MSW of
  34 kts or higher, and (1) regardless of its mode of formation lies
  over fairly warm subtropical waters (even if not quite 26 deg C),
  (2) is essentially non-frontal in character nor has any obvious
  source of significant baroclinic energy, and (3) has organized
  central (or near-central) convection which has persisted for about
  18 to 24 hours or more; then the system is most likely being
  primarily driven by the latent heat of condensation in a manner
  similar to cyclones in the deep tropics and should be considered
  a tropical cyclone.   And if a given system does not completely meet
  all three criteria, but comes rather close to doing so, then I am
  in favor of somewhat liberally "stretching" the definition of
  a tropical storm in the interest of enhancing public concern and
  also in the interest of helping to maintain consistency in historical
  statistics of tropical cyclones.  In pre-satellite days, no doubt
  there were occasionally some hybrid-type storm systems in the
  subtropics of most all oceanic basins which were treated as tropical
  cyclones and included in the seasonal statistics.  End editorial!

     The narrative presented below on the three unnamed/unnumbered
  systems which I have identified with Greek letters is taken in toto
  from information provided by Dr. Mark Landers.    Information on
  Typhoons Tanya and Virgil and TC-18W is based upon Mark's notes
  plus information gleaned from the Remarks section of JTWC warnings.

                      Tropical Cyclone "Alpha"
                            8 - 11 August

     This midget system possibly originated from a small, persistent area
  of disturbed weather in the subtropics as early as 5 Aug located near
  27N, 163E.  From 5 to 7 Aug this area of disturbed weather (associated
  with a TUTT) drifted slowly northward and by the morning of 8 Aug
  had developed a well-defined LLCC associated with a consolidated area
  of deep convection.     At 08/0000 UTC the circulation was centered
  about 750 nm north-northwest of Wake Island--far to the east of Japan.
  A small shield of persistent deep convection moved over the LLCC by
  the evening of the 8th and pushed intensity estimates to around 35 kts.
  The intensity hovered near minimal tropical storm intensity for about
  12-18 hours as the system drifted generally eastward.

     After this time the small cyclone moved eastward and weakened as
  the LLCC became exposed on the west side.  The system remained poorly
  organized for over a day until a fast re-intensification commenced
  on the night of 10 Aug.  "Alpha" began to accelerate rapidly to the
  northeast and reached an estimated peak intensity of 55 kts around
  0000 UTC on 11 Aug as it crossed the International Dateline at a
  point about 625 nm north-northwest of Midway Island.    A visible
  satellite image taken about this time appears to reveal (in the
  opinion of the author and of some others with whom I have shared the
  image) a quite well-organized tropical storm.  After peaking "Alpha"
  raced northeastward into the mid-latitudes of the North Pacific and
  weakened several hundred miles south of the southwestern Aleutian

                         Tropical Cyclone "Beta"
                               9 - 11 August

     The second midget tropical system of the Western Pacific subtropics
  was quite short-lived.  It formed along an east-west cloud band with
  TD-14W to its west and "Alpha" to its east.  At 0000 UTC on 9 Aug a
  LLCC was evident and was located almost 900 nm east-southeast of Tokyo
  and an equivalent distance east-northeast of Iwo Jima.  The midget
  system moved southeastward as central convection increased.    By
  10/0000 UTC Mark Lander estimates that minimal tropical storm intensity
  was achieved.  "Beta" reached an estimated peak MSW of about 45 kts
  by 1200 UTC on the 10th.  Early on 11 Aug (local time) the central
  convection collapsed, leaving behind a well-defined LLCC.  This center
  began to move off to the northeast and during the night of the 11th
  the remnant vortex of "Beta", located nearly 500 nm north-northwest of
  Wake Island, was swept into the primary cloud band of a rapidly  
  deepening and very large subtropical cyclone located to its northwest.

