Tropical Cyclones
Main Index Home Page Stock Weather Photos Australian Severe Weather Forum Storm News and Storm Chasing Reports Tropical Cyclones / Hurricanes / Typhoons Weather Data and Links Wild Fires / Bushfires Weather Observation Techniques Weather Picture Catalogue Tornado Pictures and Reports Stock Video Footage and DVDs for sale
Monthly Global Tropical Cyclone Summary March 2004
[Summaries and Track Data] [Prepared by Gary Padgett]


                               MARCH, 2004

  (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.)


                            MARCH HIGHLIGHTS

  --> Intense tropical cyclone makes devastating strike on Madagascar
  --> Long-lived and erratic intense cyclone strikes Western Australia
  --> Significant hybrid storm affects eastern Australia
  --> First observed South Atlantic hurricane on record makes damaging
      strike in southeastern Brazil


                 ***** Feature of the Month for March *****


     Tropical cyclones in the Atlantic Ocean, Gulf of Mexico, and
  Caribbean Sea are assigned names by the Tropical Prediction Center/
  National Hurricane Center in Miami, Florida.   A separate alphabetical
  set of alternating male/female names is used each year with the name
  of the first tropical storm beginning with the letter "A".  Names are
  repeated every six years.  The names of hurricanes which cause a lot
  of damage and/or fatalities are usually retired from the list with
  another name of the same alphabetical rank and gender replacing it.
  Following the 2003 season, the names Fabian, Isabel and Juan were
  retired.  Fabian has been replaced with Fred, Isabel with Ida, and
  Juan with Joaquin in the list of names for 2009.

     The highest number of tropical storms named in one season in the
  Atlantic was 19 during the very active 1995 season.  The most active
  Atlantic tropical cyclone season on record was 1933, in which 21 storms
  were charted, but of course that season pre-dates the formal naming of
  tropical cyclones.  The active 1969 season is credited with 17 tropical
  cyclones (plus one subtropical storm), but only 13 were actually named
  operationally.    Several of the systems began as hybrid/subtropical
  storms and forecasters at the time were still debating how to classify
  this type of storm system, and so they remained unnamed.   A few years
  later several tracks were added to the official Best Tracks database.
  Two of these unnamed storms were hurricanes, thus giving 1969 a total
  of 12 hurricanes--the current record for the Atlantic.

     The list of names for 2004 is the same one used during the active
  hurricane season of 1998 when fourteen tropical cyclones were named.
  Georges and Mitch were the destructive hurricanes of 1998, and those
  names have been replaced with Gaston and Matthew in the list for 2004.

     TPC/NHC also has warning responsibility for the Eastern North
  Pacific Ocean from the west coast of Mexico out to longitude 140W.
  Six separate alphabetical sets of names are used for this basin in
  the same manner as in the Atlantic.  Initially, the Eastern Pacific
  name sets contained only 21 names, omitting "Q" and "U" and ending
  with the letter "W", as in the Atlantic.  When the active 1985 season
  threatened to exhaust the list, the names Xina, York and Zelda were
  drafted to accommodate any additional storms which might develop.
  (Hurricane Xina was named in late October, 1985.)  The decision was
  made sometime in the latter 1980s to extend the list with these three
  names in odd-numbered years, and to add the names Xavier, Yolanda and
  Zeke in even-numbered years (to preserve the alternating gender
  scheme).  During the Northeast Pacific's year of record activity in
  1992, all 24 names were allotted to tropical cyclones forming east of
  140W, ending with Tropical Storm Zeke in late October.  Had more storms
  developed, they would have been named with the letters of the Greek
  alphabet (Alpha, Beta, etc), which is also the backup plan for the
  Atlantic basin in case more than 21 tropical storms develop in a single

    The list for this year was last used in 1998 when thirteen tropical
  cyclones were named, the last one being Madeline.  However, six years
  earlier, in 1992, all 24 names were used during the most active
  Eastern Pacific on record since the advent of meteorological satellites.

     The Central Pacific Hurricane Center, located in Honolulu, has
  tropical cyclone warning responsibility for that portion of the North
  Pacific Ocean lying between longitudes 140W and 180.  The majority of
  the tropical storms and hurricanes seen in that region are visitors
  from east of 140W, but on the average about one tropical storm forms
  in the Central Pacific each year, and when this happens, the storm is
  given a Hawaiian name.   The list consists of four sets of twelve
  names each, using only the letters of the Hawaiian alphabet.  All the
  names are used--the first storm to form in a given year is assigned
  the next available name on the list.  No tropical cyclones were named
  by CPHC in 2003.  The last storm to form in Central Pacific waters
  was Hurricane Huko in late October, 2002, so the next name to be
  assigned will be Ioke.

     Names for 2004 are (** indicates name has already been assigned):

            ATLANTIC                EASTERN PACIFIC        CENTRAL PACIFIC

    Alex           Lisa          Agatha         Madeline        Ioke
    Bonnie         Matthew       Blas           Newton          Kika
    Charley        Nicole        Celia          Orlene          Lana
    Danielle       Otto          Darby          Paine           Maka
    Earl           Paula         Estelle        Roslyn          Neki
    Frances        Richard       Frank          Seymour         Oleka
    Gaston         Shary         Georgette      Tina            Peni
    Hermine        Tomas         Howard         Virgil          Ulia
    Ivan           Virginie      Isis           Winifred        Wali
    Jeanne         Walter        Javier         Xavier          Ana
    Karl                         Kay            Yolanda         Ela
                                 Lester         Zeke            Halola


     For the past few years I have included tables of Atlantic and
  Northeastern Pacific monthly net tropical activity (NTC).  As part of
  the monthly feature for March, I have included tables for the Atlantic
  basin--the Northeast Pacific will follow in a later summary.  When
  breaking up a tropical cyclone season temporally (into months), some
  decisions have to be made regarding inter-monthly cyclones.  I have
  previously explained in detail my reasoning here, and interested
  persons can find this in the March, 2002, summary, which can be
  obtained from any of the websites listed at the end of this summary.

     The 2003 Atlantic tropical cyclone season has the distinction of
  being the longest lasting season on record as measured from the
  beginning of the first named cyclone to the dissipation of the final
  storm.  Subtropical Storm Ana was reclassified as a tropical storm
  on 21 April, making it the first known Atlantic tropical storm to
  have formed in the month of April.  Tropical Storm Peter weakened
  into a depression on 10 December--234 days after Ana had formed.
  Tropical Storms Odette and Peter were the first Atlantic December
  tropical cyclones since Hurricane Lili in 1984, and constitute the
  first occurrence of two December tropical storms since 1887.  The
  first table below lists the monthly statistics for the individual
  months as well as the seasonal totals.  The second table lists the
  monthly figures over the period 1950-2003, inclusive.

                    Atlantic Hurricane Season of 2003

  Month    NS      H       IH       NSD       HD       IHD      NTC

  APR       1      0        0       3.25     0.00      0.00     2.8
  MAY       0      0        0       0.00     0.00      0.00     0.0
  JUN       1      0        0       1.50     0.00      0.00     2.2
  JUL       2      2        0      12.25     2.00      0.00    14.4
  AUG       3      2        1       6.75     2.25      1.25    25.6
  SEP       4      3        1      29.75    23.00     14.00    92.2
  OCT       3      0        1      21.50     5.50      1.50    28.0
  NOV       0      0        0       0.00     0.00      0.00     0.0
  DEC       2      0        0       4.25     0.00      0.00     4.8

  TOTAL    16      7        3      79.25    32.75     16.75     170

                    Atlantic Basin Monthly NTC Table

  Month    NS      H       IH       NSD       HD       IHD      NTC
  JAN       0      0        0       4.50     3.50      0.00     0.07
  FEB       1      0        0       1.50     0.00      0.00     0.04
  MAR       0      0        0       0.00     0.00      0.00     0.00
  APR       1      0        0       3.25     0.00      0.00     0.05
  MAY       5      2        0      18.50     6.25      0.00     0.45
  JUN      28     10        2      73.75    13.25      0.75     2.32
  JUL      44     19        1     135.25    34.25      0.50     3.78
  AUG     144     81       31     625.25   298.75     64.50    24.25
  SEP     189    132       67    1201.50   674.50    171.00    47.66
  OCT      90     57       19     496.25   237.75     41.75    16.81
  NOV      27     20        4     126.75    46.00      6.25     4.14
  DEC       5      2        0      17.00     3.75      0.00     0.41

  TOTAL   534    323      123    2703.50  1318.00    284.75

  AVG     9.9    6.0      2.3      50.0     24.4       5.3

     The following table contains the annual statistics for the Atlantic
  basin for the period 1950-2003:

                   Atlantic Basin Annual NTC Table

  Year     NS      H      IH        NSD       HD      IHD     NTC

  1950     13     11       7       98.00    59.50    15.50    225
  1951     10      8       2       57.75    36.25     5.00    113
  1952      7      6       3       39.75    22.75     4.00     91
  1953     14      6       3       64.50    18.00     5.75    113
  1954     11      8       2       51.75    31.50     9.00    122
  1955     12      9       5       82.75    46.75    14.25    185
  1956      8      4       2       30.00    12.75     2.25     64
  1957      8      3       2       38.00    21.00     5.25     79
  1958     10      7       4       55.50    30.25     8.50    131
  1959     11      7       2       40.00    22.00     3.75     92
  1960      7      4       1       29.50    18.25     9.00     80
  1961     11      8       6       70.75    47.50    21.50    207
  1962      5      3       0       22.25    10.75     0.00     31
  1963      9      7       2       52.00    37.25     5.75    109
  1964     12      6       5       71.25    43.00     9.75    156
  1965      6      4       1       39.50    27.25     6.25     79
  1966     11      7       3       64.00    41.75     7.75    133
  1967      8      6       1       58.00    36.25     5.75     99
  1968      7      4       0       26.50    10.00     0.00     38
  1969     17     12       3       83.25    40.00     6.25    158
  1970     10      5       2       23.25     6.75     1.00     60
  1971     13      6       1       63.00    28.75     1.00     89
  1972      4      3       0       21.00     6.25     0.00     26
  1973      7      4       1       32.50    10.00     0.25     48
  1974      7      4       2       31.50    14.25     4.25     70
  1975      8      6       3       42.50    20.50     2.25     87
  1976      8      6       2       44.75    25.50     1.00     80
  1977      6      5       1       13.75     6.75     1.00     43
  1978     11      5       2       40.50    13.50     3.50     80
  1979      8      5       2       44.25    21.50     5.75     89
  1980     11      9       2       60.00    38.25     7.25    126
  1981     11      7       3       59.75    22.50     3.75    106
  1982      5      2       1       16.25     5.75     1.25     34
  1983      4      3       1       13.50     3.50     0.25     30
  1984     12      5       1       51.25    18.25     0.75     73
  1985     11      7       3       51.25    21.25     4.00    103
  1986      6      4       0       23.25    10.50     0.00     36
  1987      7      3       1       37.25     5.00     0.50     44
  1988     12      5       3       47.00    21.25     9.25    114
  1989     11      7       2       66.00    31.75     9.75    126
  1990     14      8       1       66.25    26.75     1.00     96
  1991      8      4       2       22.25     8.25     1.25     56
  1992      6      4       1       38.75    16.00     3.50     63
  1993      8      4       1       30.00     9.50     0.75     50
  1994      7      3       0       27.75     7.25     0.00     34
  1995     19     11       5      121.50    61.75    11.50    217
  1996     13      9       6       79.00    45.00    13.00    188
  1997      7      3       1       28.75     9.50     2.25     50
  1998     14     10       3       88.00    48.50     9.50    165
  1999     12      8       5       78.50    41.00    14.25    177
  2000     14      8       3       67.00    32.75     5.00    127
  2001     15      9       4       64.25    25.50     4.25    131
  2002     12      4       2       53.50    10.75     3.00     80
  2003     16      7       3       79.25    32.75    16.75    170

