Tropical Cyclones
[Index]
Main Index Home Page Stock Weather Photos Australian Severe Weather Forum Storm News and Storm Chasing Reports Tropical Cyclones / Hurricanes / Typhoons Weather Data and Links Wild Fires / Bushfires Weather Observation Techniques Weather Picture Catalogue Tornado Pictures and Reports Stock Video Footage and DVDs for sale
Monthly Global Tropical Cyclone Summary July 2003
[Summaries and Track Data] [Prepared by Gary Padgett]

                  MONTHLY GLOBAL TROPICAL CYCLONE SUMMARY
                      
                                JULY, 2003
                                

  (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.)

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

                              JULY HIGHLIGHTS

  --> First Atlantic hurricane of season strikes Texas
  --> Several weak Eastern Pacific tropical storms form
  --> First super typhoon in five years strikes Philippines--also strikes
      southern China
  --> Minimal typhoon affects southern Philippines, China and Vietnam

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

                ***** Feature of the Month for July *****

          FORECASTING AUGUST ATLANTIC TROPICAL CYCLONE ACTIVITY
          -----------------------------------------------------

     This monthly feature is based upon research conducted by Eric S.
  Blake, currently an employee of TPC/TAFB, and formerly a graduate
  research student at Colorado State University, Fort Collins, CO.
  A special thanks to Eric for granting permission to feature his
  work in this summary.  Much of the information was taken from the
  abstract of Eric's paper appearing in the book from the AMS 25th
  Conference on Hurricanes and Tropical Meteorology, held in San
  Diego, CA, in April-May, 2002.

     It has been 20 years since Dr. William Gray of Colorado State
  University (CSU) began his pioneering efforts in forecasting seasonal
  levels of tropical cyclone activity in the Atlantic basin.  The CSU
  seasonal forecasts show skill in the 3-10 month time frame in
  predicting overall seasonal activity.  However, significant month-to-
  month and multi-week variability exists within most seasons.  Very
  active hurricane seasons may contain quiet 2-4 week periods, while
  inactive years can exhibit short active periods.

  (NOTE:  In the discussion below I have freely used the terminology
  employed in the CSU forecast, e.g., NS, H, IHD, NTC, etc.  A definition
  of these terms can be found in last month's (June) monthly feature.
  Also, all the NTC values referenced in this article are based on the
  period 1950-2002.)

     A few examples of this intraseasonal variability follow:

  (1) 1995 - During the hyper-active 1995 Atlantic season (19 NS, 11 H,
      5 IH), there were two very active periods separated by a long
      quiet period.   Between 28 July and 29 August, nine NS formed
      (Dean through Luis), and from 27 September to 27 October, six
      NS developed (Noel through Tanya).  However, between 29 August
      and 27 September, only one cyclone developed--Marilyn.

  (2) During the inactive 1941 season (6 NS, 4 H), no tropical storms
      or hurricanes were detected prior to 11 September.  However,
      between that date and 15 October, all of the season's six NS 
      developed.

  (3) In 1976, which was a below-normal year in the Atlantic (NTC of
      81), the month of August was very active, producing five NS with
      a NTC of 59.  Prior to August, one minor storm developed, and
      following August, there were only two more NS.    That year
      experienced a very quiet September with only one NS forming
      during the month.

  (4) During the period 1950-2002, there were two Augusts with zero
      tropical cyclone activity:  1961 and 1997.    Yet the character
      of these two seasons was drastically different.  The stormless
      August of 1997 had been preceded by a very active July with four
      NS, but only three more storms followed after August, yielding a
      very low NTC for the season of 51.   However, the storm-free August
      of 1961 was followed by one of the most active Septembers on record,
      and that year's NTC of 212 is exceeded by only 1995 (220) and
      1950 (229).

  (5) The year 1966 experienced a very active early season with five NS
      forming by late July.  Yet, the month of August produced only one
      storm, Hurricane Faith, followed then by four NS in September.

  (6) The year 1981 was a very spread-out year, with tropical storm
      activity beginning with Arlene in early May and continuing through
      early November when Hurricane Katrina formed.  The seasonal NTC
      was 108, slightly above normal, but nearly 80% of this total NTC
      occurred during the month of September alone, when five hurricanes
      developed--there were hardly any days during the month in which
      there was not a tropical cyclone operating.

     Since there were no operational techniques for predicting the
  occurrence of Atlantic tropical cyclones above or below climatology
  over periods of a few weeks to a month, Eric set out a few years ago
  to see if one could be developed.    He chose the month of August
  because of its large average NTC (25%) and large year-to-year
  variability.  He has developed a forecast procedure which explains
  about 55-70 percent of the variance of August tropical cyclone activity
  one month in advance using regression techniques.  This is substantially
  better than simply climatology, or by assuming that in a given year
  August will produce 25% of the activity of the CSU seasonal forecast.

     Some of the atmospheric predictors identified by Dr. Gray in his
  seasonal forecasts were first utilized in a provisional first trial
  for forecasting August NTC.  This set failed to produce skillful fore-
  casts, and it was immediately clear that a new set of predictive
  factors would be required to produce skillful August-only forecasts.
  Eric's analysis utilized NCEP re-analysis fields for the years 1949-
  1999 to identify potential predictors.  Another group of provisional
  predictors was found by correlating tropical cyclone activity indices
  with global re-analysis atmospheric fields for 1958-1999.

     Eric developed hindcast schemes for the 1949-1999 period using the
  best predictors for each of the dependent variables.  Twelve parameters
  were determined to be related to some aspect of August tropical cyclone
  activity.   The two most-utilized predictors, appearing in every fore-
  cast equation, are the westerly and southerly July 200-mb winds over
  the equatorial Pacific just west of South America.  When this region
  experiences winds that are anomalously westerly and southerly, tropical
  cyclone activity is generally suppressed in the Atlantic basin.  This
  predictor is linked to ENSO, mid-latitude features in the Southern
  Hemisphere, and the Madden-Julian Oscillation.

     Increased August Atlantic basin activity usually follows anomalously
  low pressure in the mid-latitudes of the western Pacific Ocean and high
  pressure in the western Pacific tropics.  During the early summer before
  active Augusts, the Atlantic experiences a reverse synoptic-scale
  pattern as compared with the Pacific.   In particular, anomalously low
  pressure is noted in the tropical Atlantic with a diminished Bermuda
  HIGH and slightly increased pressures in higher latitudes.  These lower
  pressures in the Atlantic subtropical HIGH are usually accompanied by
  low vertical wind shear in the main tropical cyclone development region
  (south of 20N and east of the Lesser Antilles) with reduced trade winds
  and more easterly upper-level flow.

     The details of Eric's prediction scheme are better described with
  charts and tables which would be difficult to reproduce in a plain text
  document.  Much more information on the August-only forecast scheme, and
  its evolution since the first cut in 2000, may be found on the Colorado
  State University website at the following links:

     http://hurricane.atmos.colostate.edu/Forecasts/2000/august2000/>
     http://hurricane.atmos.colostate.edu/Forecasts/2001/aug2001/>
     http://hurricane.atmos.colostate.edu/Forecasts/2002/aug2002/>
     http://hurricane.atmos.colostate.edu/Forecasts/2003/aug2003/>

     The table below summarizes Eric's August-only forecasts from the
  first one in 2000 up through this year's forecast:


              FORECAST OF AUGUST TROPICAL CYCLONE ACTIVITY

     Year     NS     H     IH       NSD       HD       IHD      NTC
     --------------------------------------------------------------

     2000      3     2      1      14.25     8.25      1.25      33
     2001      3     1      1       7.00     2.50      0.50      22
     2002      4     1      0      10.00     4.00      0.00      18
     2003      3     1      1       8.00     4.00      0.50      22


     The following table summarizes actually observed August tropical
  activity (2000-2002 based on official Best Track, 2003 based on
  operational intensities):


                OBSERVED AUGUST TROPICAL CYCLONE ACTIVITY

     Year     NS     H     IH       NSD       HD       IHD      NTC
     --------------------------------------------------------------

     2000      4     2      1      26.25    14.00      1.00      41
     2001      3     0      0      10.25     0.00      0.00       9
     2002      3     0      0       5.00     0.00      0.00       7
     2003      3     1      1       6.00     2.25      1.25      23


     As can be seen, the results for 2003 were exceptional, and the
  initial forecast for August, 2002, was actually very good.  The
  NTC for that month was bumped up considerably by the anomalously
  long-lived Hurricane Alberto.   The forecasts for 2001 and 2002
  were not so close, but were both on the correct side of climatology,
  i.e., both called for below-normal activity, which was what was
  observed.  In fact, all four of the forecasts were on the correct
  side of climatology--an important benchmark--especially considering
  that two of the four seasons (2001 and 2003) had below-normal August
  activity while the seasonal activity was well above normal.

     Since 1950, in only 10 years has August failed to produce a
  hurricane, and 2001-2002 is the first occurrence of two consecutive
  hurricane-free Augusts.  Also, in only two prior years (since 1950)
  have there been as many as 3 NS in August with none reaching hurricane
  intensity:  1984 and 1988.

    (About three months ago Eric informed me that his paper was in the
  review process pending publication in _Weather and Forecasting_.  If
  and when the paper is published, I shall include the particulars in
  a future summary.)

  *************************************************************************
  
                            ACTIVITY BY BASINS

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

  Activity for July:  2 tropical depressions
                      2 hurricanes


                         Sources of Information
                         ----------------------

     Most of the information presented below was obtained from the
  various tropical cyclone products issued by the Tropical Prediction
  Center/National Hurricane Center (TPC/NHC) in Miami, Florida:
  discussions, public advisories, forecast/advisories, tropical weather
  outlooks, special tropical disturbance statements, etc.    Some
  additional information may have been gleaned from the monthly
  summaries prepared by the hurricane specialists and available on
  TPC/NHC's website.     All references to sustained winds imply a
  1-minute averaging period unless otherwise noted.


                    Atlantic Tropical Activity for July
                    -----------------------------------

     July of 2003 turned out to be the most active month of July in six
  years in the Atlantic basin.  Two named tropical storms developed, but
  both reached hurricane intensity, and one was rather long-lived for a
  July cyclone, lasting almost 8 days as a tropical storm or hurricane.
  Claudette moved on a rather long trajectory from the central Caribbean
  where it developed until its final landfall on the middle Texas coast,
  reaching hurricane intensity twice--first very briefly in the north-
  western Caribbean, and then again shortly before making landfall in
  Texas.  Danny formed at a fairly high latitude from the northern
  extension of a tropical wave and reached minimal hurricane status for
  24 hours after it had recurved well to the south of Newfoundland.
  The average Net Tropical Cyclone activity parameter (NTC) for July has
  averaged 3.6% over the period 1950-2002.  For this month the NTC was
  14.2%--well above the average.  (This includes 0.25 NSD contributed
  by Tropical Storm Bill on 1 July.)