                       Tropical Cyclone "Gamma"
                            11 - 17 August

     This system formed about 200 nm east-northeast of Okinawa in a
  monsoon trough displaced to an unusually high-latitude.    It was
  preceded in this area by three other tropical cyclones--Tropical Storm
  Paul, Tropical Storm Rachel, and TD-14W.  After TD-14W crossed Japan
  and died in the Sea of Japan and the remnants of Rachel were near
  Shanghai, the monsoon trough still stretched eastward from central
  China to the waters south of Japan.  The tropical disturbance which
  became "Gamma" formed at the eastern end of the trough and moved to
  the northeast, then more to the east, for the next two days.  During
  the night of 13 Aug the small system slowed, turned to the north,
  and intensified.  By the local morning hours of 14 Aug the well-
  defined and partially exposed LLCC could be seen located about 60 nm
  south-southeast of the mouth of Tokyo Bay.   "Gamma" then turned to
  the northwest and made landfall just to the southwest of Tokyo at
  approximately 0500 UTC on 14 Aug.   The system weakened and crossed
  Japan, emerging into the Sea of Japan on 15 Aug, where it meandered
  about for 36 hours in a possible interaction with TD-15W to its
  south-southwest.   After 16/0600 UTC the weakened vortex accelerated
  to the northeast, losing its identity on the 17th west of northern

     Mark Lander had access to some synoptic reports in the Tokyo area
  during the passage of this system:  Tokyo Heliport (RJTI), Tokyo
  International Airport (RJTT), Tateyama (RJTE), and Yokota AB (RJTY).
  As the center passed nearby each of these stations on its north-
  northwest trek, the minimum sea-level pressure fell to 998 mb and
  the peak sustained 10-min wind (on-the-hour reports) reached 24 kts
  and 25 kts at RJTT and RJTI, respectively.   These stations had an
  abrupt shift of the wind from east to southeast at the time of the
  minimum pressure and the peak sustained wind.  Very heavy rain was
  experienced with "Gamma", the highest being 168 mm in 24 hours at
  Yokota AB.  The highest winds and deepest convection were on the
  north and east side of the LLCC.  Ship observations from the offshore
  waters to the east indicated east to east-southeast 10-min avg winds
  of 30 kts.   By using the most commonly used conversion factor, this
  would translate into a 1-min avg MSW of 34 kts, which would suggest
  that "Gammma" likely briefly reached minimal tropical storm intensity
  from a 1-min avg perspective.

                 Typhoon Sam/Luding  (TC-16W / STS 9910)
                             18 - 23 August

     While the subtropics were literally crawling with midget tropical
  cyclones, Typhoon Sam was a large, sprawling tropical cyclone which
  began in the monsoon trough in the deep tropics.     An area of
  convection was noted on 11 Aug far to the southeast of Guam, apparently
  associated with an easterly wave.    A Significant Tropical Weather
  Outlook (STWO) issued by JTWC on 12 Aug relocated the disturbance
  farther west, to the south of Guam.   Isolated pockets of convection
  were flaring up along the periphery of a broad LLCC with diffluence
  aloft.    The system remained quasi-stationary for two or three days,
  and a STWO on the 14th indicated that the LLCC was very weak and it
  appeared to be more like a wave in the tropical easterlies.  On 16 Aug
  the system resumed moving westward and on the 17th JTWC issued a
  Formation Alert.   Most of the convection was east of the LLCC and the
  system displayed monsoon depression characteristics with most of the
  stronger winds along the periphery.

     By 0000 UTC on 18 Aug the disturbance was sufficiently organized
  that PAGASA initiated advisories, naming the system Luding when it
  was approximately 300 nm east of Catanduanes Island in the central
  Philippines.  JTWC began writing warnings six hours later.     The
  depression at this time had a very large circulation with sporadic
  regions of deep convection.  Tropical Depression Luding began moving
  on a northwesterly course, steered by a subtropical ridge to its
  northeast.   PAGASA upgraded Luding to a tropical storm at 1800 UTC
  on 18 Aug when the storm was centered about 200 nm east-northeast of
  the aforementioned Catanduanes Island.   JTWC followed suit six hours
  later, naming the system Tropical Storm Sam.    The storm had been
  moving on a northwesterly track, but after 19/1200 UTC it turned to
  more of a west-northwesterly course, apparently due to a ridge building
  to the north.

     Tropical Storm Luding reached the extreme northeastern coast of the
  island of Luzon during the late morning (local time) of 20 Aug with
  the MSW estimated at 50 kts (both by JTWC and PAGASA).  The storm
  passed about 40 nm south of Cabo Engano, crossed the extreme northern
  end of Luzon, and exited the island just north of Laoag around 1200
  UTC.  Convection had become well-organized around the LLCC and the
  storm maintained its intensity (or even increased a little) even
  though its inflow was interrupted somewhat by a mountain range in
  northern Luzon.   After leaving Luzon behind, Sam/Luding continued
  to increase in intensity as it drew a bead on Hong Kong.  JTWC briefly
  upgraded the storm to a minimal typhoon at 21/0600 UTC but downgraded
  it back to a 60-kt tropical storm on the next warning.