  AVG     9.9     6.0     2.3      50.0     24.4      5.3

                            ACTIVITY BY BASINS

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

  Activity for March:  No tropical cyclones


  SOUTH ATLANTIC (SAT) - Atlantic Ocean South of the Equator

  Activity for March:  1 hurricane

                            CYCLONE CATARINA
                              19 - 28 March

  A. Introduction

     According to many old meteorological textbooks, many things just
  don't happen.  In particular, essentially all books dealing with the
  global climatology of tropical cyclones unanimously state that tropical
  cyclones do not develop in the South Atlantic Ocean.  This assertion,
  however, was proven false 13 years ago when a strong tropical depression
  or possibly even minimal tropical storm formed just off the west
  African coast a few degrees south of the equator in April, 1991.  And
  just a couple months earlier, in January, what appeared to be a minimal
  tropical storm developed off the coast of Brazil around 15S.  Yet neither
  of these events grabbed the headlines as did a system which formed off
  the southeastern Brazilian coast in late March.  A system which had just
  about all the essential features of a tropical cyclone evolved out of
  an extratropical LOW and headed for the Brazilian coast.  Satellite
  intensity estimates indicated a strong Category 1 or weak Category 2
  hurricane on the Saffir/Simpson scale, and the widespread damage which
  occurred when the cyclone made landfall in the Brazilian state of Santa
  Catarina was consistent with a storm in this intensity range.

     A word about the names, numbers, and nomenclature for this cyclone.
  The name by which the storm came to be known--Catarina--appears to have
  first been utilized by the Brazilian press.   It seems likely that the
  designation "Furacao Catarina" was intended to mean the furacao (hurri-
  cane) threatening (Santa) Catarina (the state), but it was taken by the
  international press to mean a name for the storm in the sense in which
  tropical cyclones are normally named in the other oceanic basins.  The
  numerical designator '01T' was applied by the UK Meteorological Service
  when a number-plus-suffix was needed for the UKMET modeling program.
  The Monterrey NRL website for a time used the designator '01L' ('L' being
  the suffix normally applied for Atlantic tropical systems).  One other
  name should be mentioned.  Roger Edson of the University of Guam
  suggested the old Portuguese name Aldonca for the cyclone (before
  Catarina had become established as the somewhat official name), and this
  name was widely used for a couple of days by some discussion groups and

     All the available satellite data products and the damage incurred
  onshore more or less prove that Catarina was a cyclone of hurricane
  intensity.  But should Catarina be classified as a tropical cyclone?
  That is a question for which there is a significant divergence of
  opinion among the world's foremost tropical meteorologists.  This topic
  will be discussed further after the storm's synoptic history and damage
  have been covered.   For purposes of the header above, I chose to simply
  call the storm "Cyclone Catarina".   All parties in the tropical vs sub-
  tropical vs non-tropical debate would agree that Catarina was a cyclone.

  B. Storm Origins

     Much of the information in this section was taken from a discussion
  by Dr. Lance Bosart of the University at Albany/SUNY.  Lance's full
  synoptic discussion can be accessed at the following link:>

     Beginning around 12 March a quasi-stationary weak cold-core upper-
  level trough became established east of southern Brazil.  SSTs in the
  area were around 24-25 C and deep layer (850-200 hPa) vertical shear
  was generally less than 10 m/sec.  The upper-level cyclone was isolated
  from the westerlies by a persistent ridge that lay poleward of the
  system.  A series of troughs crossing the Andes during the period led
  to weak surface baroclinic cyclone formation over central Argentina.
  These systems would usually undergo modest re-intensification when they
  reached the eastern coast of South America and moved offshore.  Catarina
  formed from the last of these baroclinic systems which crossed the coast
  of South America and moved over SSTs of 24-25 C.

     The transition to a tropical-like system appeared to occur beneath a
  narrow ribbon of Amazonian moisture that turned westward and then north-
  westward from the main column of moisture that swept poleward along the
  eastern flank of the cut-off cyclone.   Otherwise, the environment over
  and to the west of the storm was quite dry in mid and upper levels.  The
  available evidence suggests that the precursor disturbance to Catarina
  was an ordinary, small-scale system that formed through ordinary baro-
  clinic processes within the envelope of a larger-scale upper-level
  trough.  Similar developments in the Northern Hemisphere have been
  observed to occur in conjunction with polar LOW formation beneath high-
  latitude cut-off cyclones, small-scale cyclone formation beneath cut-off
  cyclones over the Mediterranean Sea, and cyclogenesis beneath cut-off
  cyclones over the western Atlantic Ocean.

  C. Synoptic History

     Information relating to the track and intensity of Catarina in this
  section is based upon the track supplied by Roger Edson and which was
  included in the March global tropical cyclone tracks file prepared by
  the author.

     At 1800 UTC on 19 March a weak 25-kt LOW was located near the east
  coast of Brazil near the city of Florianopolis.  After an initial
  northerly jog (or relocation), the LOW moved in an east-southeasterly
  direction for the next three days, reaching a point approximately
  775 nm east-southeast of Florianopolis by 22/1800 UTC.  The maximum
  winds were estimated at 30 kts.  This was the easternmost point in the
  system's track--it subsequently began to move very slowly toward the
  west.  Gales were estimated to have developed by 23/1800 UTC, and the
  LOW began to exhibit increasing subtropical/hybrid characteristics on
  the 24th.  At 24/0000 UTC the LOW was located approximately 550 nm
  east-southeast of Florianopolis and was moving slowly but steadily
  westward.  Roger's track suggests that the system had become a 45-kt
  tropical storm by 0600 UTC on 25 March when it was centered roughly
  425 nm east-southeast of Florianopolis, and a minimal hurricane 24 hours
  later when located about 300 nm east-southeast of the same city.

     The estimated MSW reached 70 kts at 26/1200 UTC and then leveled off
  for about 18 hours before undergoing a modest intensification.  The
  system had by this time been dubbed Catarina by the Brazilian media,
  and, guided by a persistent ridge to the south, continued inexorably
  westward toward the coastline of Santa Catarina state.  As it approached
  the Brazilian coastline, Cyclone Catarina passed over a strip of slightly
  warmer SSTs and responded by intensifying yet further.  Roger estimates
  that the peak 1-min MSW reached about 85 kts shortly before landfall
  around 0600 UTC on 28 March.  Catarina's eye crossed the coast about
  25 nm north of Torres and just south of Ararangua.  The final data point
  of Roger's track at 28/1800 UTC places a weakening 45-kt tropical storm
  inland about 165 km north of the city of Porto Alegre.  Catarina began to
  rapidly weaken after making landfall in the manner of a normal tropical

  D. Meteorological Observations

     There were few meteorological observations available from near the
  center of Catarina.  The only strong wind report I'm aware of was an
  unofficial report of a gust to 82 kts, but the time and location and
  other particulars are unknown.   Based on a couple of reports from
  one coastal site near the landfall point and another from a station
  about 100 km inland, David Roth has estimated that the SLP in the eye
  at landfall was probably at least down to 991 hPa.

  E. Damage and Casualties

     In the state of Santa Catarina approximately 32,000 homes were
  damaged with 393 fully destroyed.  Two persons lost their lives with
  11 reported missing.  About 75 persons were severely injured, and over
  400 persons were injured after the storm while making repairs (from
  falls and other construction-related accidents).  The total damage
  was estimated at around US $350,000,000.  In the state of Rio Grande
  do Sul 31,500 people were adversely affected with 4500 homes damaged.
  Also, one health center and 16 schools were damaged, and 150,000 square
  metres of streets had to be re-paved.

     Agriculturally speaking, the banana crop in the region suffered
  an 85% loss, while other tropical fruit crops experienced losses of up
  to 60%.  Rice plantations experienced crop losses of around 40%.

     A special thanks to Alexandre Aguiar of Sao Leopoldo and Luiz Gava
  or Porto Alegre for sending the above information.  More information
  can be obtained from the following websites:>>>

     Additionally, Julian Heming of the UK Meteorological Office supplied
  a list of links containing further information on Catarina:>

  F. Discussion

     I have in my files many, many e-mails discussing the nature of Cyclone
  Catarina, i.e., whether or not the system in its later stages should be
  classified as a tropical cyclone.  I had intended to try to summarize
  many of the opinions and comments, but I have decided against that.  It
  would make the report unduly long, and furthermore, many of the comments
  were made by professional meteorologists on a non-public discussion list,
  and I don't have the time to try to contact all of them asking for
  permission to air their opinions in this summary, which is available to
  the public.   So I intend to just hit a few of the high points here, and
  perhaps quote a few persons whom I know would not mind.

     The debate over whether or not Catarina should be called a tropical
  cyclone at times seemed almost as stormy as the cyclone itself.  What is
  interesting is that almost everyone is in agreement over the salient
  characteristics of the system:

     (1) It was definitely of baroclinic origin, but once fully developed,
         appeared to be essentially completely isolated from any
         baroclinic systems.

     (2) It had well-organized convection surrounding a well-defined
         eye in the manner of a tropical cyclone, even though the
         convection was somewhat more shallow than what is typically
         seen in association with cyclones in the deep tropics.

     (3) It was warm-core in the mid and upper-levels.

     (4) It had a tight wind core similar to tropical cyclones.

     (5) It exhibited anti-cyclonic outflow.

     (6) It spent most of its life, especially the latter portion, moving
         over SSTs of around 24-25 degrees Celsius.

     (7) It definitely had winds exceeding hurricane intensity, as
         evidenced by the magnitude of damage when it moved onshore.

     Sound like a tropical cyclone?  Just about all U. S. tropical cyclone
  forecasters and researchers would (and did) answer that question with an
  unequivocal and resounding "yes".   But there were a few dissenting, at
  least questioning, voices also, primarily coming from the Australian
  sector of the globe.  What it really boils down to, in essence, is not
  so much a disagreement about the characteristics of Catarina, but about
  the application of the label 'tropical cyclone' to systems of baroclinic
  origin forming and moving over waters cooler than the traditional
  empirical threshold of 26 degrees Celsius.

     For more than 30 years, such systems in the North Atlantic basin have
  routinely been classified as tropical storms or hurricanes; hence, it is
  no surprise that the U. S. tropical cyclone community was all but
  unanimous in considering Catarina a tropical cyclone.  During its latter
  stages the cyclone certainly did not appear to be frontal, and with a
  warm core, well-organized moderately-deep convection, an eye and anti-
  cyclonic outflow, it possessed all the essential characteristics of a
  tropical cyclone as defined by TPC/NHC's operational policy.  Regarding
  the SST issue, systems forming over sub-27 C waters have many times been
  classified as tropical cyclones.  In November, 1980, small Hurricane
  Karl formed near the center of a large, weakening, occluded extratropical
  cyclone over SSTs near 20 degrees Celsius.