     As the month of July opened, Bill was a weakening tropical storm 
  inland over Louisiana.  Bill was downgraded to a tropical depression
  at 0600 UTC on 1 July, and the remnants tracked northeastward across
  the Southeast to the Mid-Atlantic coast, first as a tropical depression,
  then as an extratropical LOW.  During the early morning hours of 4 July,
  three very small low-level circulations formed along a stationary front
  extending between Bermuda and the Azores.  None of the systems possessed
  organized convection, and strong upper-level winds and cool SSTs
  suppressed development.  Nonetheless, the three little "neutercanes"
  made an interesting sight in visible imagery on the 4th of July.

     In addition to Claudette and Danny, two tropical depressions formed
  but failed to reach tropical storm status.  Tropical Depression 06 formed
  on 19 July approximately 950 nm east of Barbados.   Initially, the
  depression was considered a good candidate for intensification, but as
  it continued rapidly westward, environmental conditions became less
  favorable (increasing shear and dry Saharan air) and the system was
  unable to develop.  A QuikScat pass on the morning of the 21st reported
  winds near 45 kts, but a reconnaissance plane investigating the
  depression that afternoon could not find a closed surface circulation, so
  the system was downgraded to a tropical wave about 60 nm north-northwest
  of Barbados.  This tropical depression, like Claudette, was hampered in
  its efforts to develop a closed surface circulation due to its rapid
  translational speed.  Whereas the pre-Claudette system persevered and
  finally developed, this system was unable to do so in the face of other
  adverse factors.

     Tropical Depression 07 was a small circulation which formed on the
  morning of 25 July only about 50 nm north-northeast of Cape Canaveral,
  Florida.  This depression's roots more than likely stemmed from the
  tropical wave which had formerly been Tropical Depression 06.  After
  the demise of the earlier depression, the northern end of the wave
  continued moving rapidly west-northwestward across the eastern Bahamas
  and Cuba and moved into a small area of upper-level diffluence between
  a couple of TUTT cells.  On 24 July a large mesoscale convective system
  blew up southeast of Miami and persisted for several hours, exhibiting a
  circular shape and some evidence of symmetrical outflow.  After the
  system dissipated in the afternoon, a mesoscale vortex was clearly
  evident in the cloud debris.  This shallow system initially had no
  associated deep convection as it began drifting northward in response to
  the deep-layer mean steering.

     By the morning of the 25th the system had developed enough deep
  convection to warrant its being classified as a tropical depression.
  The initial MSW was set at 25 kts, and an afternoon reconnaissance flight
  confirmed that the winds likely did not exceed that value.    The
  depression was never forecast to reach tropical storm intensity, and that
  prediction verified.  Winds were estimated to have briefly reached 30 kts
  at 26/0000 UTC over a very small area northeast of the center, based on
  WSR-88D radar images which showed a banding feature becoming better
  organized.  By 0900 UTC on the 26th the depression's center had moved
  inland about 40 km southwest of Savannah, Georgia.   The weak cyclone
  continued moving farther inland and had dissipated into a trough of
  low pressure over central Georgia by the morning of the 27th.  The peak
  48-hour rainfall amounts reported in the HPC advisories were 45 mm and
  43 mm from Charleston and Orangeburg, South Carolina, respectively.

     The report on Hurricane Danny was written by Kevin Boyle, and the
  portions of the report on Hurricane Claudette detailing meteorological
  observations and damage and casualties were based largely upon infor-
  mation sent to the author by John Wallace.  A special thanks to John 
  and Kevin for their assistance.


     
                           HURRICANE CLAUDETTE
                                 (TC-04)
                               8 - 17 July
                 ---------------------------------------

  A. Storm Origins
  ----------------

     A tropical wave with an associated 1013-mb LOW moved off the west
  African coast on 1 July.  The wave was rather well-defined with some
  low-level rotation, but convection weakened during the day due to a
  fairly stable airmass over the Atlantic.   The next day the wave
  appeared even less impressive, becoming elongated southwest/northeast,
  and the convection weakened even more.   Over the next few days the
  system moved westward with little convection--what convection remained
  was located to the south along the ITCZ.  By 5 July the wave had reached 
  the vicinity of 46W, and while still embedded in the ITCZ, had the 
  clearest signature of the waves currently being tracked.  The Tropical 
  Weather Outlook issued by TPC/NHC at 1530 UTC mentioned the wave, but no
  significant development was expected at the time.

     The Tropical Weather Outlook at 0930 UTC on the 6th mentioned the
  potential for some slow development over the next couple of days.  The
  wave was still embedded in the ITCZ, but a low-level cyclonic rotation
  was noted near 10N.  By 1800 UTC a 1008-mb LOW was located near 10.5N,
  52.5W, or approximately 525 nm east of the Windward Islands.  The cloud
  pattern was well-organized with a curved band feature wrapping around
  the northern semicircle, but deep convection was limited.  A Special
  Tropical Disturbance Statement was issued at 1745 UTC, indicating the
  possibility of tropical cyclone development.  Dry mid-level air to the
  west was considered a possible inhibiting factor, but southerly inflow
  of moist, unstable air to the east was a favorable factor.

     By the morning of 7 July the disturbance was located about 260 nm
  east of the Windwards, moving westward at 17 kts.   The appearance was
  somewhat ragged in satellite imagery with convection less concentrated
  about the LLCC.  There was no conclusive evidence that the system had
  a closed surface circulation; nonetheless, the Tropical Weather Outlook
  indicated that the islands were likely to see squalls to tropical storm
  force as the wave passed through.   A reconnaissance plane reached the
  area around 1800 UTC and confirmed that there was yet no closed low-
  level circulation.  The lowest pressure found was 1009 mb, and the
  wave axis had by this time already passed through the Lesser Antilles.
  Wind gusts to 45 kts were reported on St. Lucia during the afternoon
  as the wave passed over, moving at speeds of up to 20 kts into the
  eastern Caribbean Sea.

     Satellite imagery on the morning of the 8th indicated that the
  convection had become more concentrated and organized and that cloud
  tops had cooled.   Visible imagery revealed increased banding as the
  system continued racing westward at 26 kts.     A reconnaissance plane
  arrived in the area during the morning, but as the 1500 UTC advisory hour
  approached, no closed surface circulation had yet been found.  However,
  winds on the poleward side of the system were above tropical storm
  strength.  Regular gale warnings were issued for portions of the Greater
  Antilles likely to be affected by the system.  Thus, this wave had
  become one of those problematic, but fortunately rare, systems which
  can produce winds exceeding gale force but without a closed circulation
  at the surface.  The system had the appearance in satellite imagery of
  a tropical storm, and no doubt had a well-developed mid-level
  circulation, but did not (or so was thought) meet the basic requirement
  for a tropical cyclone of having a ground-relative low-level west wind.

     Finally, after searching for nearly seven hours, the reconnaissance
  plane at 1744 UTC found a south-southwest wind of 21 kts along with peak
  flight-level winds (FLW) of 57 kts and a minimum pressure of 1006 mb.
  Therefore, a Special Tropical Disturbance Statement was issued at 1810
  UTC upgrading the system directly to Tropical Storm Claudette with
  40-kt winds.  (More discussion on the subject of these gale-bearing
  tropical waves can be found at the end of this report in Section E.)
  The first regular advisory on Claudette, at 08/2100 UTC, placed the
  center of the new storm about 365 nm east-southeast of Kingston, Jamaica,
  whizzing along to the west at 25 kts south of a strong subtropical ridge.
  The MSW was increased slightly to 45 kts.


  B. Storm History
  ----------------

     Claudette's history over the next few days was characterized by
  recurring cycles of intensification followed by weakening, followed by
  another round of intensification, etc.  The storm's rapid motion carried
  it into close proximity to an upper-level LOW to the west which induced
  unfavorable shear over the cyclone.   Initially, the upper-level LOW was
  forecast to weaken, and it was expected that Claudette would reach the
  Yucatan Peninsula as a hurricane, but this scenario did not materialize.
  The reconnaissance flight around 09/0000 UTC found a peak FLW of 72 kts
  with a CP of 1000 mb, and a later mission into the storm around 0400 UTC
  found a peak FLW of 85 kts northeast of the center.     However, a
  subsequent flight found only 55 kts, and also reported that the pressure
  had risen to 1004 mb.  Claudette's intensity was increased to 60 kts at
  0900 UTC when the center was located about 175 nm south-southeast of
  Kingston, but was lowered to 55 kts at 1500 UTC.  Southwesterly shear had
  caused the center to become intermittently exposed on the edge of the
  convection.

     At 09/2100 UTC Claudette was located about 220 nm south-southeast of
  Grand Cayman and moving westward at a slightly slower speed of 21 kts.
  The cyclone was producing bursts of convection with the LLCC moving 
  underneath the convection during the bursts, but subsequently becoming 
  exposed as the convection was sheared away.     A reconnaissance plane 
  found peak FLW of 65 kts on two passes with the CP around 1001-1002 mb.
  The MSW was decreased to 50 kts at 10/0300 UTC, and at 0900 UTC Claudette
  was located 330 nm southeast of Cozumel, Mexico, moving west-
  northwestward at 15 kts.    An earlier reconnaissance flight had found a
  peak FLW of 57 kts at 850 mb northeast of the center with a CP of 996 mb.
  In the meantime a large, cold, distorted CDO-feature had developed over
  the cyclone.  Vertical shear was in the 20-25 kt range and was expected
  to continue for the next several days.

     By 1500 UTC Claudette's intensity had made a turnaround.  A
  reconnaissance plane at 1201 UTC reported a 10-nm diameter eye with a
  CP of 988 mb and a peak FLW of 69 kts at 700 mb.  The plane reported a
  wind of 78 kts at 850 mb, but this was considered somewhat suspect.
  The MSW was raised back to 60 kts at 1500 UTC, although on the second
  pass through the storm the reconnaissance plane reported that the CP had
  risen back to 993 mb.   At 2100 UTC Tropical Storm Claudette was located
  about 185 nm southeast of Cozumel and moving northwestward at 12 kts.
  After its early morning intensification spurt, the storm had become less
  organized and the winds were lowered back to 50 kts.  The discussion
  bulletin, however, noted that Claudette may have reached hurricane
  intensity around 1530 UTC--a reconnaissance plane reported a peak FLW of
  76 kts and an eyewall dropsonde indicated sustained winds of hurricane
  force at the surface.  Two hours later the plane found a fragmented wind-
  field near the decayed eyewall.  It should be noted that, based on infor-
  mation in the monthly summary for July prepared by TPC/NHC, it has been
  determined that Claudette did indeed briefly reach hurricane intensity
  on the morning of 10 July.