     The weakening was only temporary, however, and the storm moved under
  an upper-level anticyclone.  Deep convection was present in the
  southeast quadrant and also west of the center.  JTWC reclassified
  Sam as a typhoon at 22/0000 UTC when the center was only about 50 nm
  southeast of Hong Kong.  Typhoon Sam reached its peak intensity of
  75 kts shortly before making landfall near Hong Kong.  The typhoon
  crossed right over the city and continued northwestward into southern
  China.  At 1800 UTC the cyclone, now a weakening tropical storm with
  40-kt MSW, was located very near the city of Canton.    By 0600 UTC
  on 23 Aug the system was dissipating inland in southern China.

     In northern Luzon one fatality was attributed to the storm--a
  woman drowned while crossing a swollen river on her way home to
  fetch clothes for her sick child.  Nine were reported injured and
  over 4000 people were temporarily displaced by floodwaters.  Most
  of the major roads to and from Baguio City were closed due to the
  rains and also to landslides.   (Thanks to Michael V. Padua for
  sending me this information.)

     Unfortunately, at the time Typhoon Sam was making landfall near Hong
  Kong, I was "up to my ears" just retrieving and downloading real-time
  warnings and advisories for about six simultaneous tropical cyclones,
  so I missed locating any reports of damage or casualties on the web.
  Patrick Hoareau, of Rennes, France, sent me a little bit of
  information.   One person died while attempting to windsurf, and one
  other person died due to an airliner crash at Hong Kong which was
  related to the typhoon.  Cheung Chau, just outside the harbor, reported
  a peak wind gust of 84 kts.  (Thanks to Patrick for sending me this

     If anyone has any more information on the effects of Typhoon Sam,
  please e-mail it to me and I will include it in a future summary.

                    Typhoon Tanya  (TC-19W / STS 9912)
                              18 - 25 August

     Typhoon Tanya appeared to be a TUTT cell-related development which
  occurred near the southwestern end of an old washed-out frontal
  boundary which trailed across the subtropical ridge axis.  On 18 Aug
  there was a LLCC associated with a small area of convection related
  to the TUTT cell.  The system was then located east of the Dateline
  approximately 200 nm north of Midway Island.      In Mark Lander's
  opinion, the system had already reached tropical storm intensity
  by 18/1800 UTC.   The small system initially moved southwestward
  and then swung to a westward course, crossing the Dateline about
  150 nm northwest of Midway around 0300 UTC on 19 Aug.  Environmental
  sea level pressures in the area where the storm developed were around
  1023 mb--somewhat akin to that of similar tropical storms which have
  formed in the Atlantic subtropics.  During Tanya's initial tropical
  storm stage, the diameter of the area encompassed by tropical storm
  force winds was less than 75 nm.

     JTWC issued the first depression warning at 1800 UTC on 19 Aug
  shortly after the system had entered the Northwest Pacific basin.
  (Mark's MSW estimate at this time was 60 kts.)  Six hours later the
  system was upgraded to Tropical Storm Tanya with 35-kt winds, while
  Mark's track has Tanya reaching typhoon intensity at this time.
  Dvorak current intensity estimates at this time available to JTWC
  were 25 kts and 55 kts, and at 20/0600 UTC were 25 kts and 65 kts!
  JTWC upgraded Tanya to a typhoon at 21/0000 UTC when a 15 nm-wide
  eye became visible in satellite imagery.   Dr. Lander cited a study
  by Ray Zehr which indicates that based upon Atlantic reconnaissance
  investigations, when an eye first becomes visible in a storm with
  a CDO-type pattern of convection, the average intensity (1-min MSW)
  is usually around 80 kts.