     The Australian experience, however, has been somewhat different.  In
  the Coral Sea and South Pacific waters off southern Queensland and
  northern New South Wales, many subtropical/hybrid and even extratropical
  systems have been noted to form at fairly low latitudes, i.e., a whole
  continuum between severe extratropical cyclones and severe classic
  tropical cyclones occurs.  Deciding at what point to make the decision to
  declare a named tropical cyclone has often proved to be very problematic.
  Jeff Callaghan points out that early on 28 March Catarina seemed to be
  straddling an increasing low to mid-level thermal gradient between a warm
  thermal HIGH over land to its southwest and a cold 700 to 500-hPa cold
  LOW near and northeast of the center.  Jeff further indicates that about
  all systems in the Brisbane AOR (tropical cyclones and otherwise)
  straddle such thermal gradients, but with tropical cyclones the shear
  and vertical tilt are usually much weaker.  In summary Jeff states that
  in his opinion Catarina was similar to the hybrid systems located towards
  the tropical cyclone end of the spectrum; operationally, it would likely
  have been named as a tropical cyclone.

     Greg Holland, a leading Australian tropical cyclone researcher, is not
  in favor of classifying Catarina as a tropical cyclone.   With Greg,
  however, it is not a question of the cyclone being frontal.  He sees a
  continuous spectrum of warm-core, convectively-driven cyclones ranging
  from polar LOWs through Mediterranean cyclones to hybrid systems like
  Catarina.  In his opinion, the term 'tropical cyclone' is best restricted
  to systems actually forming in the tropics over SSTs 26 C or higher.

     Generally speaking, NHC forecasters years ago rather liberalized the
  definition of a tropical cyclone to include systems of subtropical
  origin, as long as they were not frontal, in the interest of reducing
  confusion in public warnings.   Australia, on the other hand, seems to
  have somewhat restricted the definition of a tropical cyclone, likely
  after Dvorak analysis became the primary intensity estimation tool, in
  order to weed out some of the numerous hybrids (and possibly monsoon
  depressions) which were not handled too well by the Dvorak method.

     In the western Atlantic, 26 C SSTs extend northward to latitude 40N
  with even warmer waters in the Gulf Stream just off the U. S. East Coast,
  so hurricanes from the deep tropics often maintain their intensity to
  latitudes well outside the tropics.  So with systems of baroclinic origin
  sometimes forming over these same waters and acquiring most (if not all)
  of the features of cyclones forming in the deep tropics, it is not
  surprising that NHC forecasters in the main opted for classifying these
  systems as tropical cyclones, even if a few systems lacked some of the
  characteristics of classic tropical cyclones.  However, off eastern
  Australia true tropical cyclones rarely affect the coastline south of
  latitude 25S, and almost never beyond 30S, but hybrid storm systems tend
  to be rather abundant in those latitudes.  The subtropical storms,
  though, while occasionally capable of producing hurricane-force winds,
  rarely attain the intensity of even a moderately intense tropical
  cyclone, so it does seem to make sense to try to differentiate between
  those storms from the deep tropics which can upon occasion become very
  intense as opposed to those from the subtropics which rarely strengthen
  beyond storm intensity.

     I recently checked the official definitions of the term 'tropical
  cyclone' in several of the WMO regions, and there are subtle differences
  which, if interpreted very literally, would admit such systems as
  Catarina in some basins and not in others.   In short, there is no
  universally agreed-upon, detailed definition of a tropical cyclone;
  hence, it is not surprising that there would be major differences of
  opinion regarding a system like Catarina.

  G. Another Possible Subtropical System

     For an ocean which is not considered a tropical cyclone basin to have
  had two possible tropical cyclones in the same season is remarkable, but
  there was yet a third system which deserves mention.   This system
  occurred several days before the pre-Catarina LOW formed.  I received an
  e-mail from David Roth at HPC on the morning of 15 March regarding a
  LOW which was developing off southern Brazil.  To quote part of David's
  message:  "It looks subtropical in satellite pix due to the shear present
  and weak frontal tail extending NNW of the LOW despite the convection
  pulsing SE of the center (there is a broad, weak 5 C gradient across the
  frontal feature)...but when looking at 500 hPa the temps are up to -5 C
  and the SSTs are near 25 C, so it appears to be warm core.  It could
  merely be a sheared TC...but that weak front keeps me from thinking that
  it's fully tropical.

     "A buoy a bit south of the center has had sustained winds of 20-30 kts
  since midday Sunday (14 March), and pressures a bit west of the center
  are down to 1009 hPa at 1000 UTC, so the subtropical definition might be
  a better fit.  The satellite imagery looked best at 0300 UTC, when
  convection entirely covered the center and a weak feeder band attempted
  to form on the convection's north side."

  H. Link to Further Information

     Chris Velden at the University of Wisconsin and his associates have
  made available a webpage devoted to coverage of Cyclone Catarina.  Many
  satellite images, as well as graphs of data from other sensors, may be
  found.  The link to this outstanding compendium of Catarina data is:>

  (Report written by Gary Padgett)


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

  Activity for March:  No tropical cyclones

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

  Activity for March:  1 tropical storm **

  ** - Classified as a tropical storm by JTWC and PAGASA only

               Northwest Pacific Tropical Activity for March

     One tropical cyclone formed during the month of March in the Northwest
  Pacific basin.  JTWC assigned the designator 02W to the system when
  warnings were initiated, and PAGASA named the system Butchoy when it
  trespassed into that agency's AOR.  These also were the only TCWCs to
  upgrade 02W/Butchoy to tropical storm intensity--all the other Asian
  warning centres classified the system as only a tropical depression.  A
  report follows on Tropical Storm 02W/Butchoy, written by Kevin Boyle.  A
  special thanks to Kevin for his assistance.

                           TROPICAL STORM
                         (TC-02W / BUTCHOY)
                            16 - 23 March

  A. Storm Origins

     Tropical Storm 02W formed from an extensive area of cloudiness
  associated with a near-equatorial trough.  It was to become the third
  unnamed system on the trot, the previous being Tropical Storm 27W
  (December, 2003) and Tropical Storm 01W (February, 2004).  JTWC began
  issuing STWOs on a persistent area of convection within the trough at
  1130 UTC on 14 March located near 3.0N/142.6E, or approximately 460 nm
  southeast of Yap.  Earlier, at 14/0812 UTC, a QuikScat pass showed a
  very large and elongated LLCC consolidating under the large area of
  convection which was situated south of the circulation near the
  strongest winds.  The environment was assessed as moderately favourable
  with good diffluence and weak to moderate wind shear.

     The potential for development at this point was judged to be poor,
  but this was upgraded to fair at 14/1400 UTC.  The area of convection
  at this time was relocated to a position closer to Yap Island,
  approximately 350 nm to the southeast.  The development potential
  remained fair through the 15th, and then was upgraded to good at 16/0900
  UTC.   A TCFA was issued as animated multi-spectral imagery revealed a
  better-defined LLCC associated with the system.  The first warning on
  Tropical Depression 02W was issued three hours later.

  B. Storm History

     At 1200 UTC on 16 March Tropical Depression 02W was moving westward
  at 6 kts, located approximately 150 nm east-southeast of Palau.  The
  system continued to consolidate and was upgraded to tropical storm
  intensity at 16/1800 UTC.  The third warning (issued at 17/0000 UTC)
  placed the centre 18 nm east of Palau, but the combination of visible
  and microwave satellite images and synoptic data from Palau indicated
  that the LLCC was further to the southeast than previously advertised.
  Tropical Storm 02W was thus relocated to a position 150 nm southeast of
  Palau.  It was now moving somewhat faster (at 10 kts) towards the west-
  northwest, embedded in the steering flow of a mid-level ridge to the
  north.  However, the LLCC soon separated from its associated area of
  deep convection, and TS-02W was downgraded to a depression at 17/0600
  UTC.  The earlier relocation meant that Palau had to endure a second
  approach, but this time not so close.  By 17/1200 UTC the centre of
  the cyclone had passed the island to its south and was located 70 nm to
  the south-southwest.  (PAGASA had initiated bulletins by this time,
  naming the system Butchoy.)

     At 0000 UTC 18 March Tropical Depression 02W was moving west-
  northwestward at 17 kts from a position roughly 180 nm west of Palau.
  Animated multi-spectral imagery at this time depicted a complex system
  with up to three fully-exposed, weak LLCCs.  An 18/0413 UTC AMSU pass
  gave the best indication of where the main LLCC was located, near
  9.0N/130.0E, or approximately one degree southeast of the displaced deep
  convection.  The synoptic situation with TD-02W began to improve as the
  system turned northwestward, and after the convection began to redevelop
  over the LLCC, the MSW was upped to 30 kts at 1200 UTC, and further to
  35 kts (tropical storm status) at 18/1800 UTC.

     At 19/0000 UTC Tropical Storm 02W/Butchoy was moving northwestward
  some 425 nm east-southeast of Manila with 35-kt sustained winds.   At
  this time, animated multi-spectral imagery once again suggested the
  presence of multiple centres, and the 0000 UTC position was based on
  the LLCC nearest the deep convection.  Forward motion fluctuated from
  4 to 9 kts over the next couple of days as the difficult-to-pinpoint
  centre of TS-02W continued trekking to the northwest.  A 19/2202 UTC
  AMSU pass showed the LLCC to be exposed on the south side of the deep

     By 0000 UTC on 20 March Tropical Storm 02W/Butchoy had closed the
  distance to Manila down to 295 nm (to the east-southeast).  Initially,
  forecast models were predicting the system to cross the Philippines,
  but the 20/0000 UTC Prognostic Reasoning message called for a change
  to a northward track east of the Philippines, through a weakness, and
  finally recurving northeastward into the westerlies.  But before that
  was to happen, TS-02W had one last, final fling.  A TRMM pass at 20/1641
  UTC showed improved organisation of the deep convection and the formation
  of banding features.  The MSW was raised to 45 kts, and this was to be
  the peak intensity.

     Microwave imagery at 21/0000 UTC showed that the LLCC had become
  partially-exposed again, indicating that shear was increasing from the
  west.  Weakening had begun, and TC-02W was barely holding on to tropical
  storm status by 1200 UTC as it veered northeastwards, passing
  approximately 260 nm east-northeast of Manila, Philippines.  The MSW
  fell below 35 kts and TS-02W was downgraded to a depression at 21/1800
  UTC.  At 0000 UTC 22 March the centre of the ailing tropical depression
  was relocated to a position approximately 240 nm northeast of Manila
  with the nearest deep convection sheared over 125 nm away to the north-
  east.  Movement had become sluggish towards the north-northwest at 2 kts.
  As cool, dry air entrainment and high vertical wind shear were expected
  to continue, the final warning was issued by JTWC at 22/0600 UTC,
  locating the slow-moving and completely exposed LLCC of the dissipating
  storm approximately 220 nm northeast of Manila.

  (Editor's Note:  PAGASA was the only one of the Asian TCWCs to upgrade
  TC-02W to tropical status, assigning a peak MSW (10-min avg) of 45 kts.)

  C. Damage and Casualties

     No damage or casualties are known to have resulted from Tropical
  Storm 02W/Butchoy.