     The MSW was reduced to 45 kts at 0300 UTC on 11 July as Claudette
  did not have a well-defined center.  The storm was just north of Cozumel
  at 0900 UTC, moving northwestward at 12 kts.  The 0600 UTC reconnaissance
  mission found a peak FLW of only 45 kts, but ship ELWX5 reported 50-kt
  winds northeast of the center, so the MSW was raised back to 50 kts.
  At 1500 UTC Claudette's center was over the northeastern tip of the
  Yucatan Peninsula.  The center was located to the southwest of a broadly
  curved band of deep convection in which an earlier reconnaissance plane 
  had found winds to 55 kts.  By 11/2100 UTC Claudette had moved out over 
  the Gulf of Mexico about 130 nm northwest of Cancun.  The cyclone was 
  quite disorganized--the LLCC had become elongated east-northeast/
  west-southwest, and several smaller cloud swirls could be seen rotating 
  around the mean center.  The MSW was lowered back to 45 kts once more.

     The 12th of July was an uneventful day in the life of Claudette.  The
  storm moved slowly northwestward to north-northwestward in the Gulf of
  Mexico with little change in intensity, the MSW remaining at 45 kts all
  day.  At 2100 UTC the center was located approximately 350 nm east-
  southeast of Brownsville, Texas.   Peak winds at flight level reported
  by reconnaissance missions were near 55 kts most of the day.  The center
  of the storm took a slight jog to the north, but then moved back to the
  southwest early on the 13th, becoming more or less stationary about
  270 nm east-southeast of Brownsville.  During the morning of 13 July the
  center was pulled into the deep convection once again, but this time
  there was a response in the pressure field.  The CP dropped as low as
  994 mb before rising to 997 mb during the afternoon as the convection
  was once more sheared away from the center.    An early morning
  reconnaissance flight found a peak FLW of 60 kts, so the MSW was upped 
  to 50 kts at 1500 UTC.  A reconnaissance plane around midday found peak
  winds of 69 kts at flight level.

     Another reconnaissance plane during the evening hours found that the
  LLCC had become well-established in all quadrants with the center
  embedded in deep convection.  A CP of 991 mb and a peak FLW of 68 kts
  were measured; therefore, the MSW was increased to 55 kts at 14/0300 UTC.
  The storm was still quasi-stationary at this time, but by 14/0900 UTC had
  begun a slow northwesterly drift at 5 kts.  The center was then fixed
  about 270 nm east-southeast of Corpus Christi.   A reconnaissance plane
  around 1200 UTC reported winds at flight level of 65-70 kts, and the 
  crew noted that an open eyewall was present.  At 2100 UTC Claudette's
  center was located approximately 220 nm east of Corpus Christi, moving
  north-northwestward at 6 kts.  The MSW remained at 55 kts with the CP
  having fallen slightly to 989 mb.  The storm was still fighting shear
  with the eye less defined than previously; however, several oil rigs 
  were reporting hurricane-force winds several hundred feet above the 
  surface.

     The 15/0300 UTC advisory upped the winds to 60 kts--Claudette was
  almost a hurricane.   A reconnaissance plane had recently reported a
  FLW of 77 kts in the northwest quadrant, and data from the NOAA Gulf-
  stream jet indicated that shear was decreasing.  At 15/0500 UTC an
  intermediate advisory was issued upgrading Claudette to a hurricane.
  The cyclone was then centered 150 nm east of Corpus Christi, moving
  west at 7 kts.   A reconnaissance flight around 0600 UTC recorded winds
  of 84 kts in the northeast quadrant, and a dropsonde reported a pressure
  of 982 mb, although the dropsonde missed the center of the eye.
  Claudette's 25-nm eye was by now visible on coastal radars.  At 1300 UTC
  a NOAA buoy (WMO 42019), located a short distance south of the eye,
  reported sustained winds to 47 kts (8-min avg), gusting to 60 kts.

     At 1500 UTC Claudette's eye was closing in on the Texas coast, being
  centered only about 17 nm east of Port O'Connor.  A reconnaissance plane
  had reported finding winds of 80-85 kts in the northeastern eyewall, and
  the CP had fallen to 981 mb.  WSR-88D data from Houston indicated winds
  near 90 kts.  Based on this, the MSW was upped to 70 kts at 1500 UTC,
  and the advisory noted that winds could reach 75-80 kts before Claudette
  began to weaken.  A late-received eye dropsonde report indicated that the
  pressure had fallen to 979 mb.  The monthly report for July on TPC/NHC's
  website indicates that Claudette's official intensity at landfall will
  be increased to 80 kts.  There was an unofficial report of sustained
  winds to 82 kts, gusting to 90 kts, at Point Comfort, Texas.

     Claudette continued moving inland across southern Texas and began to
  slowly weaken.  At 2100 UTC the cyclone was downgraded to a 60-kt
  tropical storm located about 135 km southeast of San Antonio.  By 0300
  UTC on 16 July Claudette was located 110 km south-southwest of San
  Antonio and NHC issued their final advisory, estimating the MSW still at
  45 kts.  With the primary threat from Claudette now inland flooding,
  warning responsibility for the weakening cyclone was turned over to the
  Hydrometeorological Prediction Center at Camp Springs, Maryland.   The
  first HPC advisory downgraded Claudette to a 25-kt depression near the
  Rio Grande northwest of Laredo.  The system maintained a fairly impres-
  sive satellite signature for another day or so as it continued moving
  generally up the Rio Grande valley.  The final HPC advisory, issued at
  0900 UTC on 17 July, placed the weakening circulation in northern Mexico
  south of El Paso, Texas.


  C. Meteorological Observations
  ------------------------------

  (1) Wind Reports
  ----------------

     As noted above, NOAA buoy (WMO 42019) reported an 8-min avg sustained
  wind of 47 kts, gusting to 60 kts, while the eye was passing a short
  distance to the north.  The Port Aransas CMAN reported a MSW of 33 kts
  (presumably a 1-min avg) with a gust to 38 kts at 15/2000 UTC.  Victoria
  reported a MSW of 45 kts with a peak gust of 59 kts at 1731 UTC, along
  with a minimum SLP of 996 mb.  Contact with this station, however, was
  lost shortly thereafter for the duration of the storm.

     A sustained wind of 68 kts, gusting to 83 kts, was recorded at a
  Dow Chemical plant in Long Mott, Calhoun County, around 1700 UTC.  Also,
  an unofficial gust of 77 kts was recorded at a power plant in Matagorda
  County.    Wind gusts to 73 kts were recorded at Wadsworth and Port
  Lavaca, where a tornado reportedly made a brief touchdown, causing
  some damage.  An unofficial gust of 70 kts was reported at Goliad,
  Texas, about 30 km south-southwest of Victoria.  Finally, as noted
  above, there was an unofficial report of 82-kt sustained winds, gusting
  to 90 kts, at Point Comfort.

  (2) Rainfall Reports
  --------------------

     Rainfall totals from Claudette, as reported in advisories issued by
  HPC, were rather light as far as tropical cyclone rainfall goes.  Some
  selected 6-hourly totals include:

     Location              Amount               Date and Time Period
     ------------------------------------------------------------------

     Corpus Christi         36 mm               16 July - 0000-0600 UTC
     Cotulla                37 mm               16 July - 0600-1200 UTC
     Angleton/Lake Jackson  39 mm               16 July - 1200-1800 UTC
     Houston/Hull Field     33 mm               16 July - 1200-1800 UTC

     A 12-hourly total of 26 mm was recorded at the Terrell County Air-
  port between 16/1800 UTC and 17/0600 UTC.  (Terrell County is located
  along the Mexican border approximately 100-150 km northwest of the
  city of Del Rio, Texas.)


  D. Damage and Casualties
  ------------------------

     The monthly summary for July on TPC/NHC's website indicates that
  three persons lost their lives during Claudette, either directly or
  indirectly.  According to information sent to the author by John
  Wallace, around 72,000 persons were without power during the storm.
  About 70% of the businesses in Calhoun County sustained damage, and
  75% of those in Port Lavaca were damaged.   In Victoria County the
  damage has been assessed at $10.6 million.  The Victoria Airport
  sustained damage of around $1.8 million while losses in the city of
  Victoria have been estimated at $3.5 million, mostly to private
  property.  A very preliminary estimate places the total damage at
  $17 million.  This is likely to rise as it does not include flood
  damage or any possibly significant damage to crops.  Fifteen Texas
  counties were declared federal disaster areas in the aftermath of
  Hurricane Claudette.

     In addition, the NHC report mentions that some minor damage was
  incurred on the island of St. Lucia due to gusty winds and locally
  heavy rainfall as the pre-Claudette wave passed through the Windwards.


  E. Discussion
  -------------

     As noted earlier, the pre-Claudette tropical wave was one of those
  problematic waves with tropical storm-force winds but without a
  closed surface circulation--a basic requirement for a tropical cyclone.
  The problem does not lie with nature in producing such systems, but
  with our man-made classification and nomenclature system.   By
  definition (and I'm using U. S. terminology here), the weakest type of
  tropical cyclone is a tropical depression, which must have maximum
  sustained winds less than 34 kts.   And for decades the U. S. public
  has been informed that a tropical disturbance (or tropical wave) is
  weaker than a tropical depression.  As an example, the final public
  advisory on Tropical Depression 06, issued at 2100 UTC on 21 July,
  contains as the header line: "depression weakens to a tropical wave."

     In June I sent out a tropical cyclone survey to a few dozen persons.
  The survey treated several tropical cyclone classification issues, and
  one of those was this issue of how to best handle these strong, rapidly-
  moving tropical waves with tropical storm-force winds.  A summary of
  the responses will appear in a monthly feature in a few months, so the
  comments here will be brief.  One distinction which should be made is
  that between simply gale-bearing tropical waves with little evidence of
  a storm-relative circulation, and those (such as pre-Claudette) which
  appear to have a definite circulation relative to the "center", but
  which doesn't extend down to the surface sufficiently to produce
  westerly winds relative to a ground-based observer on the equatorward 
  side of the wave.

     The responses to the above-mentioned survey tended to be in favor
  of "bending" the rules for tropical storm classification in regard to
  such systems, although there were some voices adamantly opposed to
  this.  Most of those in favor of naming such systems as tropical storms
  base this on the argument that whenever a system is named, the attention 
  of the public and mariners jumps up by an order of magnitude (something
  which I've found to be true in my experience).   On the other hand,
  there are those who point out that weather systems other than tropical
  storms can produce gale-force winds in the deep tropics and that
  ordinary gale warnings should be sufficient.