     With an anticyclone in place over the storm and steered by a 
  subtropical ridge to the north, Tanya continued to intensify.  JTWC's
  peak intensity for Tanya was 70 kts from 0000 to 0600 UTC on 22 Aug,
  while Mark Lander's peak was 90 kts from 0000 to 1200 UTC on 21 Aug.
  At Tanya's peak intensity, the diameter of the area experiencing
  tropical storm-force winds was less than 150 nm, and the maximum radius
  of 50-kt winds was only 20 nm.   The typhoon was located about 750 nm
  north of Wake Island at the time it reached its peak intensity.  By
  22/1200 UTC the convection was less organized and Tanya was beginning
  to weaken.  The storm, which had been moving west-northwestward, began
  to turn more to the northwest after 1800 UTC on 22 Aug.  Vertical shear
  associated with an upper-level anticyclone over Japan increased and
  Tanya continued to weaken.  By 23/0000 UTC the LLCC was completely
  exposed with the nearest deep convection about 20 nm to the south.

     The weakening storm recurved to the north and northeast on 23 Aug,
  passing almost 900 nm east of Japan.   Rather suprisingly, JTWC's
  intensity estimate at this time was significantly higher than Mark's.
  However, JTWC quickly weakened the storm, downgrading it to a tropical
  depression at 23/1800 UTC and writing the last warning at 24/0000 UTC, 
  whereas Mark's track maintains Tanya as a minimal tropical storm all
  the way to latitude 43.0N, 172.5E at 0000 UTC on 25 Aug, by which time
  the storm had likely become extratropical.  This position placed the
  center over 1000 nm northwest of Midway Island.    (Note: The peak
  10-min MSW assigned to Tanya by JMA was 50 kts on 21 and 22 Aug.)

                        Tropical Cyclone  (TC-18W)
                              19 - 25 August

     As Tropical Storm Tanya was gaining strength near the International
  Dateline, another small-scale tropical system was beginning to take
  shape about 750 nm east of Iwo Jima.    This system was treated as
  a 25-kt tropical depression (one warning gave the MSW as 30 kts) by
  JTWC, but Mark Lander estimates the system to have been an unnamed
  tropical storm with a peak MSW of 45 kts.  (I have no information
  from JMA on this system.)  From its beginning far to the east of Iwo
  Jima on 19 Aug, the system moved slowly in a general northwesterly
  direction.   A SWTO issued by JTWC at 19/2300 UTC mentioned an area
  of convection with a LLCC evident beneath the convection; yet no
  warnings were initiated on the system.

     Dr. Lander believes the system reached tropical storm intensity
  at around 1800 UTC on 20 Aug when it was located about 650 nm
  northeast of Iwo Jima.  Mark sent me a high-resolution visible picture
  of the system at about that time which reveals a small area of deep
  convection with considerable banding, and the LLCC appears to be under
  or very near the deepest convection.   As noted in a JTWC warning,
  the convection does seem to be elongated somewhat due to some vertical
  shearing.   JTWC initiated depression warnings at 21/1800 UTC--at the
  same time Mark Lander's track shows the cyclone to be at its peak
  intensity of 45 kts.  The system was being steered west-northwestward
  at this time by a low-level ridge to its north.      The modest
  intensification was enhanced by the storm's being located in a region
  of diffluence generated by a TUTT to the southwest and an upper-level
  anticyclone to the northeast.

     By 0000 UTC on 23 Aug the center had become fully exposed with the
  convection sheared off to the west.     The system was at this time
  located about 300 nm east-southeast of Tokyo and began to move slowly
  to the north.  Vertical shear had increased and the system quickly
  weakened.  JTWC dropped warnings at 24/0000 UTC, but Mark's track
  carries the residual depression for another 24 hours to a position
  about 600 nm east-northeast of Tokyo.

                    Hurricane Dora  (TC-07E / TS 9911)
                              6 - 23 August

     Hurricane Dora was an unusally long-lived and far-travelled North
  Pacific tropical cyclone which trekked from near 100W south of
  Acapulco, Mexico, all the way across the Northeast Pacific basin
  and entered the Northwest Pacific basin.   Along the way Dora became
  a rather severe hurricane with MSW estimated at 120 kts, but by the
  time the storm crossed the International Dateline into the Northwest
  Pacific basin on 20 Aug, winds were down to 60 kts and the cyclone
  continued to weaken and dissipate over the next three days.   The
  full history of Hurricane Dora, including the short time it was
  west of the Dateline, is described in the section of this summary
  covering the Northeast Pacific basin.