  (Report written by Kevin Boyle)


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

  Activity for March:  No tropical cyclones


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

  Activity for March:  2 tropical disturbances
                       2 severe tropical storms **
                       1 very intense tropical cyclone

  ** - Systems formed east of 90E and moved into basin

                        Sources of Information

     The primary sources of tracking and intensity information for
  Southwest Indian Ocean tropical cyclones are the warnings issued by
  the Tropical Cyclone Warning Centre on La Reunion Island, part of
  Meteo France (MFR), and the Regional Specialised Meteorological Centre
  for the basin.    However, tropical cyclones in this region are named 
  by the sub-regional warning centres on Mauritius and Madagascar with
  longitude 55E being the demarcation line between their respective
  areas of warning responsibility.  The La Reunion centre only advises
  these agencies regarding the intensity of tropical systems.  References
  to sustained winds imply a 10-minute averaging period unless otherwise

     In the companion tropical cyclone tracks file, I occasionally
  annotate positions from warnings issued by the Joint Typhoon Warning
  Center (JTWC) of the U. S. Air Force and Navy, located at Pearl
  Harbor, Hawaii, when they differ from MFR's coordinates by usually
  40-50 nm or more.  The JTWC warnings are also the source of the
  1-minute average maximum sustained wind values included in the
  tracks file.    Additionally, information describing details of
  satellite imagery and atmospheric circulation features included in
  the narratives is often gleaned from the JTWC warnings.

             Southwest Indian Ocean Tropical Activity for March

     After a quiet February, tropical cyclone activity returned to the
  Southwest Indian Ocean in March.  The most intense cyclone of the season,
  Gafilo, formed well to the northeast of Mauritius and trekked westward
  toward a rendezvous with Madagascar.  Striking the northeastern portion
  of the island at peak intensity, Gafilo was very destructive and deadly
  to the cyclone-prone island.  After reaching the Mozambique Channel,
  Gafilo re-intensified and struck Madagascar a second time along the
  western coast.   Severe Tropical Storms Helma and Itseng stirred up
  waters in the eastern extremity of the basin, both having moved in
  from the Australian Region where they were known as Nicky and Oscar,
  respectively.   (Reports on these two storms follow in the next section
  of this summary, covering the Northwest Australia/Southeast Indian
  Ocean region.)

     There were a couple of other disturbances numbered by MFR.  The first
  of these, Tropical Disturbance 11, certainly wins the award for being the
  most tenacious tropical system of the year.     According to Huang 
  Chunliang, who monitored this system on a daily basis, it was very likely
  a continuation of former Tropical Cyclone Evan, which formed on the first
  day of the month in the northwestern Gulf of Carpentaria.  Perth dropped
  bulletins on ex-Evan on 6 March while it was located off the Western 
  Australian coast after it had failed to re-intensify.  The daily Tropical
  Weather Outlooks began mentioning a tropical LOW on the 13th about 65 nm
  southwest of the Cocos Islands.  The LOW had crossed 90E by the 15th and 
  MFR began issuing sporadic bulletins on the system, numbering it as 
  Tropical Disturbance 11.  At 1200 UTC on the 15th it was located about 
  625 nm west-southwest of the Cocos Islands.   JTWC issued a TCFA for the
  disturbance at 0900 UTC on 16 March, but cancelled it a few hours later.

     MFR estimated the peak 10-min avg winds at 25 kts during this period,
  but by 1200 UTC on the 21st the system had weakened while located about
  80 nm east-northeast of Agalega and bulletins were dropped for a few
  days.    Interestingly, during the period when MFR was not issuing
  bulletins, JTWC upgraded the disturbance to TC-21S with the MSW (1-min
  avg) estimated at 30 kts.  Per JTWC's first warning, the center of TC-21S
  was located approximately 225 nm south of the Seychelles at 23/1800 UTC.
  However, a special warning (the third) was issued at 24/1200 UTC down-
  grading the system.   After JTWC had dropped the system, MFR on the 25th
  issued their first bulletin in four days as convection flared once more
  in the vicinity of the weak low-level center.  During this last gasp of
  Tropical Disturbance 11's life, the peak 10-min avg winds near the center
  were estimated at no higher than 20 kts.  The final MFR bulletin, issued
  at 0600 UTC on 28 March, placed the weak LLCC approximately 225 nm
  northeast of the northern tip of Madagascar.

     On the 31st of March, MFR issued two bulletins on a system located
  roughly 160 nm south-southeast of Agalega, designating it as Tropical
  Disturbance 13.  However, the system subsequently weakened and no more
  bulletins were issued.  A report on the very intense Tropical Cyclone
  Gafilo follows.

                       TROPICAL CYCLONE GAFILO
                          (MFR-09 / TC-16S)
                             2 - 15 March

  Gafilo: contributed by Madagascar

  A. Storm Origins

     An area of convection developed approximately 360 nm south of Diego
  Garcia on 29 February.  Animated enhanced infrared satellite imagery
  revealed cycling convection beginning to develop near an organizing
  LLCC.  An upper-level analysis indicated favorable divergence aloft
  but marginal vertical wind shear.  JTWC assessed the potential for
  development over the next 24 hours as poor.   This was upgraded to
  fair at 0400 UTC on 1 March when the system was relocated about 60 nm
  to the west of the previous day's position.  A 01/0114 UTC QuikScat
  pass depicted a well-defined LLCC with improving convective organization.
  A 200-mb analysis indicated that the disturbance was equatorward of an
  upper-level ridge axis with good diffluence and weak vertical shear.

     The system was relocated to a position about 300 nm south-southwest
  of Diego Garcia at 1800 UTC on 1 March.  MFR issued the first bulletin
  on Tropical Disturbance 09 at 0600 UTC on the 2nd, estimating the MSW
  (10-min avg) at 25 kts.  A couple of hours later JTWC issued a TCFA for
  the system since deep convection was continuing to become better
  organized.  The first JTWC warning on TC-16S, locating the center about
  740 nm east-northeast of Mauritius, was issued at 1800 UTC.  The system
  was moving west-southwestward at 12 kts, tracking along the equatorward
  periphery of a mid-level steering ridge located to the southeast.  MFR
  upgraded the disturbance to a tropical depression (30 kts) at 03/0000
  UTC, and at 03/0600 UTC Tropical Storm Gafilo was christened by the
  Meteorological Service of Mauritius.  Gafilo was located some 545 nm
  to the northeast of Mauritius with 40-kt winds, and the storm's motion
  had become west-northwesterly at a spiffy 20 kts.

  (NOTE:  According to Philippe Caroff, Chief Forecaster of Meteo France
  La Reunion, the name of this particular cyclone is correctly pronounced
  "gah-FEEL", i.e., the final 'o' is silent.)

  B. Storm History            

     The rapid west-northwestward motion continued through 3 March, but
  by 0600 UTC on the 4th Gafilo's translational speed had slowed to 7 kts
  and the storm had turned more to the west.   By 04/1800 UTC the cyclone
  was moving west-southwestward from a position 415 nm north of Mauritius.
  Intensification was steady once the system reached tropical storm
  status, and an AMSU pass at 04/0616 UTC revealed the presence of a small,
  symmetrical eye.  MFR upgraded Gafilo to tropical cyclone (hurricane)
  status at 1200 UTC on 4 March.  (JTWC had done so 18 hours earlier.)
  Throughout the 5th and 6th the rapidly intensifying cyclone trekked
  steadily in the direction of northern Madagascar.   Gafilo had reached
  the "intense cyclone" stage (winds 90 kts or higher) by 1800 UTC on
  5 May.  Twelve hours later the cyclone had reached its peak estimated
  intensity of 125 kts (10-min avg)--a "very intense cyclone" per MFR's
  terminology.  MFR estimated the minimum CP at a very low 895 hPa.
  (JTWC's peak 1-min avg MSW of 140 kts is in excellent agreement with
  MFR's reported peak intensity).  SSM/I imagery revealed a very distinct
  eye 20 nm in diameter during Gafilo's most intense stage.  Storm-force
  winds reached outward from the center 75 nm in all quadrants, and
  gales covered an area over 350 nm in diameter.

     The center of the very intense Gafilo made landfall near Antalaha,
  Madagascar, shortly after 0000 UTC on 7 March with winds near the peak
  intensity of 125 kts (10-min avg).  The storm's winds dropped sharply
  after landfall as the cyclone moved at a fairly good clip across
  northern Madagascar.   At 07/1800 UTC, MFR was reporting the intensity
  at 45 kts, but JTWC's 1-min MSW estimate was 75 kts.   By 1200 UTC on
  the 8th Gafilo's center was back over water in the Mozambique Channel
  roughly 200 nm west of Antananarivo.  The storm by then was tracking
  slowly south-southwestward along the western periphery of the mid-level
  steering ridge located to the southeast.   By 1800 UTC on 9 March Gafilo
  had rounded the western extremity of the ridge and was moving south-
  eastward at 10 kts back toward the southwestern coast of Madagascar.
  The storm had slowly intensified and the MSW had reached 60 kts by
  this time.  (JTWC upgraded Gafilo to 65 kts for a 24-hour period
  commencing at 08/1800 UTC.)  Severe Tropical Storm Gafilo made landfall
  a second time in Madagascar around 09/2300 UTC just north of the town of
  Morombe with peak winds estimated at 60 kts.   The 10/0000 UTC synoptic
  observation from Morombe reported sustained winds of 27 kts and a MSLP
  reading of 986.8 hPa.

     Gafilo's center meandered around over southern Madagascar for over
  three days.   MFR downgraded the system to a depression at 11/1800 UTC,
  and JTWC wrote their final warning at the same time.   The center of
  the weak tropical disturbance finally moved back over water around 0600
  UTC on the 13th, but conditions did not favor re-intensification, so
  MFR issued the final warning on Gafilo at 13/1200 UTC, placing the
  center approximately 375 nm west-southwest of Reunion Island.  However,
  warnings were revived 24 hours later on the system as a subtropical
  depression, located about 300 nm south-southeast of Reunion Island and
  moving rapidly east-southeastward.  The LOW, however, did not strengthen
  significantly and had begun to weaken again by the 15th.  The final
  bulletin at 15/0600 UTC placed the center about 550 nm south of Rodrigues

  C. Damage and Casualties

     Tropical Cyclone Gafilo was extremely destructive to Madagascar.  A
  report dated 30 March 2004 on the ReliefWeb internet site gave the death
  toll at 237 with 181 missing.  A ferry en route from the Comoros Islands
  to Madagascar during the time the cyclone was located off northwestern
  Madagascar never arrived and was presumed sunk.  The ferry was carrying
  113 passengers and crew, but it is not clear if these are included in
  the above quoted death toll or number of missing persons, or either.
  A majority of the deaths occurred in associated with extensive flooding
  caused by the torrential rains.  Over 300,000 hectares of productive
  agricultural land were damaged, and an estimated 150,000 hectares of
  crops were lost.  The most heavily damaged crops were rice, corn and
  vanilla.  More than 200 schools and 200 health centers were either
  damaged or destroyed.

     The city of Majunga on the northwestern coast suffered heavy damage
  and facilities along the waterfront were destroyed.  The city reported
  peak gusts of 88 kts during the storm.  Morondava, a city prone to
  coastal flooding, was almost 100% destroyed.  Approximately 95% of
  Antalaha was demolished, and the town of Maroantsetra located on
  Antongil Bay was inundated.   Gafilo passed overland to the south of
  Majunga, but the storm maintained intensity rather well in passing
  across the island.  Morondava is located along the central western
  coast of Madagascar near where Gafilo made its second landfall.  However,
  Antalaha is near the northeastern coast where the cyclone made its
  initial and most intense landfall.   This area suffered greatly in April,
  2000, from the intense Tropical Cyclone Hudah, and was also affected
  by intense Tropical Cyclone Hary in March, 2002, but with relatively
  minor damage since that particular cyclone only brushed the coast and
  the more intense left semicircle remained offshore.