     The following quote from an e-mail written by Kenneth J. Schaudt
  of Marathon Oil Company illustrates how the attention of the public
  and other concerned parties spikes up when a tropical storm is named:
  "Much of the United States' natural gas is produced in the Gulf of
  Mexico.  Since the threat of tropical storms may shut down production
  briefly, the market for natural gas responds to the perceived threats.
  On the 8th (of July) at 1806 UTC, the Special Tropical Disturbance
  Statement declaring Tropical Storm Claudette hit the wire.  Within
  three minutes, the price of natural gas had jumped 10 cents."

     As I've already stated, I'll wait until I prepare a monthly feature
  on this topic before going into more details.  If anyone who did not
  receive the aforementioned survey would like to state his/her opinion
  on this topic, please e-mail your comments to me and I'll include them
  in the future monthly feature.

  (Report written by Gary Padgett with contributions by John Wallace)



                           HURRICANE DANNY
                               (TC-05)
                             16 - 23 July
                 -----------------------------------

  A. Storm Origins
  ----------------

     The origins of Hurricane Danny stemmed from the northern part of a
  rather large tropical wave that left the African coast on 10 July.  NHC
  included this large tropical wave (devoid of thunderstorm activity) in
  their Tropical Weather Outlook as early as 13 July (issued at 2130 UTC),
  stating that there was some potential for development over the next few
  days as the system moved westward over warmer waters.  The wave moved
  uneventfully westward across the tropical Atlantic during the 14th.  A
  NCEP/MPC 0000 UTC preliminary surface analysis chart on 15 July showed
  a trough splitting from the northern portion of the wave, which, by that
  time, lay along 49W.  An increase in shower/thunderstorm production was
  also noted in satellite images.  By early on the 16th a LOW had formed
  in association with the area of disturbed weather, which had drifted
  northwest to a position located roughly 700 nm east-southeast of Bermuda.
  Satellite images depicted the weak LLCC located southwest of the main
  area of deep convection induced by southwesterly vertical shear.
  However, they also revealed an organizing system that was moving north-
  westward at 17 kts toward a progressively more favourable upper-level
  environment.  Based on satellite images and ship reports, NHC issued the
  first advisory on Tropical Depression Five at 1500 UTC on 16 July,
  located near 31.5N, 54.5W, or about 525 nm east of Bermuda.  The initial
  intensity was set at 25 kts and some additional strengthening was fore-
  cast for the following two to three days as the system traversed the
  warm Atlantic waters that lay ahead of the depression.


  B. Storm History
  ----------------

     Overall organization of Tropical Depression Five continued to improve
  with deep central core convection developing and banding evident in the
  northern semicircle.  Outflow was good to the northwest and excellent to
  the northeast.  The only negative influence was the presence of south-
  easterly shear which tilted the vertical profile of the storm slightly
  towards the northwest.  By 17/0300 UTC satellite CI estimates had reached
  35 kts and the depression was upgraded to Tropical Storm Danny when
  located approximately 475 nm east of Bermuda.   This was increased to
  45 kts six hours later based on the appearance of a 10-nm diameter eye
  feature as indicated by a 17/0339 UTC TRMM overhead pass.   The next
  six-hour increment saw another 10 kts added, bringing the MSW up to
  55 kts.  After that, intensification eased off but the cyclone remained
  impressive enough with good outflow and two main hooking bands.  Danny
  was moving on a northward path at around 10 kts at 17/2100 UTC.

     A little more strengthening was expected before Danny ran out of warm
  water, but judging by 18/0000 UTC satellite imagery, it appeared that the
  cyclone was not going to reach hurricane strength.    The images showed
  that banding was minimal, and most of the convection was limited to a
  burst near the centre caused by westerly shearing.  Additionally, Dvorak
  numbers were falling.  Visible satellite images and microwave data at
  18/1500 UTC, however, contradicted this fact and showed that Danny (now
  moving northeastward at 14 kts) was nearing hurricane strength.  Well-
  defined curved bands were noted wrapping around a small CDO with
  excellent outflow aloft, although convective tops were not very cold.
  Satellite CI estimates from TAFB, AFWA, and SAB had increased to 60 kts
  by this time.   Just as Danny had begun to trek northeastward over colder
  waters at 18/2100 UTC, the MSW was raised to 65 kts based on satellite
  microwave data showing a definite eye.    Danny had become the second
  hurricane of the 2003 season while located about 450 nm south of
  St. John's, Newfoundland.
 
     Hurricane Danny held its MSW of 65 kts for much of the 19th as it
  moved on a northeasterly heading at 16 kts over increasingly cooler
  waters.    (NOTE: Normally a 65-kt Atlantic hurricane has a CP near
  987 mb.  However, because Danny was located in a high surface pressure
  environment all its life, the CP was likely much higher than 987 mb.
  Unfortunately, there was no data to determine the exact pressure, so
  1005 mb was used as the minimum CP with a greater than normal uncertainty
  about the value.)  Weakening began at 19/2100 UTC as the MSW dropped
  below hurricane intensity.  Danny had curved onto an easterly path by
  this time, and the main emphasis was on what could happen to the
  tropical cyclone in the days to come.
 
     Danny's track was completely influenced by a large, semi-permanent
  subtropical HIGH.  During its life it had travelled in a huge clockwise
  loop around this HIGH and was forecast to sink to the south on the
  eastern side of the anticyclone as a remnant LOW, with the possibility
  of regeneration over increasingly warmer waters.  Its track was somewhat
  reminiscent of the long-lived Hurricane Alberto in August, 2000.  
  However, Danny was not destined to emulate the long, meandering exploits
  of Alberto nor that of Hurricane Kyle (2002).

     A dramatic deterioration occurred early on the 20th as the deep
  convection was ripped off the LLCC by northwesterly vertical shear,
  leaving the centre completely exposed.   Even though CI estimates were
  between 55-65 kts from all three agencies, the MSW was dropped to 50 kts
  at 20/0300 UTC.  For the rest of the 20th it was a case of winding the
  storm down:  first to 45 kts at 20/0900 UTC (based on QuikScat data and
  a buoy MSL pressure report of 1008.0 mb at 0600 UTC), followed by a
  further drop to 35 kts six hours later.   The weakening tropical storm
  began to move more south of east around this time.  The MSW fell below
  the tropical storm threshold of 34 kts at 20/2100 UTC, which was the time
  of the final advisory issued by NHC.   At the time Danny was centred
  approximately 360 nm west-northwest of Flores in the Azores.  Little
  convection was associated with Danny and it was far from the exposed
  LLCC.

     NHC continued to monitor Danny's remnants in their Tropical Weather
  Outlooks until 24 July.  The remnant LOW was tracked south-southeastward,
  then southward during the 21st before turning southwest the next day.
  The LOW was centred about 420 nm southwest of the northern Azores at
  22/1530 UTC.  Associated convective activity was sporadic during this
  period.  Having almost completed a full circuit around the HIGH, Danny
  finally gave up the ghost when it was last mentioned in the 24/2130 UTC
  Tropical Weather Outlook.  Its final reference point was roughly 435 nm
  west-southwest of Flores in the Azores.


  C. Damage and Casualties
  ------------------------

     Hurricane Danny spent its entire life over the subtropical Atlantic
  Ocean.  No damage or casualties have been reported.

  (Report written by Kevin Boyle)

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

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

  Activity for July:  3 tropical storms


                         Sources of Information
                         ----------------------

     Most of the information presented below was obtained from the
  various tropical cyclone products issued by the Tropical Prediction
  Center/National Hurricane Center (TPC/NHC) in Miami, Florida (or the
  Central Pacific Hurricane Center (CPHC) in Honolulu, Hawaii, for
  locations west of longitude 140W):  discussions, public advisories,
  forecast/advisories, tropical weather outlooks, special tropical
  disturbance statements, etc.  Some additional information may have
  been gleaned from the monthly summaries prepared by the hurricane
  specialists and available on TPC/NHC's website.  All references to
  sustained winds imply a 1-minute averaging period unless otherwise
  noted.


              Northeast Pacific Tropical Activity for July
              --------------------------------------------

     Over the period 1971-2002 the month of July in the Northeast Pacific
  basin has averaged about 4 named storms, 2 hurricanes and 1 intense (or
  major) hurricane, and has produced an average NTC of 23.5%.  July of 2003
  turned out to be one of the most inactive Julys on record.  Three rather
  short-lived tropical storms developed, and none reached hurricane
  intensity.  The total NTC of 4.2% was the lowest since 1977.  Dolores
  was a tropical storm for only 12 hours, while Enrique and Felicia were
  both at tropical storm intensity for about two days each.   None of the
  cyclones affected the Mexican coastline, nor did any enter the Central
  Pacific (west of 140W).



                         TROPICAL STORM DOLORES
                                (TC-04E)
                               6 - 8 July
               ------------------------------------------

     The precursor of Tropical Storm Dolores was a tropical wave which
  moved off the African coast on 20 June and moved westward across the
  Atlantic with little associated convection.  On 25 June the system
  entered the eastern Caribbean Sea, bringing widely scattered showers
  and thunderstorms to the Windward Islands.  The environment was hostile
  for tropical cyclogenesis with an upper-level trough and moderate
  subsidence over the region.    The tropical wave entered the Eastern
  Pacific on 30 June, and by 1 July an area of disturbed weather had formed
  about 435 nm south-southwest of Acapulco.  A Tropical Weather Outlook
  issued by TPC/NHC at 01/2300 UTC noted that upper-level winds were
  marginally favorable for development.  The disturbance moved westward
  over the next few days, very gradually increasing in organization.
  By the morning of 5 July the system had reached a point approximately
  700 nm south-southwest of Cabo San Lucas, and the 1700 UTC Tropical
  Weather Outlook indicated that a tropical depression could form during
  the afternoon or night.

     By 0600 on 6 July an impressive CDO feature had developed.  With
  Dvorak T-numbers of 2.5/2.0/2.0 from TAFB, SAB, and AFWA, respectively,
  advisories were initiated on Tropical Depression 04E at 06/0900 UTC,
  located roughly 400 nm southwest of Socorro Island.   The depression
  moved on a general northwesterly track, reaching tropical storm intensity
  around 1800 UTC when located about 450 nm west-southwest of Socorro.
  The system was upgraded to Tropical Storm Dolores with 35-kt winds at
  2100 UTC.  From the outset Dolores did not have great prospects for a
  long life--it was battling northeasterly shear and was forecast to move
  over sub-25 C SSTs after about 48 hours.  After only twelve hours as a
  tropical storm, Dolores was downgraded to a depression at 0900 UTC on
  7 July.  The LLCC was exposed and displaced to the east-northeast of
  decreasing deep convection.  By 1500 UTC the weakening cyclone was over
  sub-25 C water and devoid of any deep convection.  A 07/1358 UTC Quik-
  Scat pass indicated winds of only 20-25 kts.