                     Typhoon Virgil  (TC-19W / TS 9913)
                          22 August - 3 September

     Typhoon Virgil was the final midget tropical cyclone to form in
  subtropical latitudes of the Northwest Pacific during the month of
  August, and perhaps became the most intense.   The storm originated
  from an isolated area of disturbed weather on 22 Aug located about
  225 nm northwest of Iwo Jima.   The system drifted south-southeastward
  and on 23 Aug acquired a well-defined partially exposed LLCC and a
  small area of persistent deep convection.      According to Mark
  Lander's track, the system had reached minimal tropical storm
  intensity by 23/0600 UTC about 150 nm northwest of Iwo Jima.   At
  about the same time JTWC issued a Formation Alert on the system.
  JTWC began issuing depression warnings at 0000 UTC on 24 Aug.

     The depression/tropical storm began to move generally in an
  east-northeastward direction which it followed for most of its life.
  The system appeared to be sheared from the east on 24 Aug, but on
  the morning of 25 Aug (local time) it acquired a CDO as the shear
  lessened.   Around 0530 UTC (afternoon locally) a small eye became
  visible in the CDO.   At 25/0600 UTC JTWC upgraded the system
  from TD-19W directly to Typhoon Virgil with 65-kt MSW.    (Mark's
  intensity estimate is similar at this time--70 kts, while JMA
  was still calling the system a depression.  JMA's 10-min MSW was
  never higher than 45 kts for Virgil.)   JTWC's peak MSW of 75 kts
  for Virgil was reached at 1800 UTC on the 25th.   Around the time
  Virgil reached typhoon intensity the radius of 35-kt winds was
  estimated to be 40 nm and radius of 50-kt winds only 20 nm.  At
  peak intensity Virgil displayed a 14-nm wide eye with a maximum
  radius of gale-force winds of 70 nm.  At the time that Virgil was
  upgraded to typhoon intensity it was located only about 100 nm 
  northeast of Iwo Jima.

     Mark Lander's analysis assigns a peak intensity of 100 kts to
  Virgil from 1200 UTC through 1800 UTC on 25 Aug.  According to Mark,
  manual Dvorak intensity estimates for Virgil reached T5.5 (102 kts),
  and a solid T5.0-T5.5 with a peak of T5.8 from 1030 UTC to 1730 UTC
  on 25 Aug using Ray Zehr's objective digital Dvorak method.  Virgil's
  eye persisted through the night, and then on 26 Aug the LLCC became
  partially exposed on the north side of the deep convection due to
  northeasterly shear from an anticyclone to the north.  By 1800 UTC
  the MSW was down to 50 kts and the LLCC was about 70 nm north of the
  deepest convection.   Dvorak current intensity estimates received
  by JTWC at 27/0000 UTC were 35 kts and 65 kts, so JTWC maintained
  the MSW at 50 kts. (Actually, Mark Lander's estimate at this time was
  lower--down to 45 kts.)

     JTWC downgraded Virgil to a tropical depression at 27/1800 UTC
  when it was located roughly 375 nm northeast of Iwo Jima, and issued
  the final warning at 29/0000 UTC.  Mark Lander's track keeps Virgil
  as a minimal tropical storm through 1200 UTC on the 29th, and then
  tracks the very weak residual depression on a slow eastward to east-
  northeastward course through 0600 UTC on 3 Sep when it was located   
  almost 1500 nm to the east of Japan.   The depression moved very
  slowly during its decaying phase as it was steered by a low-level
  ridge to the east-southeast but running counter to a low-level ridge
  to the north.

               Tropical Storm Wendy/Mameng  (TC-20W / TS 9914)
                            30 August - 4 September

     A STWO issued by JTWC on 29 Aug mentioned an area of convection
  located northwest of Yap that was associated with a rapidly moving
  easterly wave.   A broad LLCC had formed by the next day and the
  upper-level environment was favorable for strengthening.    PAGASA
  initiated warnings at 1800 UTC when the broad center was located
  about 400 nm northwest of Palau, naming the system Mameng.  Mameng
  was a classic example of a monsoon depression with two large, primary
  clusters of deep convection and with the strongest winds located on
  the periphery of the circulation.  (According to some e-mail from
  Mark Landers, one of these clusters of convection passed over Palau,
  dropping almost 100 mm of rain in less than 6 hours.)  JTWC issued
  a Formation Alert at 31/0230 UTC.  Animated visible imagery suggested
  that a separate LLCC had started to develop in association with one
  of the two main clusters of convection within the broader circulation
  of Mameng.