     One press report mentioned wind gusts of 179 kts in association with
  Gafilo, but no information was given regarding time, location, and
  whether or not they were recorded or estimated.     In the author's
  opinion, they were likely not reliably recorded, since this would indeed
  probably be a world record for winds recorded in a tropical cyclone.
  Another press report stated that wind gusts of 135 kts were recorded.
  This value seems much more believable, although no information was given
  regarding time and location.

     Many articles about the effects of destructive Tropical Cyclone
  Gafilo may be found at the following link:>

  (Report written by Gary Padgett)



  Activity for March:  1 tropical cyclone
                       2 severe tropical cyclones (hurricanes)

                        Sources of Information

     The primary sources of tracking and intensity information for
  Northwest Australia/Southeast Indian Ocean tropical cyclones are 
  the warnings and advices issued by the Tropical Cyclone Warning
  Centres at Perth, Western Australia, and Darwin, Northern Territory. 
  References to sustained winds imply a 10-minute averaging period
  unless otherwise stated.

     In the companion tropical cyclone tracks file, I occasionally
  annotate positions from warnings issued by the Joint Typhoon Warning
  Center (JTWC) of the U. S. Air Force and Navy, located at Pearl
  Harbor, Hawaii, when they differ from the Australian centres' coor-
  dinates by usually 40-50 nm or more.  The JTWC warnings are also the
  source of the 1-minute average maximum sustained wind values included
  in the tracks file.   Additionally, information describing details of
  satellite imagery and atmospheric circulation features included in
  the narratives is often gleaned from the JTWC warnings.

                Northwest Australia/Southeast Indian Ocean
                       Tropical Activity for March

     Three tropical cyclones traversed waters off Western Australia and
  the Southeast Indian Ocean during March.  Two of these became intense
  tropical cyclones (peak 10-min avg winds >= 90 kts) while the other
  almost reached hurricane intensity (it was considered a minimal hurri-
  cane by JTWC).  Tropical Cyclone Nicky formed in the western extremity
  of Perth's AOR and had almost reached hurricane status by the time it
  crossed 90E into the Southwest Indian Ocean and was re-named Helma.
  Severe Tropical Storm Nicky-Helma moved generally southward and main-
  tained intensity for a couple of days before rapidly weakening.

     Severe Tropical Cyclone Oscar formed during the fourth week of the
  month and moved harmlessly westward over the open South Indian
  Ocean.   Oscar peaked at 95 kts (110 kts 1-min avg per JTWC) while
  still in the Australian Region, but had begun weakening rapidly by the
  time it crossed into the Southwest Indian Ocean basin and was renamed
  Itseng by the Meteorological Service of Mauritius.  The third cyclone,
  Fay, was named by the Darwin TCWC but spent most of its life in Perth's
  AOR.  Severe Tropical Cyclone Fay was the most intense cyclone of the
  season in the entire Australian Region, reaching Category 5 status on the
  Australian scale.  Fay moved slowly and erratically off the coastline
  of Western Australia, and weakened to a Category 2 cyclone at one point,
  but recovered and regained Category 4 status before making landfall
  along a sparsely-inhabited stretch of coastline.   Heavy rains attending
  Fay's landfall caused widespread flooding in portions of Western

     Reports follow on Nicky-Helma, Oscar-Itseng, and Fay, all authored
  by Simon Clarke.

                          (TC-17S / MFR-10)
                             8 - 13 March

  A. Storm Origins

     Nicky was the fifth tropical cyclone of the 2003/2004 season named by 
  the Perth Tropical Cyclone Warning Centre.  While Nicky formed in the 
  Australian basin, it moved into the Southwest Indian Ocean basin where, 
  according to convention, it was renamed Helma.

     Nicky-Helma was first detected on 8 March 2004 within a persistent 
  area of convection.  QuikScat imagery depicted an organised LLCC 
  approximately 370 nautical miles west of the Cocos Islands (near 
  13.4S/90.7E).  The system was located within an environment of good 
  upper-air diffluence and moderate wind shear.  The initial depression 
  was slow and erratic in movement due to the surrounding weak steering 
  environment.  By 09/0400 UTC the developing tropical LOW was located 
  near 13.2S/90.7E and had deepened to 995 hPa.  At this time the Perth 
  TCWC officially named the system Tropical Cyclone Nicky.     

  B. Storm History

     Nicky soon commenced a southwesterly path along the western periphery 
  of a low to mid-level ridge located to its southeast.  This south-
  westerly path at a pace of between 7 and 10 knots was to be maintained
  more or less for the remainder of the cyclone's life.  Initially, upper-
  level wind shear slackened, allowing Nicky to slowly intensify.  By
  10/1800 UTC Nicky had passed into the Southwest Indian Ocean basin near
  16.5S and was renamed Helma.  Peak intensity was attained at 11/0600
  UTC with Helma (972 hPa) located near 17.8S/88.1E.  Meteo France in La
  Reunion estimated the maximum 10-min avg winds at 60 knots, while JTWC's
  peak 1-min avg MSW was 70 knots.

     Thereafter, increasing vertical wind shear, constricting upper-air 
  ventilation and cooler sea surface temperatures steadily weakened 
  Helma.  By 12/1800 UTC Helma's LLCC was exposed on the northwestern 
  edge of the deep convection.  Increasing westerly wind shear prevented
  any further re-coupling of the LLCC with the deep convection and by
  13/0600 UTC, Helma had lost tropical cyclone status approximately
  1000 nautical miles southwest of the Cocos Islands (near 22.7S/84.2E).

  C. Damage and Casualties

     Nicky-Helma remained over the open ocean for its entire life.  The 
  Perth and La Reunion warning centres issued warnings for shipping. 
  However, there are no known reports of any incidents arising from the 

     Satellite imagery of the system can be found at the following web-

  (Report written by Simon Clarke)

                     SEVERE TROPICAL CYCLONE FAY
                            14 - 28 March

  A. Storm Origins

     Fay was the second tropical cyclone of the 2003/2004 season to be 
  named by the Darwin Tropical Cyclone Warning Centre.  However, Fay 
  moved into Perth’s area of responsibility soon after naming and it 
  was in Western Australia that Fay had its greatest impact, reaching 
  Category 5 status on the Australia scale with wind gusts estimated to 
  be as high as 163 knots. 

     The pre-cursor of Fay was detected as early as 14 March 2004 in the 
  northern Gulf of Carpentaria as a tropical disturbance moving towards 
  the west at approximately 10 knots.  After skirting offshore parallel 
  to the Top End of the Northern Territory, the tropical LOW had reached
  the Timor Sea to the west of Darwin by 16 March 2004.  Deep convection
  had increased but remained at its strongest to the northwest of, and
  slightly removed from, the LLCC.    Under a favourable upper-level
  environment of low vertical wind shear and fair diffluence, the tropical
  LOW was officially named Fay at 16/1200 UTC.

     At this time, Fay was located approximately 180 nautical miles west- 
  northwest of Darwin and 170 nautical miles north-northeast of Kalumburu
  (near 11.7S/127.9E). The system was travelling in a west-southwesterly
  direction at 6 knots along the northwestern periphery of a mid-level
  steering ridge located to the southeast.

  B. Storm History

     Following naming, Fay travelled toward the southwest closer to the 
  Kimberley coast.    However, the cyclone turned towards the west on
  18 March (prior to making landfall) as it was steered by a mid-level
  ridge to the south.

     Fay then drifted out into the Indian Ocean and intensified at a near 
  climatological rate under favourable conditions of moderate upper-level
  outflow and weak vertical wind shear.   Fay reached its peak intensity
  of 910 hPa at 21/1000 UTC while located approximately 150 nautical miles
  north-northwest of Cape Leveque (near 14.1S/122.4E) with the peak MSW
  estimated at 115 knots (120 knots 1-min avg per JTWC) near the centre.
  Around this time the cyclone became quasi-stationary as a passing short-
  wave trough began to weaken the mid-level steering ridge to the south. 

     Peak intensity was sustained only briefly.  Fay abruptly turned to 
  the south-southeast toward the west Kimberley coast as a mid-level 
  trough extending northwest to southeast over central Australia filled 
  and the steering ridge to the northeast began to dominate.  This path 
  was to be maintained for almost 48 hours, during which time Fay 
  weakened to a category 2 cyclone on the Australian scale as dry air 
  originating from the Great Sandy Desert was entrained into the system.  

     On 25 March animated water vapour imagery revealed that a poleward 
  outflow channel extending to the southeast of Fay was counterbalancing
  the effects of the continental dry air entrainment.  Accordingly, the
  cyclone commenced re-intensification, reaching category 3 status as it
  passed within 50 nautical miles of Broome before veering away on a more
  west-southwesterly track.   At 25/1000 UTC Fay was estimated to be
  75 nautical miles north-northeast of Pardoo and 115 nautical miles
  northeast of Port Hedland (near 18.9S/119.9E) and had re-intensified
  into a Category 4 cyclone.  Fay was about to commence on a track to the
  south at 3-5 knots which would be maintained until landfall.  Fay's
  coastal crossing was in a remote part of the coastline between the
  pastoral stations of Pardoo and Wallal at approximately 26/2200 UTC.
  Fay possessed an estimated central pressure of 940 hPa just prior to
  landfall with estimated peak 10-min avg winds of 90 knots (115 knots
  1-min avg per JTWC) near the centre. 

     After making landfall, the cyclone began weakening as it moved to
  the south-southeast farther inland, passing close to the Yarrie Mine.  
  The cyclone eventually lost cyclone status between Nullagine and 
  Telfer (21.6S/121.1 E) at 28/1000 UTC, approximately 36 hours after  
  moving onshore.

  C. Warnings

     The Darwin and Perth Tropical Cyclone Warning Centres issued a total 
  of 109 Tropical Cyclone advices for Fay collectively.  Advisories were
  discontinuous as Fay reached peak intensity well out in the Indian
  Ocean.  Shipping forecasts were issued on a regular basis, however,
  during this period.

  D. Meteorological Observations

     The Perth Bureau of Meteorology has issued a report on Fay that can 
  be found at:>

  This report provides a snapshot of the cyclone, including pictorial 
  representations of Fay’s track, satellite and radar imagery.  As Fay 
  did not pass over any wind or pressure recording sites during its 
  life, other observations within this report are limited.  

  E. Damage and Casualties

     The body of an SES volunteer was found in a remote West Australian 
  gorge after he was swept to his death in a flash flood while trying 
  to save an injured tourist.  This was the only reported casualty as a 
  result of Fay.

     There was little damage of any consequence.  The Manager of the 
  Kooljaman resort at Cape Leveque on the Peninsula reported to local 
  media that some of the resort's camping accommodation had been damaged.
  Strong winds uprooted between 50 and 100 trees with some of the fallen
  trees blocking the access road to the resort.

     The resort town of Broome was all but shut down as the cyclone passed 
  by.    Schools, shops and businesses were closed and flights were
  cancelled.  Broome experienced strong winds with gale-force gusts, some
  heavy rain and heavy seas but escaped serious damage.    The cyclone
  uprooted trees and closed roads in the region, but left little in the
  way of structural damage.