     The final advisory was issued at 0300 UTC on 8 July, placing the
  center approximately 750 nm west-southwest of Cabo San Lucas.  A few
  cells of deep convection had redeveloped to the southwest of the LLCC,
  but Dolores was in the process of degenerating into a remnant LOW.

     No damage or casualties are known to have resulted from Tropical
  Storm Dolores.

  (Report written by Gary Padgett)



                         TROPICAL STORM ENRIQUE
                                 (TC-05E)
                               10 - 13 July
               ------------------------------------------

     Like its predecessor Tropical Storm Dolores, Tropical Storm Enrique's
  origins can be traced back to a tropical wave which emanated out of the
  African continent.  On 25 June a tropical wave moved off the west coast
  of Africa and embarked on its trek across the Atlantic.  The wave had
  little associated convection and was often difficult to keep track of in
  satellite imagery.   By 28 June the wave had developed an associated
  1012-mb LOW near 7N, and was improving in organization with the formation
  of curved banding features and plenty of convection.  The LOW was rather
  far south for tropical cyclone formation, but QuikScat data confirmed the
  existence of a surface circulation.   Conditions were marginally
  favorable for some development, but this did not materialize.  On 30 June
  visible satellite imagery indicated a small-scale LLCC embedded in the
  ITCZ, but convection was virtually non-existent.

     The wave continued westward across the southern Caribbean/northern
  South America area and had entered the Eastern Pacific by 6 July.   At
  07/0000 UTC a broad, diffuse cyclonic circulation was located southwest
  of Costa Rica but with no significant convection.  At 0600 UTC the
  system was repositioned westward to 89W based on a well-defined cloud
  pattern and low-level turning in satellite imagery.  A Tropical Weather
  Outlook issued by TPC/NHC at 1100 UTC indicated that an area of disturbed
  weather about 220 nm south of the Guatemalan coast had some potential for
  development over the next few days.   By the afternoon of 8 July the
  disturbance was located about 325 nm south of Acapulco.    Ship H3KF
  reported winds of 25 kts and a SLP of 1009.5 mb at 09/0000 UTC in the
  vicinity of the disturbance.  By the morning of 9 July the system had
  reached a location about 250 nm southwest of Acapulco, and the 1700 UTC
  Tropical Weather Outlook indicated that it had become better organized
  with environmental conditions favoring further development.

     Visible satellite imagery on the morning of 10 July revealed a well-
  defined circulation, although it was exposed and well-removed from the
  deep convection.  Nonetheless, there was enough organization to warrant
  upgrading to Tropical Depression 05E.  The first advisory at 10/1500 UTC
  placed the center roughly 350 nm south-southwest of Manzanillo, Mexico.
  The depression was battling strong northeasterly shear, and was initially
  forecast not to strengthen above 40 kts before encountering cooler SSTs.
  Some consolidation of the cyclone was noted during the evening as a few
  bursts of deep convection fired off near the LLCC and convective banding
  observed earlier well-removed from the center diminished.

     During the night some rather intense convection developed over the
  LLCC.  A SSM/I pass indicated that the LLCC was embedded at least 30 nm
  within the eastern portion of some -70 C cloud tops.   Based on a
  consensus of 30 kts satellite intensity estimates, the rapidly improving
  convective pattern, and the new center location, the cyclone was upgraded
  to Tropical Storm Enrique at 0900 UTC on 11 July, located approximately
  500 nm south of Cabo San Lucas.  As Enrique moved northwestward the
  shear lessened and the storm became better organized.   Winds reached an
  estimated peak of 55 kts at 0300 UTC on 12 July with the CP estimated at
  994 mb.  Data T-numbers were 55 kts from all three satellite agencies
  at the time.   Enrique was forecast to briefly reach minimal hurricane
  intensity, but this did not materialize.

     Twelve hours later Enrique still maintained an impressive convective
  pattern with cloud tops to -80 C near the center.  The MSW was still
  estimated at 55 kts, but a TRMM overpass at 12/0802 UTC showed no eye
  and a partly-exposed center west of the main convective bands.  Cirrus
  outflow was good in all directions, but there were hints in satellite
  imagery that weak westerly flow was undercutting the outflow.  Shortly
  thereafter, Enrique moved into colder waters and began to steadily
  weaken.  By 13/0900 UTC the LLCC was rapidly separating from the mid/
  upper-level circulation with the deep convection displaced well to the
  northeast.  First visible images on the 13th showed a LLCC displaced 
  about 150 nm southwest of diminishing deep convection.  Enrique was 
  reduced to depression status at 1500 UTC, and the final advisory was 
  issued at 2100 UTC, locating the convection-free low-level cloud swirl 
  about 650 nm west-southwest of Cabo San Lucas.

     No damage or casualties are known to have resulted from Tropical
  Storm Enrique.

  (Report written by Gary Padgett)



                         TROPICAL STORM FELICIA
                                (TC-06E)
                              18 - 23 July
               ------------------------------------------

     Unlike Enrique and Dolores, the progenitor of Tropical Storm Felicia
  cannot be traced with certainty to a tropical wave of African origin.
  On 13 July a low-pressure area was located south of El Salvador with
  scattered moderate convection within 150 nm of the center.  The LOW
  had moved across Costa Rica the previous night, so it had obviously
  moved from the Caribbean.  Whether or not it was related to a tropical
  wave is unknown to the author.   By mid-day on 14 July the LOW was
  centered south of the Gulf of Tehuantepec and was better organized,
  although most of the deep convection was located west of the LLCC.  The
  system passed about 280 nm south of Acapulco on 15 July, and was better
  organized by the morning of the 16th, but the upper-level environment
  was only marginally favorable for development.   By 0500 UTC on 17 July
  the LOW was located about 390 nm south of Manzanillo, Mexico, and the
  upper-level winds were becoming somewhat less hostile for tropical
  cyclogenesis.

     The first advisory on Tropical Depression 06E was issued at 0300 UTC
  on 18 July.  Dvorak classifications from the three agencies were all
  T2.0, so advisories were initiated.  TD-06E was then located roughly
  325 nm south-southwest of Manzanillo, Mexico.   Like its predecessors,
  the depression developed under substantial easterly shear, although this
  was forecast to lessen with time.   Satellite intensity estimates had
  reached 35 kts six hours later based on a small, cold CDO feature, but
  the forecaster was reluctant to upgrade the system based on nighttime
  infrared imagery.   TD-06E was upgraded to Tropical Storm Felicia with
  45-kt winds on the third advisory at 1500 UTC on 18 July.  The storm was
  located approximately 500 nm south of Cabo San Lucas, moving westward
  at 16 kts.  This was very near the same locale where Enrique had reached
  tropical storm intensity one week earlier.

     The 18/1500 UTC advisory had forecast Felicia to briefly reach
  minimal hurricane intensity, but this was not to be.  Shortly after the
  1500 UTC advisory was released, a SSM/I pass over the cyclone indicated
  that the center was located to the northeast of the convection and not
  embedded as previously estimated.  The initial position was adjusted
  slightly to the north and the winds were decreased to 40 kts.  The new
  forecast (at 2100 UTC) called for Felicia to remain at 40 kts for about
  24 hours and then begin to weaken.  This was exactly what happened.
  Felicia's center remained partly-exposed on the north side of the deeper
  convection as it moved westward at around 15 kts.  By 2100 UTC on the
  19th the center had become exposed to the north of limited deep
  convection and the intensity was reduced to 35 kts.

     Satellite intensity estimates at 20/0000 UTC ranged from 25 to 35 kts,
  but considering the fairly rapid translation speed (16 kts), the MSW
  was retained at 35 kts.  Felicia was still generating limited deep
  convection on the south side, and the environment was still marginally
  favorable for intensification with low shear and SSTs around 26-27 C,
  but the surrounding air was rather dry.  A QuikScat pass at 20/0200 UTC
  showed only a few rain-flagged 35-kt vectors in the convection, so the
  system was downgraded to a depression at 0900 UTC, located about 775 nm
  southwest of Cabo San Lucas.  Felicia continued westward for the next
  three days in advisory status, mainly because of sporadic bursts of
  convection and an occasional T2.5 rating from a satellite analyst.  By
  0900 UTC on 22 July there was no significant convection within 60 nm of
  the fully-exposed circulation center, so the MSW was decreased to 25 kts.
  QuikScat data also supported this intensity.   The weakening cyclone
  still produced some isolated convection for another 24 hours, but by
  0900 UTC on the 23rd it had been convection-free for six hours, and the
  mid/upper-level circulation had been sheared off about 200 nm to the
  east.  With upper-level westerly shear forecast to increase to 50 kts
  within 24 to 48 hours, the final advisory on Felicia was issued.  The
  dissipating center was located approximately 1500 nm west-southwest of
  Cabo San Lucas, or about 1100 nm east-southeast of Hilo, Hawaii.

     There are no reports of damage or casualties associated with Tropical
  Storm Felicia.

  (Report written by Gary Padgett)

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

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

  Activity for July:  2 tropical depressions **
                      1 typhoon ++
                      1 super typhoon

  ** - one classified as a tropical depression by PAGASA--the other by
       PAGASA and JMA

  ++ - classified as a typhoon by JTWC only


                        Sources of Information
                        ----------------------

     Most of the information presented below is based upon tropical
  cyclone warnings and significant tropical weather outlooks issued
  by the Joint Typhoon Warning Center of the U. S. Air Force and
  Navy (JTWC), located at Pearl Harbor, Hawaii.   In the companion
  tropical cyclone tracks file, I normally annotate track coordinates
  from some of the various Asian warning centers when their center
  positions differ from JTWC's by usually 40-50 nm or more.   All
  references to sustained winds imply a 1-minute averaging period
  unless otherwise noted.

     Michael V. Padua of Naga City in the Philippines, owner of the
  Typhoon 2000 website, normally sends me cyclone tracks based upon
  warnings issued by the Japanese Meteorological Agency (JMA) and the
  Philippines' Atmospheric, Geophysical & Astronomical Services
  Administration (PAGASA).  Also, Huang Chunliang of Fuzhou City, China,
  sends me each month tracks obtained from warnings issued by the
  National Meteorological Center of China (NMCC), the Central Weather
  Bureau of Taiwan (CWBT) and the Hong Kong Observatory (HKO).  A very
  special thanks to Michael and Chunliang for the assistance they so
  reliably provide.