     The first depression warning was issued by JTWC at 0000 UTC on
  1 Sep, placing the center about 175 nm east-southeast of Catanduanes
  Island.   On 1 and 2 Sep there was considerable divergence between
  JTWC's and PAGASA's positions with the latter agency's positions
  being somewhat to the north and west of JTWC's.  Since the convection
  at this stage was somewhat "chunky" in nature, it is likely that the
  two TCWC's were focusing on different clusters of convection.  PAGASA
  upgraded Mameng to a tropical storm at 1800 UTC on 1 Sep when it was
  located about 100 nm east of Catanduanes Island.     Tropical Storm
  Mameng was in many ways a clone of Typhoon Sam/Luding two weeks
  earlier, both in its area and mode of formation and in its subsequent
  track.     The storm began to track generally in a northwestward
  direction which carried it just off the northeastern tip of Luzon
  (Cabo Engano) and through the Babuyan Islands.    At 02/0000 UTC JTWC
  relocated the center almost 200 nm farther to the north, due apparently
  to a consolidation of convection around another LLCC within the
  larger circulation.     This northward relocation brought JTWC's
  coordinates much more in line with PAGASA's, and after 02/0600 UTC
  the two TCWC's were in close agreement on the location of the center.

     JTWC upgraded the system to Tropical Storm Wendy at 02/1800 UTC
  based upon satellite imagery and synoptic reports of winds to 35 kts.
  Wendy/Mameng was located in the Luzon Strait between Taiwan and Luzon
  at this time.  An anticyclone east of the storm interacting with a
  TUTT to the northwest had caused the convection to become rather
  elongated.  Tropical Storm Wendy continued moving to the northwest
  toward the coast of China, making landfall around 2130 UTC on 3 Sep
  about 140 nm east-northeast of Hong Kong.   As Wendy approached the
  Chinese coast, a CDO feature developed near the LLCC and the storm
  intensified slightly.  The highest MSW reported were 40 kts by JTWC
  (1-min avg) and 45 kts by PAGASA (10-min avg).  The final warning
  from JTWC indicated that winds may have been briefly higher just
  prior to landfall.

     I have received no reports of damage or casualties resulting from
  Tropical Storm Wendy/Mameng.   If anyone has any information on the
  effects of the storm, please e-mail it to me and I will include
  it in a future summary.

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

  Activity for August: 1 tropical depression **

  ** - classified as a depression by IMD only

                   North Indian Ocean Activity for August

     As in July, there were no tropical cyclones of tropical storm or
  hurricane intensity in the North Indian Ocean, nor were warnings
  issued on any tropical depressions by JTWC.  There was one system
  that was carried as a tropical depression by the IMD.  This system
  formed on 6 Aug in the northern extremities of the Bay of Bengal
  about 85 nm south of Calcutta.   The weak depression subsequently
  moved west-northwestward and inland into Orissa state on the 7th.
  By 1200 UTC on 8 Aug the weakening LOW was located about 80 km
  south of Pendra.   The MSW was likely no greater than 25 kts.


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

  Activity for August:  No tropical cyclones


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

  Activity for August:  No tropical cyclones


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

  Activity for August: 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, 1998 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: []
       (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
           August as an example:   aug99.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:  aug99.sum, for

    Back issues can also be obtained from the following websites
  (courtesy of Michael Bath, Michael V. Padua, Tom Berg, Michael
  Pitt, and Rich Henning):>>>>>

     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:>


     I have discovered that JTWC now has available on its new website
  the complete Annual Tropical Cyclone Report for 1997 (1996-1997 season
  for the Southern Hemisphere).   Also, tracks only for the 1998 tropical
  cyclones are currently available.

     The URL is:>

     Also, TPC/NHC has available on its webpage nice "technicolor"
  tracking charts for the 1998 Atlantic and Eastern North Pacific
  tropical cyclones; also, preliminary storm reports for all the 1998
  Atlantic and Eastern North Pacific cyclones are now available.

     The URL is:>

  Prepared by: Gary Padgett
  E-mail:  [email protected]
  Phone:  334-222-5327 (nights & weekends) / 850-882-2594 (weekdays)


Document: summ9908.htm
Updated: 4th January 2007

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