     The BHP Billiton iron ore mine at Yarrie, 150 kilometres east-
  southeast of Port Hedland, was shut down during the passage of the 
  cyclone.  Some 200 workers were locked down for 8 hours in two squash 
  courts as accommodation units were overturned, water tanks "shredded" 
  and power lines cut as the cyclone passed by.

     In the 24 hours to 0100 UTC on 28 March 2004, heavy rain was recorded 
  in the Pilbara, especially in the De Grey River catchment southeast 
  of Port Hedland.  Highest registrations included 134 mm at Warrawagine,
  230 kilometres east-southeast of Port Headland, and 111 mm at Mandora,
  on the coast 250 kilometres east-northeast of Port Hedland.    Wallal
  Downs, about 30 kilometres from Mandora, recorded 197.6 mm in the
  48 hours to 0100 UTC.  These falls are considered to represent the best
  rains in four years, if not a decade, in central and western parts of
  the Pilbara and Gascoyne.

     At Nullagine, 150 kilometres farther south, flooding in the Nullagine 
  River split the town into four sections, resulting in the evacuation 
  of the town's population of 140 to the police station, court house 
  and buildings on the outskirts of the town.  Heavy rain in the Oakover
  and Nullagine River catchments produced moderate flooding in the De Grey
  River. The wind uprooted trees and disrupted phone and power services
  throughout the area.

     At the time of writing this report, there was no information 
  available regarding stock losses.  However, the accruing benefits of 
  the useful rain will likely outweigh any immediate short-term losses.

  F. Further Information / Web Links

     Further information, including satellite imagery, track details and 
  photographs of the event can be found at the following web-links:

     Courtesy to Laurier Williams for part of the material contained 
  within this report. For further information please refer to Laurier 
  Williams excellent web page at:>

  (Report written by Simon Clarke)

                          (TC-20S / MFR-12)
                            20 - 28 March

  A. Storm Origins

     Oscar was the sixth tropical cyclone of the 2003/2004 season to be 
  named by the Perth Tropical Cyclone Warning Centre.  Following on 
  from Nicky two weeks earlier, Oscar also moved into the Southwest 
  Indian Ocean basin where it was renamed Itseng.

     Oscar-Itseng was first detected as a westerly moving depression on
  22 March 2004, located approximately 285 nautical miles east-southeast 
  of the Cocos Islands (near 14.9S/100.9E).  At the time, animated 
  infrared and microwave imagery depicted deep convection cycling over 
  a well-defined LLCC.  Upper-level conditions were somewhat favourable 
  with weak to moderate vertical wind shear and good diffluence aloft. 
  By 23/2200 UTC, the developing depression was located near 14.8S/97.4E 
  and had rapidly deepened to 985 hPa.  It was officially named Tropical
  Cyclone Oscar by the Perth TCWC at this time.

  B. Storm History

     Oscar continued on a 6-knot westerly path under the steering 
  influence of a mid-level ridge to the south while steadily 
  intensifying in a favourable environment consisting of warm sea 
  surface temperatures and good upper-level outflow in the poleward 
  direction.   A banding eye formed as the cyclone passed approximately 
  130 nautical miles to the south of the Cocos Islands on 25 March. 

     Peak intensity was attained at 25/2200 UTC with Oscar (935 hPa) 
  located approximately 375 nautical miles southwest of the Cocos 
  Islands (near 17.0S/92.1E).  Perth estimated the maximum 10-min avg 
  winds at 95 knots while JTWC's peak 1-min avg MSW was 110 knots.  
  Satellite imagery depicted a well-defined eye.  At this time, Oscar 
  was moving toward the south-southwest at approximately 5 knots in 
  response to a migratory short-wave trough that had weakened the 
  mid-level steering ridge to the south. 

     A mid-level ridge to the east eventually recurved Oscar toward the 
  south.  However, by 27/1800 UTC Oscar had edged sufficiently to the 
  west to move into the Southwest Indian Ocean basin (near 19.4S). 
  Accordingly, Oscar was renamed Itseng.  By this time Itseng had 
  encountered moderate to strong upper-level wind shear and cool sea 
  surface temperatures and as a consequence had weakened dramatically. 
  The cyclone lost deep convection over its LLCC and was soon 
  downgraded below cyclone status at 28/1200  (near 19.7S/89.2E).  The 
  spectacular loss in structure is depicted at the following weblink:

  C. Damage and Casualties

     As with Nicky-Helma earlier in March, Oscar-Itseng remained in open 
  ocean for its entire life.  The Perth and La Reunion warning centres 
  issued warnings for shipping.  However, there are no known reports of 
  any incidents arising from the cyclone.

     Satellite imagery of the system can be found at the following web-

  (Report written by Simon Clarke)



  Activity for March:  2 tropical cyclones
                       1 significant hybrid storm

                     Northeast Australia/Coral Sea
                      Tropical Activity for March

     The fairly inactive (at least in recent years) portion of the Southern
  Hemisphere lying between east longitudes 135 and 160 produced three
  rather noteworthy systems during the month of March, two of which brought
  severe weather and/or very heavy rainfall to the coasts of Queensland
  and northern New South Wales.  A subtropical hybrid-type storm early in
  the month brought damaging winds, heavy rainfalls and high seas to the
  coastlines of the aforementioned states.   A few weeks later, another
  storm system took shape right along the tropical Queensland coast,
  bringing extremely heavy rainfalls and gale-force winds.  After moving
  away from the coast, this LOW developed enough central convection and
  typical tropical cyclone features to be designated Tropical Cyclone
  Grace.  After earning a name, Grace sped away from Australia, entering
  the Fiji AOR east of 160E shortly before being downgraded.

     The Gulf of Carpentaria was the spawning ground for a tropical LOW
  which became a tropical cyclone.     A LOW began showing signs of
  strengthening on the last day of February, and on 1 March became the
  ephemeral Tropical Cyclone Evan for a few hours before making landfall
  in the Northern Territory.  Evan's remnant LOW re-emerged into the Timor
  Sea and was expected for a few days to re-intensify into a tropical
  cyclone, but this never materialized.  The system continued westward
  across the South Indian Ocean, characterized by occasional convective
  flare-ups.  There is a possibility that the brief TC-21S in late March
  northeast of Madagascar was related to the remnants of Evan.  Another
  tropical LOW formed in the Arafura Sea around mid-month and drifted
  westward north of the Top End.  This system ultimately became Severe
  Tropical Cyclone Fay.  (A report on Fay will appear in Part 2 of the
  March summary.)

     A special thanks to Simon Clarke, who was near the center of action
  for the unnamed hybrid storm, for writing the report on that system, as
  well as the summaries for Tropical Cyclones Evan and Grace.  Also, many
  thanks to Jeff Callaghan for sending observations and damage reports
  related to these systems.

                         TROPICAL CYCLONE EVAN
                        29 February - 06 March

  A. Storm Origins

     Tropical Cyclone Evan, the second cyclone to be named by the Darwin 
  Tropical Cyclone Warning Centre, briefly attained tropical cyclone status
  in the Gulf of Carpentaria.  The initial depression formed in the north-
  eastern Gulf on 29 February from a broad area of monsoonal activity
  stretching from Tropical Cyclone Monty off the Pilbara Coast of Western
  Australia across the Cape York Peninsula, Queensland, and into the
  northern Coral Sea.

     The tropical depression (TC-15P per JTWC) maintained a general west- 
  southwesterly track with convection initially displaced slightly to 
  the west of the LLCC.   By 01/0630 UTC and despite continuing shear, 
  organization improved sufficiently for the depression to be upgraded
  to tropical cyclone status.

  B. Storm History

     At the time, the newly-named Evan was centred in the Gulf of 
  Carpentaria about 70 nautical miles east of Alyangula and 110 
  nautical miles south-southeast of Nhulunbuy (13.9S/137.5E) and moving 
  west-southwestward at 10 knots.  Evan maintained a central pressure 
  of 994 hPa with a MSW (10-min avg) of 40 knots until landfall 
  approximately three hours later on the eastern coast of Groote 

     After crossing Groote Eylandt, Evan maintained a westerly path, 
  crossing the mainland Northern Territory coastline to the near north 
  of Numbulwar at 01/1530 UTC.  Evan rapidly weakened to below tropical 
  cyclone intensity, but the remnant depression continued to track 
  westwards across the Top End at (or near) 10 knots, re-emerging over 
  water in the Joseph Bonaparte Gulf near Port Keats by 03/0030 UTC. 
  Evan moved overland again, clipping the far northern Kimberley 
  coastline of Western Australia, before final re-emergence into the 
  Indian Ocean at 04/0030 UTC.

     Warnings were issued by the Perth Tropical Cyclone Warning Centre 
  for ex-tropical cyclone Evan at this time.  However, the system
  remained weak and poorly organised, suffering from dry air entrainment
  that effectively stifled any further opportunity for redevelopment.
  Convection was lost and the remnant tropical LOW moved out across the
  Indian Ocean, showing varying degrees of convective bursts without
  ever regaining sufficient momentum for redevelopment as a tropical

  C. Damage and Casualties

     There were no casualties as a consequence of Evan and any damage 
  reported was inconsequential, being confined primarily to power loss 
  and flooding on Groote Eylandt.  As Evan crossed Groote Eylandt, the 
  storm dumped a record 316 mm of rain on the island.  The island's 
  previous 24-hour rainfall record was 158 mm.  (Not a BoM source--from 
  EO Natural Hazards Website.)

     Evan's main impact was to contribute further to the Australian Top 
  End's very much above-average wet season rainfall.  The local media 
  reported that the Stuart Highway south of Katherine was closed for 
  several days due to these rains, and several businesses and homes 
  experienced localised flooding. 

  (Report written by Simon Clarke)

                         SUBTROPICAL STORM
                            2 - 5 March

  A. Storm History

     An unusual event was to unfold in the Coral Sea during the first
  week of March, 2004, as a tropical LOW formed in the northern Coral Sea.
  The LOW subsequently moved toward the south-central Queensland coast as 
  a hybrid system induced by a 500-hPa LOW near the Tropic of Capricorn. 
  This was not to be a typical tropical cyclone, but rather a sheared-type
  hybrid with storm-force winds in its southern quadrant.

     The first gale warning was issued by the Bureau of Meteorology, 
  Queensland, at 02/0648 UTC as a 1002-hPa tropical LOW near 14S/154E 
  slowly deepened in conjunction with a developing pressure gradient 
  associated with a large HIGH located over the Tasman Sea.  By 03/0149 
  UTC, a complex area of low pressure with two centres had become
  established in the Coral Sea with one LOW (1000 hPa) located near
  14S/151E and another LOW (1002 hPa) located near 18S/159E. 
     These two LOWs gradually interacted with each other as a trough dug 
  southwards through the Coral Sea into the strong high-pressure system 
  to the south.  Eventually a new 999-hPa low-pressure centre developed 
  on the trough line at 04/1334 UTC near 20.5S/157E (approximately 380 nm 
  northeast of Fraser Island).  This LOW then commenced a southwestward
  motion at approximately 22 kts while rapidly intensifying.  Satellite
  imagery at the time clearly showed a south to east bias in the upper-
  level cloud structure, indicating a hybrid system with tropical
  characteristics coupled to an upper-level system of more mid-latitude

     At the time, Jeff Callaghan of the Bureau of Meteorology, Queensland,
  explained to the local media that SSTs of around 28 C were fuelling the 
  storm and that it was being steered by an upper-level LOW centred near 
  Charleville, Queensland, giving the storm a fairly predictable south-
  westerly path towards the Fraser Island area.   Also at this time,
  gales became established across the entire southern Coral Sea to
  New Caledonia.