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

                    
                Northwest Pacific Tropical Activity for July
                --------------------------------------------

     July activity was close to normal in the Northwest Pacific basin.
  According to information from JTWC's website, the average number of
  tropical cyclones during the 1959-2001 period was 4.7, but this number
  includes tropical depressions plus those reaching tropical storm
  intensity.  The average number of typhoons was listed as 2.7 for the
  period 1960-2001.  July of 2003 produced either 2, 3 or 4 tropical
  depressions, depending on which agency's classifications one follows.
  JTWC issued warnings on only two cyclones, both of which reached typhoon
  status, although JTWC was the only agency to upgrade Koni to a typhoon.
  PAGASA classified and named two other systems as depressions--one of
  these was treated briefly as a tropical depression by JMA also.

     Tropical Storm Koni (named Gilas by PAGASA) moved across the central
  Philippines around mid-month.  After reaching the South China Sea the
  cyclone turned to a northwesterly heading and headed for Hainan Dao
  while intensifying to a minimal typhoon (per JTWC).   Koni then turned
  due westward, crossed Hainan Dao and the Gulf of Tonkin before making
  a final landfall in Vietnam.   Much more potent and deadly was Super
  Typhoon Imbudo (known as Harurot in the Philippines).   Imbudo followed
  the classic west-northwestward "straight-shooter" track from deep in the
  Philippine Sea across northern Luzon and into southern China.  Imbudo
  developed into the year's second super typhoon with 130-kt peak winds
  (per JTWC).  The storm was deadly and destructive to both the Philippines
  and China.

     A weak circulation in the central South China Sea on 9-10 July was 
  classified as Tropical Depression Falcon by PAGASA.   JMA also treated 
  this system as a tropical depression with both agencies estimating 30-kt 
  winds at 0600 and 1200 UTC on the 9th.   Also, PAGASA designated another 
  weak system on 30 July as Tropical Depression Ineng.   Ineng was located 
  well to the east of Mindanao, and the peak MSW was estimated by PAGASA 
  at only 25 kts.  To the author's knowledge, none of the other warning 
  centers treated Falcon or Ineng as tropical depressions.

     The reports on Typhoons Koni and Imbudo were written by Kevin Boyle,
  and contain much information supplied by Huang Chunliang.   A special
  thanks to Kevin and Chunliang for their assistance.



                              TYPHOON KONI
                       (TC-08W / STS 0308 / GILAS)
                              15 - 22 July
             -----------------------------------------------

  Koni: contributed by the Republic of Korea (South Korea), is the Korean
        word for 'swan', a bird which migrates from Siberia each year to
        spend the winter in Korea.  Since a male and female swan mate for
        life, the swan is regarded as a symbol of loyalty.

  Gilas: PAGASA name, means 'amazing'
 
  A. Storm Origins
  ----------------------

     During the second week of July the monsoon trough became very active,
  spawning two disturbances, the first of which was mentioned in the 1500
  UTC 12 July STWO issued by JTWC.  The associated area of deep convection
  was situated near 5.9N, 144.1E, or approximately 440 nm south of Guam.
  Animated infrared satellite imagery revealed an area of cycling deep
  convection whilst upper-level wind shear conditions were ideal for
  further development with moderate diffluence aloft.  However, development
  potential for the next 24 hours was assessed by JTWC to be poor.  This
  was upgraded to fair at 13/0600 UTC based on observations made by multi-
  spectral satellite animations which showed cycling deep convection
  organizing around a weak LLCC.  However, the convection associated with
  the possible LLCC subsequently decreased significantly such that the
  potential for development was once again downgraded to poor status.  

     The disturbance changed little through the 14th, although late in the
  day animated multi-spectral satellite imagery revealed that cycling deep
  convection was re-organizing with spiral bands and low-level cloud lines
  becoming evident.  This observation prompted JTWC to issue a TCFA at
  15/0200 UTC which located the disturbance near 10.9N, 135.2E, or 125 nm
  north of Palau.    The STWO issued at 15/0600 UTC stated that recent
  QuikScat and synoptic data were locating the position of the surface
  circulation further south than previously thought.  By this time, the
  system had crossed 135E into PAGASA's AOR and was named Gilas by that
  agency, who considered it as a 30-kt tropical depression (10-min avg).
  Tropical Depression 08W finally made its debut at 15/1200 UTC, located
  180 nm northwest of Palau (or 9.9N, 132.6E), moving westward at 11 kts.
  The MSW was estimated to be 25 kts at this stage.  Tropical Depression
  08W/Gilas continued westward while slowly strengthening and
  consolidating.  By 16/0600 UTC the MSW had reached minimal tropical
  storm intensity.  (JMA began including the depression in their High Seas
  bulletins at 16/0000 UTC, fixing the system near 9.2N, 130.5E.).


  B. Track and Intensity History
  ------------------------------

     By 0000 UTC on 17 July Tropical Storm 08W/Gilas had moved steadily
  westward with the partially-exposed LLCC located along the east coast
  of Samar.  The MSW had, by this time increased to 45 kts, but the system
  had a more hostile environment to work with:  easterly wind shear, land
  influences, and loss of an outflow pattern that had been provided by a
  TUTT cell located to the northeast.  CIMSS Shear Products showed that
  the easterly shearing conditions were likely to persist even as Gilas
  cleared the Philippines, but the shear was considered light enough for
  some strengthening to occur.   By 17/1200 UTC TS-08W/Gilas had crossed
  the bulk of the Philippines with the centre located over Panay Island.
  The system had weakened slightly to 40 kts during its transit across
  land, but the MSW slowly began to trend upward as the cyclone moved into
  the South China Sea at 1800 UTC (after the centre had clipped northern
  Palawan).  Multi-spectral and microwave imagery indicated an increase in
  deep convection at this time.  A shortwave trough moving off the Chinese
  coast created a  weakness in the subtropical ridge during the 18th, and
  in response to this the storm (now named Koni following JMA's upgrade to
  tropical storm status at 18/0600 UTC) began a northwesterly jaunt which
  was to last through the 19th, 20th, and into the 21st.   (HKO and NMCC
  initiated bulletins at 17/0600 UTC and 18/0600 UTC, respectively,
  although NMCC had issued two irregular bulletins on the system at
  17/0000 UTC and 18/0000 UTC.  Both NMCC and HKO upgraded the system to
  tropical storm intensity at 18/0600 UTC.)

     At 19/0000 UTC Tropical Storm Koni was centred approximately 225 nm
  west-southwest of Manila (13.7N, 117.2E), well away from the interfering
  influence of the Philippines.   However, intensification was slow due to
  the effects of northeasterly shearing conditions until 1800 UTC, when
  Koni took advantage of a lull in the shear and strengthened into a 65-kt
  minimal typhoon at 20/0000 UTC.  (At 19/1800 UTC PAGASA ceased issuing
  bulletins as Koni/Gilas had moved out of their AOR, allowing the
  agency to fully concentrate on the rapidly developing Super Typhoon
  Imbudo/Harurot.)  Koni's movement at this time was still toward the
  northwest at around 10 kts, the centre fixed approximately 370 nm west-
  northwest of Manila and roughly halfway to its impending landfall on
  Hainan Dao.  Although Koni was expected to strengthen some more in a
  favourable environment, the MSW of 65 kts was to be its maximum
  intensity.  However, this intensity was maintained through the 20th and
  21st as the cyclone continued on a northwesterly heading, then westward
  toward the island of Hainan.  A 20/1309 UTC GHz microwave pass revealed
  a partial symmetrical eye.  (Note: None of the Asian TCWCs upgraded
  Koni to typhoon status.  Peak intensities include: JMA 60 kts with a CP
  of 975 mb, PAGASA 55 kts, NMCC 60 kts and HKO 55 kts.   All the MSW
  estimates represent a 10-min avg).

     At 0000 UTC on the 21st Typhoon Koni was centred approximately 90 nm
  east-southeast of Hainan Dao, moving toward the northwest at 8 kts.
  Upper-level northeasterly shear was stripping away the convection in the
  northern quadrants of the storm, as seen in water vapor imagery.
  Animated infrared satellite imagery and synoptic observations indicated
  that Typhoon Koni made landfall approximately 30 nm south-southwest of
  Qionghai on the island of Hainan at approximately 21/1200 UTC.   By this
  time Koni had taken a westerly track which was to persist for the rest
  of its existence.    After its transit across Hainan Dao the cyclone
  continued westward across the Gulf of Tonkin and began to weaken as it
  approached the Vietnamese coast.   The MSW fell below typhoon strength
  and had dropped to 55 kts by the time Koni crossed into northern Vietnam
  south of Hanoi at approximately 22/0800 UTC.   The storm rapidly
  deteriorated as it moved further inland, and was dissipating as a
  significant tropical cyclone over land by 22/1800 UTC--the time of the
  final warning issued by JTWC.  The system was tracking westward over
  Laos on this last advisory with the MSW estimated at 35 kts.  HKO issued
  their final warning at 22/0900 UTC (AOR limit) while both JMA and NMCC
  ended their coverage six hours later.


  C. Meteorological Observations
  ------------------------------

     Based on NMCC warnings, Tropical Storm Koni made landfall near
  Beipo Town, Wanning City, Hainan Province around 21/1035 UTC with the
  MSW estimated at 41 kts and a CP of 980 hPa.   Gusts of Beaufort Force
  8-10 were widely recorded on Hainan Dao, and Dofang City reported a
  gust of Beaufort Force 11.

     Following are a few rainfall totals from Chinese stations sent by
  Huang Chunliang:

  Station           Province       Rainfall (mm)   Reporting Period (UTC)
  -----------------------------------------------------------------------

  Lingshui          Hainan             177           21/0000 - 22/0000
  Lingshui          Hainan             179.5         20/0000 - 22/0600
  Sanya             Hainan             112           21/0000 - 22/0000
  Wuzhi Mt.         Hainan             188.8         20/0000 - 22/0800


  C. Damage and Casualties
  ------------------------

     In Hainan Province, China, some 249,000 residents in 122 towns of
  15 cities/counties were affected by the storm.  The number of collapsed
  houses was approximately 1400 in the province.  Direct economic losses
  were estimated at 140,270 million yuan.   The "swan" did cause some
  damage to Hainan, especially in the cities of Wanning, Baoting and
  Sanya.  However, thanks to the storm, the drought at that time in the
  province was also eased.   No casualties were reported in association
  with Severe Tropical Storm Koni.

     According to press reports, two people were killed and 18 injured as
  Tropical Storm Koni lashed northern Vietnam, uprooting trees, felling
  power lines and destroying more than 1000 houses.  There were no reports
  of damage or casualties in the Philippines.

     The HKO report on Severe Tropical Storm Koni can be accessed at the
  following URL:

     http://www.weather.gov.hk/informtc/koni/koni.htm>

  (Report written by Kevin Boyle with important contributions by
  Huang Chunliang)



                         SUPER TYPHOON IMBUDO
                     (TC-09W / TY 0307 / HARUROT)
                             15 - 25 July
           ------------------------------------------------

  Imbudo: contributed by the Philippines, means 'funnel' or 'downspout'

  Harurot: PAGASA name, means 'very fast' or 'top speed'

  A.  Storm Origins
  ----------------------

     An area of convection developed and persisted approximately 100 nm
  east-southeast of Chuuk and was first included in the 0600 UTC 14 July
  STWO.  Animated multi-spectral satellite imagery revealed disorganized
  deep convection near a possible weak LLCC.    Upper-level analysis
  indicated moderate diffluence aloft and weak vertical wind shear over
  the area.  The potential for the development of a significant tropical
  cyclone within the next 24 hours was considered  poor.  This was upgraded
  to fair at 15/0230 UTC after the system had become better organized with
  low-level cloud lines evident at the surface as seen in multi-spectral
  satellite animation.  A TCFA was issued at 16/0600 UTC on the basis of
  continued organization.  Upper-air analysis at this time revealed that
  an upper-level LOW to the northwest was aiding development by providing
  an ideal poleward outflow pattern.

     The first warning was issued on Tropical Depression 09W at 16/1800
  UTC with the system located 360 nm east-southeast of Yap Island
  (near 7.0N, 143.6E), and the initial movement was toward the north-
  northwest at 11 kts.  This heading swung to the west-northwest during
  the 17th as the cyclone's broad circulation continued its organization
  phase.  The depression was named Imbudo at 17/0600 UTC when JMA upgraded
  it to tropical storm intensity.  At the same time NMCC issued the first 
  warning after issuing an irregular bulletin at 17/0000 UTC.   Tropical
  Depression Imbudo was soon upgraded by JTWC six hours later with the MSW
  estimated at 40 kts.  Recent microwave and animated infrared imagery 
  indicated that banding features were developing and wrapping in from 
  the north of the system.       The centre of Imbudo had moved west-
  northwestward to a position 125 nm east-southeast of Yap Island by the 
  time of JTWC's upgrade.


  B. Storm History
  ----------------

     By 0000 UTC on 18 July Tropical Storm Imbudo was passing 18 nm north
  of Yap Island while moving west-northwestward with a MSW of 50 kts.
  Strengthening continued as the day wore on--the forward motion slowed to
  around 5 to 7 kts and the heading turned westerly.  This was forecast to
  persist for the near term, but a passing longwave trough was expected to
  dig into the mid-level ridge, forcing the storm northwestward for awhile.
  An upper-level LOW situated to the northeast of the tropical cyclone, as
  seen on UW-CIMSS 200-mb data, provided excellent diffluence, allowing
  Imbudo to strengthen into a 65-kt typhoon by 1800 UTC that same day.
  This signalled the beginning of a rapid intensification phase which
  brought Imbudo up to 75 kts by 19/1200 UTC.  Satellite CI estimates had
  reached 90 kts by 20/0000 UTC, and a 19/2044 UTC TRMM pass revealed a
  15-nm irregular eye.

     Only twelve hours more were required for Imbudo to attain super
  typhoon status of 130 kts--its peak intensity.  (JMA's peak of 85 kts
  10-min avg MSW with a CP of 945 mb was not reached until near landfall.)
  At this time (20/1200 UTC) Imbudo was centred approximately 595 nm east-
  southeast of the northern tip of Luzon, moving on an accelerating west-
  northwesterly track at around 10 kts.  (PAGASA had initiated warnings on
  Imbudo at 19/0000 UTC, naming the cyclone Harurot with an initial
  intensity of 65 kts.  Also, at this time NMCC upgraded Imbudo to severe
  tropical storm status, then to typhoon intensity at 19/1800 UTC.  Both
  agencies implement a 10-minute wind averaging period.)  A 20/2119 UTC
  SSM/I pass revealed well-defined concentric eyewalls, but water vapour
  imagery showed that the TUTT cell to the east of Imbudo was weakening
  and moving further away, no longer enhancing the poleward outflow
  channel.

     At 0000 UTC 21 July Imbudo had maintained its super typhoon title
  with a MSW of 130 kts.   The well-defined eye was tracking west-
  northwestward at 12 kts (controlled by the mid-level steering of a
  subtropical ridge centred near Okinawa) and was located 510 nm southeast
  of the northern tip of Luzon, i.e. near 13.7N, 128.3E.   By 21/2100 UTC
  the 6-nm eye had reached a position about 160 nm east of Luzon, or near
  15.6N, 124.4E.   Microwave imagery prior to landfall revealed that the
  system was undergoing a concentric eyewall cycle.  By 22/0600 UTC Super
  Typhoon Imbudo had struck the east coast of Luzon with all its strength
  and might.   (PAGASA's maximum 10-min avg intensity at landfall was 100
  kts with the CP estimated at 941 mb.)  Over land the cyclone's intensity
  waned quickly with the MSW dropping to 80 kts at 1200 UTC.   By 1800 UTC
  Imbudo/Harurot had moved back over water again, having tracked across
  Luzon, and with the MSW increasing to 90 kts.  (By 23/0000 UTC Imbudo had
  moved out of PAGASA's AOR and their final warning written.)
  Strengthening then ceased and this (90 kts) was to be Imbudo's peak
  intensity during the second part of its career.  

     At 0000 UTC on 23 July Typhoon Imbudo was crossing the South China
  Sea, located approximately 270 nm southeast of Hong Kong, China, as it
  continued on its west-northwesterly track at 17 kts.  (HKO and CWB of 
  Taiwan had been including the tropical cyclone in their bulletins since
  21/1800 UTC and 21/1200 UTC, respectively.)  Microwave imagery indicated 
  that the system was once again re-organizing and strengthening.  The MSW 
  remained at 90 kts, although it dipped slightly to 85 kts for one warning
  cycle at 23/1200 UTC.  At this time animated satellite imagery indicated 
  that the deep convection had decreased slightly during the previous six 
  hours.  The intensity was raised again to 90 kts at 1800 UTC, by which 
  time dry air entrainment was noted in the northwestern quadrant in 
  enhanced infrared satellite animation imagery.  The typhoon was then 
  passing 135 nm south of Hong Kong, China (21.1N, 111.9E).  (HKO's and 
  NMCC's estimated MSW at this time was 90 kts, 10-min avg.)

     The large, banding eye of Imbudo (as seen in enhanced multi-spectral
  imagery) was about to make landfall in Guangdong province near the city
  of Yangjiang at 24/0000 UTC.  The official time of landfall, as noted in
  the 24/0600 UTC warning, was 24/0100 UTC.  Upon making landfall Typhoon
  Imbudo weakened rapidly and was downgraded to a 55-kt tropical storm on
  the final JTWC warning issued at 24/1200 UTC, located inland near 23.0N,
  108.9E.    NMCC continued to issue warnings on the weakening system
  through the 24th, downgrading it first to a severe tropical storm at
  24/0600 UTC, and then to a tropical storm at 24/1200 UTC.  NMCC's final
  bulletin was issued at 25/0000 UTC after their MSW (10-min avg) had
  dropped to 30 kts (a tropical depression).  HKO was also still monitoring
  the dying storm, issuing their last bulletin at 24/1800 UTC after Imbudo
  had passed out of their AOR.  JMA followed Imbudo's progress into China,
  issuing their final warning at 24/2100 UTC.

  (Editor's Note:  Both NMCC and HKO treated Imbudo as a more intense 
  system at landfall than did JTWC.   NMCC's 10-min avg MSW of 90 kts and
  HKO's 10-min avg MSW of 95 kts equate to 1-min avg MSW values of roughly
  100 and 105 kts, respectively.   JMA, on the other hand, lowered their
  MSW estimate to 65 kts at 24/0000 UTC, shortly before landfall.)

  
  C. Meteorological Observations
  ------------------------------

     The following meteorological observations were all sent by Huang
  Chunliang.


  (1) Rainfall Observations
  -------------------------

  Station                    Province        Rainfall   Recording Period
                             or Region         (mm)          (UTC)
  ----------------------------------------------------------------------

  Ding'an                     Hainan            230    23/0600 - 24/0600
  Wenchang                    Hainan            213    23/0600 - 24/0600
  Haikou                      Hainan            163    23/2100 - 24/2100
  Dianbai                     Guangdong         108    23/2100 - 24/2100
  Zhanjiang                   Guangdong         100    23/2100 - 24/2100
  Gaozhou Met. Bureau         Guangdong         203.1  24/0000 - 25/0000
  Dongxing                    Guangxi           221    23/2100 - 24/2100
  Lingshan                    Guangxi           217    23/2100 - 24/2100
  Beihai                      Guangxi           137    23/2100 - 24/2100
  Qinzhou                     Guangxi           113    23/2100 - 24/2100
  Hepu                        Guangxi           343    23/1200 - 25/0000
  Chih Pen, Taitung Cnty.     Taiwan            137    21/1600 - 23/0600
  Cheng Kung, Taitung Cnty.   Taiwan            115    21/1600 - 23/0600


  (2) Wind Observations
  ---------------------

     The following are peak gust reports:

  Station                             Peak Gust 
  ---------------------------------------------

  Shaba Town, Yangxi County             71 kts
  Hailing Dao                           76 kts
  Shangchuan Dao                       108 kts
  Nansha, Guangzhou City                50 kts
  Dongping Town, Yangjiang County       86 kts
  Huazhou Meteorological Bureau         80 kts
  Shiji Town, Panyu District,           58 kts
     Guangzhou City

     Based on NMCC warnings, the center of Typhoon Imbudo made landfall
  near the counties of Yangxi and Dianbai, Guangdong Province, around
  24/0200 UTC with MSW (10-min avg) of 38 m/sec (74 kts) and a CP of
  960 hPa.   Danzao, Nanhai District, Foshan City, reported a peak
  sustained easterly wind of 18.9 m/sec (37 kts) at 24/0413 UTC with
  a peak easterly gust of 29.2 m/sec (57 kts) at 24/0405 UTC.  In the
  Guangxi Zhuang Autonomous Region, twenty-one counties/cities had
  reported winds of gale force or higher as of 25/0000 UTC.  Among
  those, a station in Heng County recorded a peak sustained wind of
  39 kts while Guigang City recorded a peak gust of 64 kts.


  D. Damage and Casualties
  ------------------------

  (1) Philippines
  ---------------

     Imbudo was the strongest typhoon to hit the Philippines in five
  years.  In October, 1998, Super Typhoon Zeb/Iliang claimed 83 lives,
  and later that month Super Typhoon Babs/Loleng was responsible for
  221 deaths.  The highest death toll in the Philippines noted by the
  author is twenty-one.   A few days after the storm there were six
  persons missing and unaccounted for.    Eleven of the deaths (in the
  Maguindanao and Sultan Kudarat areas) were attributed to flash flooding
  while six (from Romblon in the central Philippines region and
  Ilocos Norte province in the north) were caused by typhoon-related
  accidents.  Several areas suffered power failures (including Metro
  Manila) caused by fallen power lines.