     The subtropical storm was to remain a complex system and by 05/0600 
  UTC, the primary 994-hPa low was located near 23.0S/152.5E and moving 
  southwestward at about 20 kts, producing gales and storm-force winds in 
  the open east coast waters of Australia between Bowen and Coolangatta.
  The storm eventually crossed the coast at approximately 05/0012 UTC in
  the Hervey Bay area of southeast Queensland before tracking inland to
  the west of Brisbane and dissipating soon thereafter.

     In response to the serious nature of the storm, and in an unusual 
  move for Queensland, the emergency signal sirens normally assigned 
  exclusively to tropical cyclones were used in association with the 
  Bureau of Meteorology's Severe Weather Warnings issued for this sub-
  tropical system as it approach the south Queensland coastline.

     To add to this summary, the following comments are provided, courtesy 
  of Jeff Callaghan, in the following (slightly edited) insight:

     "Dynamically this system was developing in a similar fashion to a 
  severe East Coast LOW except for the fact it was moving over waters 
  where the sea surface temperatures were 28 C.  Often with these 
  systems QuikScat shows very tight circulations with at least storm- 
  force winds near the centre and gales right around the centre.  If 
  this were to have happened we (i.e., BoM Brisbane) would have 
  considered naming it as a tropical cyclone.  However, because of the 
  strong LOW near Willis Island, it developed as a very elongated system
  with no tight focus, and as a result the strongest winds were well-
  removed from the centre and so Severe Weather Warnings appeared to be
  the way to go."

  B. Meteorological Observations
     Jeff Callaghan also provided the following detailed material with 
  respect to observations:

  (1) Waves and Storm Surge

     Very large long-period waves were generated by the storm.  Waves with 
  significant wave heights of almost 6 m and maximum heights of almost 
  12 m were recorded on the Mooloolaba wave rider buoy around 0600 UTC on
  5 March.  On the buoy off Stradbroke Island the corresponding readings
  were 7.1 m and 14.3 m around 05/0700 UTC, coming from the east-northeast.
  The peak height measured on the Tweed Buoy was just over 14 m and this
  occurred at 1700 UTC on 6 March, when the significant wave height was
  between 5 and 6 metres.  The wave periods were almost 13 seconds and 
  approached from the east-northeast. 

     Even in Moreton Bay a maximum wave height of over 3.5 metres was 
  measured around 1500 UTC on 5 March.

    The worst effect from storm surge appears to have been in association 
  with the Friday night's high tide at Currumbin on the Gold Coast when
  cars were floating around the car park.  This was not a big tide (0.74 m
  below the Highest Astronomical Tide--HAT) and therefore the wave effects
  appear to have added over a metre to the tide. 

     Northeasterly gales in Moreton Bay early on Saturday raised the level 
  of the Bay 0.6 metres.  The winds eased somewhat but still water levels
  almost reached HAT on the 05/2347 UTC.    This, combined with waves,
  caused inundation of low-lying areas.

  (2) Wind Observations

     Cape Moreton's (WMO 94594) strongest winds were at 05/0500 UTC-- 
  130/56 knots (10-min mean) and peak gust 67 knots.  Note these Almos 
  automatic weather stations (AWS) appear to have very low gust factors 
  compared with the old Dynes anemometers.  Cape Moreton reported storm- 
  force winds from 0057 to 0530 UTC on 5 March.

     Double Island Point (WMO 94584) also reported storm-force winds from 
  0130 to 0330 UTC on 5 March from the south-southeast.  The maximum 
  10-minute mean wind was 49 knots.

     The Gold Coast Seaway reported south-easterly gales from 0400 to 1000 
  UTC on the 5th with maximum gusts to 50 knots, followed by easterly gales
  from 05/1230 to 1600 UTC.

     Moreton Bay South (AWS) reported southeasterly gales from 0555 to 
  0718 UTC on 5 March with maximum gusts to 49 knots, followed by east- 
  northeasterly gales around 05/1500 UTC.
     Moreton Bay Central (AWS) reported southeasterly gales from 0200 to 
  0728 UTC on 5 March with maximum gusts to 56 knots.  Easterly gales were 
  felt from 05/0949 to 05/1217 UTC, then turning north-northeasterly from
  05/1411 to 05/1530 UTC.  Maximum gusts on the 5th were 51 knots.

     Moreton Bay North (AWS) reported southeasterly gales from 0304 to 
  0630 UTC on 5 March with maximum gusts to 46 knots. 
     Redcliffe (AWS) reported south to southeasterly gales from 05/0300
  to 05/0733 UTC with maximum gusts to 49 knots, followed by easterly 
  and northeasterly gales from 1400 to 1522 UTC.

     Heron Island (AWS) reported gales from the south-southwest from 0035  
  to 0600 UTC on 5 March with maximum gusts to 49 knots. 

     Rundle Island (AWS) reported southerly gales from 04/2130 to 05/0200
  UTC with maximum gusts to 50 knots. 
     Cato Island (WMO 94394) AWS reported southeasterly gales from 1300 to
  1800 UTC on 4 March.

     Frederick Reef (WMO 94393) AWS reported southeasterly gales from 0500
  to 1500 UTC on the 4th with a maximum 10-minute mean wind of 39 knots. 

     Gannet Cay (WMO 94379) AWS reported south to southwesterly gales from 
  04/2200 to 05/0200 UTC with a maximum 10-minute mean wind of 42 knots.

  (3) Rainfall Observations

     Twenty-four hour rainfall totals from 04/2300 to 05/2300 UTC

  TOMEWIN                   284.0 mm 
  SPRINGBROOK               259.0 mm 
  MALENY                    239.0 mm 
  O'REILLYS ALERT           236.0 mm 
  TAMBORINE                 226.0 mm 
  MT MEE                    225.0 mm 
  MT GLORIOUS               221.6 mm 
  TALLEBUDGERA CK           197.0 mm 
  CANUNGRA                  196.2 mm 
  HINZE DAM                 194.0 mm 
  MT NEBO                   190.6 mm 
  COOROY                    182.6 mm 
  FERNY HILLS               180.0 mm 
  LAKE MANCHESTER           176.0 mm 
  BEERBURRUM                169.6 mm 
  STRATHPINE                169.0 mm 
  MORAYFIELD                164.2 mm 
  EUMUNDI                   161.0 mm 
  WOODFORD                  157.0 mm 
  HIGHVALE                  155.2 mm 
  PETRIE                    154.0 mm 
  MITCHELTON                150.0 mm 
  EVERTON HILLS             148.0 mm 
  REDCLIFFE                 144.0 mm 
  SAMFORD                   134.0 mm

  C. Casualties and Damage

     To this end, up to 106,000 customers lost power during the passage
  of the subtropical storm through the greater Brisbane Metropolitan area,
  the Sunshine Coast, and the Gold Coast with tree and associated minor 
  structural damage reported throughout the area and extending into 
  northern New South Wales.

     Tragically, at the time of reporting two casualties are known as a 
  consequence of the subtropical storm (with possibly another person
  missing) with all of the deaths associated with drowning in flooded
  waterways.  Other damage reports from local State Emergency Service
  included minor roof damage and sandbagging requests.

     Police also reported some cars bogged in flooded roadways.  The 
  Warrego Highway at Cunningham's Gap, to the west of Brisbane, was also 
  closed due to rockslides. 

    On a more positive note, the subtropical storm contributed to very 
  beneficial drought-breaking rainfalls over the southeastern half of 
  Queensland, significantly easing the persistent drought over the Gold 
  Coast region.
    An excellent summary and satellite interpretation, including 
  unofficial comments of the event, can be found at:>

  (Report written by Simon Clarke with significant contributions by
  Jeff Callaghan)

                        TROPICAL CYCLONE GRACE
                           (TC-19P / TD-07F)
                             18 - 24 March
  A. Storm Origins

     Grace is mostly remembered for its effects prior to reaching cyclone 
  status rather than during its life as an officially named tropical 

     A multi-centred tropical LOW formed adjacent to the north Queensland 
  tropical coast near Cooktown as early as 20 March within a very active
  monsoon trough that stretched across the northern Coral Sea and Cape
  York Peninsula in an exaggerated northwest to southeast band across
  the Coral Sea toward New Caledonia.    Over the ensuing days, the
  northernmost circulation became dominant and initially commenced a
  path toward the east and then southeast of the equatorward ridge.  The
  tropical LOW struggled to develop an upper-level structure under a 
  relatively unfavourable upper-level wind environment.  With hindsight 
  the storm may well have been a hybrid system rather than a classic
  tropical cyclone.  Gales were forecast well to the north and south of
  the centre.   The definition of a tropical cyclone in WMO Region V
  requires that gales be present near the centre; hence, the storm
  remained unnamed until there was evidence of gale-force winds near the
  tropical LOW's centre.

     However, by 21/1820 UTC the Bureau of Meteorology, Queensland, 
  deemed that the central circulation had consolidated sufficiently 
  enough to qualify as the second named tropical cyclone of Queensland's 
  2003/2004 season:  Grace.

  B. Storm History

     Tropical Cyclone Grace formed in the open ocean near 20.0S/155.0E
  (or approximately 330 nautical miles east-northeast of Mackay).  At
  this time Grace had a CP of 988 hPa and was moving toward the southeast
  at 15 to 20 knots.   This general motion was to continue for the 
  remainder of the cyclone's life.  Grace peaked in intensity at 985 hPa 
  with a MSW of 50 knots (10-min mean) while centred near 20.3S/155.9E at
  22/0000 UTC.  This intensity was maintained for approximately 6 hours. 
  Thereafter, Grace began to undergo extratropical transition with an
  increasingly asymmetric wind field due to a squeeze with a surface ridge
  to the south.  A vertical circulation remained present in the low levels
  but was sheared away above 500 hPa by a 30 to 50-knot northwesterly

     Grace rapidly lost its entire upper-level structure and was 
  downgraded at 23/1800 UTC from tropical cyclone status (by the TCWC
  at Nadi, Fiji) when located approximately about 400 nautical miles
  east-northeast of Sandy Cape (23.6S/162.3E).  The remnant surface wind
  field of the system meandered to the east and then to the east-northeast
  over the following days, producing a very broad area of gales to its
  south through the Tasman Sea.   Fiji continued to issue warnings on
  ex-tropical cyclone Grace for 24 hours in the event that it should 
  redevelop.  However, after 24/1800 UTC, general gale warnings were 
  issued for a few more days for the extratropical remnants of Grace. 