     A total of 60 evacuation centres were required for 10,921 persons.
  A total of 3,379 houses were partly damaged and 812 totally destroyed.
  Most of the destruction occurred across northern and eastern parts of
  the country.

  Cost estimates so far include:

  Agriculture/livestock - RDCC's report - 440.2 million pesos
                          DA's report - 1,098.6 billion pesos

  Infrastructure - RDCC's report - 259.0 million pesos
                   DPWH's report - 7.0 million pesos


  (2) China
  ---------

  (a) Hainan Province
  -------------------

     Some 249,000 residents in 65 towns of six cities/counties (including
  Haikou City, Wechang City, Ding'an County, Chengmai County, Danzhou
  City, Lingao County, especially Haikou and Wechang) were affected by
  the typhoon.  Direct economic losses there were estimated at 55.35
  million yuan.  No casualties were reported in the province.


  (b) Guangdong Province
  ----------------------

     Some 381 towns of 7 cities were stricken by the typhoon with
  4,776,100 residents being affected.  The death toll from Typhoon
  Imbudo had risen to eight when local governments assessed the cost
  of the damage.  More than 24 others were injured and three fishermen
  were still missing as of the press report, according to the Guangdong
  Provincial Anti-flooding Headquarters.   Total direct economic losses
  in the province were estimated at 1.9066 billion yuan with 595,000
  houses collapsed.  In Zhanjiang City, electricity and water supplies
  were suspended in many counties and towns after power poles and
  pumping stations were destroyed.


  (c) Guangxi Zhuang Autonomous Region
  ------------------------------------

     About 3,951,100 residents in Guangxi Region were affected by the
  typhoon.  Preliminary statistics indicate that Typhoon Imbudo killed
  12 people due to falling trees and chimneys, landslides, and drowning.
  Losses in the region amounted to 499.6 million yuan (equivalent to
  $60.2 million US).  Most areas of Guangxi, including the cities of
  Yulin, Beihai, Qinzhou, Fangchenggang, Guigang, and Nanning experienced
  heavy rainfalls.  Also, according to official figures, 130 head of
  livestock died due to Imbudo, some 4950 houses collapsed, and 116,000
  hectares of crops were ruined.

     An excellent report on Typhoon Imbudo can be found on the website
  of the Hong Kong Observatory:

     http://www.weather.gov.hk/informtc/imbudo/imbudo.htm>

     Also, some reports on the storm's effects may be found at the
  following URL:

     http://www.reliefweb.int/w/rwb.nsf/vLND>

  (Report written by Kevin Boyle with important contributions by
  Huang Chunliang)

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

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

  Activity for July:  2 tropical depressions **

  ** - classified by IMD only


               North Indian Ocean Tropical Activity for July
               ---------------------------------------------

     On 27 July JTWC re-issued a STWO for the North Indian Ocean to include
  an area of convection which had moved into the northern Arabian Sea from
  the Pakistani coast.  The development potential was assessed as fair, but
  this was downgraded to poor later on the 27th after the convection had
  significantly decreased around a possible LLCC.  However, increased low-
  level organization on the 28th brought about a re-classification of the
  potential for development to fair, but the convection had essentially 
  dissipated by 24 hours later and the system was dropped as a suspect area
  for tropical cyclone formation.

     Early in August, Huang Chunliang informed me that IMD had classified
  this system as a depression, and furthermore, that there had been a
  circulation a few days earlier in the Bay of Bengal which had also been
  treated as a depression.    Before Chunliang could create and send me
  tracks for these systems based on IMD bulletins, his computer was struck
  by a virus which wiped out most of his files.  Based on his memory and
  a few notes which did survive, Chunliang sent me some information on
  these North Indian systems.  (A special thanks to Chunliang for sending
  the information.)

     On 25 July IMD upgraded a low-pressure area to depression status and
  placed its center near 21.0N, 89.0E.  A bulletin issued around 0700 UTC
  on the 26th indicated that the depression had intensified into a "deep
  depression" (i.e., 30 kts) during the late evening, crossed the North
  Orissa coast and was centered near Balasore in Orissa state.   This
  system was never referenced by JTWC in their STWOs, and neither was it
  listed as an "invest" area on NRL's tropical cyclone page.

     The weakening system continued to move farther inland in a west-
  northwesterly direction and was downgraded to a depression (25 kts) on
  either the 26th or 27th.  A bulletin at 28/0700 UTC mentioned intense
  convective clouds over the northeast Arabian Sea off the Gujart coast
  in association with the depression over northwest Madhya Pradesh about
  100 km southwest of Guna.  The system was downgraded to a low-pressure
  area later on the 28th.  It is interesting to note that the Pakistan
  Meteorological Department (PMD) referred to the system as a monsoon
  depression in a bulletin on the 28th.

     Regarding the Arabian Sea disturbance mentioned in the first para-
  graph above, IMD upgraded this system to depression status at 29/0300
  UTC, placing the center near 21.5N, 64.5E.  However, it soon weakened
  and was downgraded to a low-pressure area on the 30th.  (JTWC had
  dropped it as a suspect area at 29/1800 UTC.)  PMD also referred to this
  system as a monsoon depression which brought heavy rains to northwest
  India and neighboring southern Pakistan, triggering severe flooding in
  Sindh Province and some other locations.  As of the first week in August,
  more than 230 deaths had been attributed to severe monsoonal flooding
  in Pakistan.  Several reports on the Pakistani flooding may be found
  at the following website:

     http://www.reliefweb.int/w/rwb.nsf/vLND>

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

  SOUTHWEST INDIAN OCEAN (SWI) - South Indian Ocean West of Longitude 90E

  Activity for July:  No tropical cyclones

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

  NORTHWEST AUSTRALIA/SOUTHEAST INDIAN OCEAN (AUW) - From 90E to 135E

  Activity for July:  No tropical cyclones

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

  NORTHEAST AUSTRALIA/CORAL SEA (AUE) - From 135E to 160E

  Activity for July:  No tropical cyclones

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

  SOUTH PACIFIC (SPA) - South Pacific Ocean East of Longitude 160E

  Activity for July:  1 subtropical LOW


                 South Pacific Tropical Activity for July
                 ----------------------------------------

     No out-of-season tropical cyclones formed in the South Pacific basin
  during July, but there was a system around mid-month with some
  characteristics of subtropical storms which brought very heavy rains and
  high winds to the islands of New Caledonia.  Thierry Lefort of Meteo
  France reported that the upper-level cyclonic circulation reached the
  surface and disturbed the tradewind flow without becoming involved with
  any pre-existing surface fronts.  Also, the strongest winds were well-
  removed from the center of the LOW.   Pete Donaldson of CPHC stated that
  the system was very much like a Kona LOW of the Hawaiian Islands region.

     Record rainfall amounts were recorded on the Loyaute Islands, which
  are flat coral islands:  420 mm in 24 hours and 530 mm for the entire
  event.  A Meteo France station on the east coast of Grande Terre
  recorded 10-min avg sustained winds of 49 kts with a peak gust of 64 kts.
  (This reading was likely orographically influenced by the nearby range
  of mountains.)  On Ile des Pins, which is flat, a MSW of 39 kts was
  recorded with gusts to 60 kts.   Noumea, on the poleward side of the LOW
  and adjacent to an anticyclone, measured peak gusts of 55 kts.  Jeff
  Callaghan reported that the 10-metre yacht Astrid had been dismasted
  near 24.6S, 153.9E (near Fraser Island) by large waves generated by
  the New Caledonia LOW, and that a large-scale air and sea search was
  underway (on 17 July) for the vessel.  (Thanks to Thierry and Jeff for
  sharing this information.)

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

                               EXTRA FEATURE

     In order to shorten the amount of typing in preparing the narrative
  material, I have been in the habit of freely using abbreviations and
  acronyms.   I have tried to define most of these with the first usage
  in a given summary, but I may have missed one now and then.  Most of
  these are probably understood by a majority of readers but perhaps a
  few aren't clear to some.  To remedy this I developed a Glossary of
  Abbreviations and Acronyms which I first included in the July, 1998
  summary.  I don't normally include the Glossary in most months in
  order to help keep them from being too long.  If anyone would like to
  receive a copy of the Glossary, please e-mail me and I'll be happy
  to send them a copy.

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

  AUTHOR'S NOTE:  This summary should be considered a very preliminary 
  overview of the tropical cyclones that occur in each month. The cyclone
  tracks (provided separately) will generally be based upon operational
  warnings issued by the various tropical cyclone warning centers.  The
  information contained therein may differ somewhat from the tracking and
  intensity information obtained from a "best-track" file which is based
  on a detailed post-seasonal analysis of all available data. Information
  on where to find official "best-track" files from the various warning
  centers will be passed along from time to time.

    The track files are not being sent via e-mail.  They can be retrieved
  from the archive sites listed below.  (Note: I do have a limited e-mail
  distribution list for the track files.    If anyone wishes to receive
  these via e-mail, please send me a message.)

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

    http://australiasevereweather.com/cyclones/>
    http://www.typhoon2000.ph> OR http://64.235.42.210>
    http://mpittweather.com>
    ftp:// ftp.aoml.noaa.gov/hrd/pub/landsea/padgett/>


     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:

    http://www.met-office.gov.uk/sec2/sec2cyclone/sec2cyclone.html>


                    TROPICAL CYCLONE REPORTS AVAILABLE

     JTWC now has available on its website the complete Annual Tropical 
  Cyclone Report (ATCR) for 2002 (2001-2002 season for the Southern 
  Hemisphere).  ATCRs for earlier years are available also.

     The URL is:  http://199.10.200.33/jtwc.html>

     Also, TPC/NHC has available on its webpage nice "technicolor"
  tracking charts for the 2002 Atlantic and Eastern North Pacific
  tropical cyclones; also, preliminary storm reports for all the 2002
  Atlantic and Eastern North Pacific cyclones are now available, as
  well as track charts and reports on storms from earlier years.

     The URL is:  http://www.nhc.noaa.gov>


     A special thanks to Michael Bath of McLeans Ridges, New South Wales,
  Australia, for assisting me with proofreading the summaries.


  PREPARED BY

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

  John Wallace  (Eastern North Pacific, North Indian Ocean, Western
                 Gulf of Mexico)
  E-mail:  [email protected]

  Kevin Boyle  (Eastern Atlantic, Western Northwest Pacific, South
                China Sea)
  E-mail:  [email protected]

  Simon Clarke  (Northeast Australia/Coral Sea, South Pacific)
  E-mail:  [email protected]

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

Document: summ0307.htm
Updated: 26th October 2006

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