  C.  Meteorological Observations

     Many thanks to Jeff Callaghan at the Bureau of Meteorology, Brisbane,
  for the following observations, summarised within this report: 


  (a) Rainfall

     Selected rainfall amounts from 2300 UTC 16 March to 2300 UTC
  19 March (all amounts in mm):

  Locataion            To 18 March   To 19 March   To 20 March   Total
  Tully-Saddle Mtn. Alert   163.0         201.0      231.0       595.0
  Copperlode Dam Alert      163.0         256.0      222.0       641.0
  Topaz                     156.0         231.0      372.0       759.0
  Daintree Tea              151.0         203.0      282.0       636.0
  Mt. Sophia                150.0         222.0      279.0       651.0
  Port Douglas              149.0         138.0      166.0       453.0
  Bartle View Alert         147.0         147.0      248.0       542.0
  Kuranda Qld               139.0         232.0      195.0       566.0
  Babinda                   128.0         224.0      334.0       686.0
  Myola Alert               123.0         160.0      172.0       455.0
  Cairns                    116.0         178.0      162.0       456.0

  (b) Storm Surge Observations

     A 0.4 m storm surge was recorded at Cooktown on the morning high
  tide (0.3 m above HAT) on 19 March.  Many boats were upturned in the 
  harbour and washed away.

    A 0.2 m storm surge was recorded at Cairns on the morning high tide 
  (just above HAT) on 19 March and a 0.3 m storm surge on the morning high
  tide (at HAT) on 20 March.  This caused the flooding around the northern
  beach suburbs to build up and required the closing of some roads.

     A 0.55 m storm surge was recorded at Clump Point on 20 March, and the 
  HAT was exceeded with a storm surge of 0.45 m at Lucinda on 20 March. 

  (c) Wind Observations

     The strongest winds from Green Island AWS recorded on 18 March

     1220 UTC 130/37, G. 47 KTS 
     1230 UTC 120/41, G. 49 KTS 
     1300 UTC 120/37, G. 42 KTS 
     2000 UTC 130/38, G. 42 KTS 

  and on 19 March 2004:

     1900 UTC 140/41, G. 47 KTS
     1930 UTC 140/37, G. 43 KTS
     2000 UTC 140/38, G. 43 KTS

  and on 20 March 2004:

     0300 UTC 160/37, G. 43 KTS
     0330 UTC 150/40, G. 46 KTS 
     0400 UTC 160/39, G. 46 KTS 
     0630 UTC 170/38, G. 45 KTS
     0900 UTC 170/38, G. 45 KTS
     1130 UTC 180/37, G. 42 KTS 

     The strongest winds from: 

     Low Isle AWS at 18 March 2004 LWI 1311 UTC 140/39, G. 50 KTS

     Bougainville Reef AWS at 20 March 2004 1900 UTC 220/32 KTS, 996.2 hPa

     Holmes Reef AWS at 18 March 2004 0700 UTC 100/31 KTS, 999.8 hPa;
                     at 20 March 2004 1900 UTC 180/32 KTS, 994.5 hPa;
                     at 20 March 2004 2000 UTC 190/32 KTS, 996.0 hPa

     Flinders Reef at 19 March 2004 1200 UTC 120/35 KTS, 999.5 hPa;
                   at 19 March 2004 1900 UTC 180/32 KTS, 994.9 hPa


  (a) Reef Wind Observations

     Frederick Reef observations:

     21/1200 UTC 140/36 knots  MSLP  997.4 hPa
     21/1300 UTC 140/34 knots  MSLP  997.6 hPa
     21/1400 UTC 140/37 knots  MSLP  996.1 hPa
     21/1500 UTC 140/43 knots  MSLP  993.9 hPa
     21/1600 UTC 140/40 knots  MSLP  993.0 hPa
     21/1700 UTC 150/40 knots  MSLP  992.9 hPa
     21/1800 UTC 130/41 knots  MSLP  993.7 hPa
     21/1900 UTC 140/45 knots  MSLP  992.2 hPa
     21/2000 UTC 130/43 knots  MSLP  992.8 hPa
     21/2100 UTC 130/43 knots  MSLP  992.2 hPa
     21/2200 UTC 130/44 knots  MSLP  993.0 hPa
     21/2000 UTC 130/43 knots  MSLP  992.8 hPa
     21/2300 UTC 130/46 knots  MSLP  992.0 hPa

     (AWS then failed)

  (b) Island Observations

     Heron Island:

     22 March 2004 0801 UTC 180/35, G. 46 knots, and 0815 UTC 180/37,
     G. 48 knots

     Rundle Island:

     21 March 2004 1730 UTC 170/34, G. 41 knots; 1800 UTC 170/35,
     G. 39 knots; and 1830 UTC 170/34, G. 39 knots 

  (c) Coastal Wind Observations
     Cape Moreton - 21 March 2004 2230 UTC 140/45, G. 54 knots 

     Double Island Point - 22 March 2004 0330 UTC 160/41, G. 51 knots 

  (d) Waves

     On 24 March 2004, the wave rider buoys near Brisbane had the 
  significant wave height increasing to 4.8 metres (maximum heights to 
  9 metres) from the northeast.  The main wind field offshore was from 
  the southeast or east-southeast, indicating that the swell arriving 
  here was subject to angular spreading and weakening.  This is an 
  indication of the vast area of gales out to sea during the previous 
  24-hour period.  

     Beaches on the Gold Coast reported waves surging up through
  vegetated dune areas on 24 March, with vertical scarping of up to
  1.0-2.0 m in many places along the open beaches.

  D. Damage and Casualties

     During its formative stages and due to its proximity to the coastline,
  the initial tropical low-pressure complex that eventually consolidated 
  into Tropical Cyclone Grace caused widespread flooding and damage to 
  roads and property along the far north Queensland coast, mainly between 
  Cooktown and Cairns.  Winds and waves brought tide levels above the 
  highest tides of the year (HAT) and this was particularly evident at 
  Cooktown.  Floodwaters closed all major roads into Cairns.

     In addition, a large section of one lane of the Captain Cook Highway 
  north of Cairns collapsed after a landslide consisting of nearly 20 
  metres of rock and boulders the size of cars destroyed the ocean-side 
  road.  The scenic coastal highway and link between Cairns and Port 
  Douglas was closed for several days.   Residents were evacuated from 
  the Whitfield range area due to landslides.

     An estimated $20,000,000 (US) damage to the Cairns region is
  attributed to pre-cyclone Grace.  There were no casualties associated
  with Tropical Cyclone Grace.  However, the State Emergency Services (SES)
  reported that impatient drivers were ignoring road closures.   Several
  roads were closed, including the Bruce Highway south of Cairns, where
  the Tully River was in flood.  On Saturday (19 March), SES workers were
  rescuing motorists who had driven around the closure signs only to be
  swept from the roads.  Also, a man was rescued clinging to the roof of
  his car in floodwaters north of Cairns after being trapped there for four
  hours.  Ergon Energy reported that more than 350 properties were without
  power for some time as crews were unable to get access to repairs, a
  problem caused mainly by fallen trees.  However, there was no significant
  damage to the overall network.

     In combination with a strong high-pressure ridge to its south, Grace 
  produced a large area of gales and high seas in a broad swath across 
  the Coral Sea, particularly south of the St. Lawrence area.  Large 
  swells battered the coast, forcing the closure of some Sunshine Coast 
  beaches.  On 23 March a helicopter and coastguard rescued a stricken 
  craft near the Gold Cost Seaway.  Following shortly after the South 
  Queensland subtropical hybrid (see separate report) some two weeks 
  earlier, Grace exacerbated significant coastal erosion along the 
  South Queensland and Northern New South Wales coastlines.  Sandbagging 
  was required to protect the Currumbin Surf lifesaving club at high tide. 

     On a lighter note, experienced surfboard riders reported having a 
  'field day'.
     Following the loss of tropical cyclone status, the remains of Grace 
  continue to have an impact.  In New Caledonia, further strong winds, 
  heavy rains and flooding were experienced.   A massive oil slick 
  threatened a popular tourist beach in New Caledonia.  Officials in 
  the French Pacific territory put up barriers around the island of 
  Amedee, which was threatened by a toxic oil slick, estimated to cover 
  an area of 20 square kilometres.  They said the oil had come from a 
  boat wrecked several decades ago on a coral reef off South Province.

  F. Further Information/Weblinks

     A satellite shot of Grace can be found at:

  G. Post-Note 

     It is evident that the local media had grasped the name 'Grace' long 
  before a named cyclone was actually a fact.  This might be due to its 
  destructive effects on the Queensland coastline prior to official naming,
  which occurred when the storm was well out to sea and moving away from
  the coast.

  (Report written by Simon Clarke with significant contributions from
  Jeff Callaghan)


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

  Activity for March:  3 tropical depressions
                       1 tropical cyclone **

  ** - System formed west of 160E and moved into basin

                South Pacific Tropical Activity for March

     No tropical cyclones graced the waters of the South Pacific east of
  160E during March except for Grace (pun intended!), which moved into
  Fiji's AOR from the Australian Region shortly before becoming extra-
  tropical.  (The report on Grace can be found above.)  However, there
  were a few tropical depressions which deserve mention.   The first of
  these was designated Tropical Depression 06F by Nadi.  TD-06F was
  first classified as a depression at 2100 UTC on 20 March when it was
  located roughly 225 nm west-southwest of Port Vila, Vanuatu.  The
  system drifted generally southeastward over the next couple of days,
  and the final bulletin at 22/1800 UTC placed the center approximately
  475 nm south-southeast of Port Vila.  Winds near the center were never
  estimated greater than 25 kts, but there were peripheral gales reported
  well to the south and east of the centre.   Some of the islands of
  Vanuatu experienced heavy rains and strong, gusty winds during the
  passage of TD-06F.   Natural disaster chaser Geoff Mackley from New
  Zealand was on the island of Ambrym at the time filming the volcanoes
  on the island.  Heavy weather from the tropical LOW wreaked havoc on
  Geoff's expedition, blowing down tents and damaging other equipment.
  An interesting account of the team's woes can be found at the following

     A weak tropical depression was analyzed at 1800 UTC on 28 March about
  125 nm northeast of Fiji, or near 17S/180E.    A QuikScat pass had
  indicated winds of 30-40 kts under convection several degrees north of
  the center, but these winds were believed to be overestimated due to
  rain effects, and were also not corroborated by surface observations.
  This LOW drifted southeastward and the final reference to it was in the
  Tropical Disturbance Summary issued at 19/0600 UTC when it was located
  near 19S/178W.  No "F-number" was assigned to this weak depression, and
  no track was given in the companion global tropical cyclone tracks file.

     Another highly-sheared system formed on 30 March to the southeast of
  Tonga.  This LOW was designated Tropical Depression 08F, and like TD-06F,
  had fairly weak winds near the center on the order of 15-25 kts, but
  generated gales well to the south and southwest of the center.  TD-08F
  was diffuse and difficult to track, experiencing two significant
  relocations.  By 0600 UTC on 1 April the LOW was becoming extratropical
  roughly 175 nm west-southwest of Rarotonga and was dropped from Nadi's
  tropical weather outlooks.


                               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):>>>

     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:>


     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 report
  for the 2002-2003 Southern Hemisphere season has also recently been

     The URL is:>

     Also, TPC/NHC has available on its webpage nice "technicolor"
  tracking charts for the 2003 Atlantic and Eastern North Pacific
  tropical cyclones; also, storm reports for all the 2003 Atlantic
  and Eastern North Pacific cyclones are now available, as well as
  track charts and reports on storms from earlier years.

     The URL is:>

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


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

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

  Huang Chunliang  (Assistance with Western Northwest Pacific, South
                    China Sea)
  E-mail:  [email protected]

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


Document: summ0403.htm
Updated: 26th October 2006

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