01 December 2002

2002 Tropical Atlantic Activity Report

The hurricane season is basically over (technically on Dec 1st), so it is time for my annual Season Summary. I sent out about 50 updates to this mailing list (which has grown from about 4 subscribers in 1996 to 230 subscribers in 2002, not to mention the updates being posted on several websites) over the past 6 months; now it's time for the final one. This report is structured in the following manner: 1) the Saffir-Simpson Scale, 2) Lifetimes and Intensities, 3) Climatology and Statistics, and 4) Landfall.
As usual, my data (which in large part comes from The National Hurricane Center and Unisys Weather) and typing could contain errors, so if you see a mistake, please point it out to me.

1. Saffir-Simpson Scale of Tropical Cyclone Intensity

CATEGORY            WINDS (mph)   PRESSURE (millibars)
------------------- ----------    ------------------
depression          23- 39        N/A
tropical storm      40- 73        N/A
1                   74- 95        > 980
2                   96-110        965-979
3                  111-130        945-964
4                  131-155        920-944
5                     >156        < 919

2. Lifetimes and Intensities
NAME           DATES OF        MAX WIND    MIN PRES
            ACTIVITY        (kts)       (mb)
------------- --------------- ------------ -------------
ARTHUR        14 JUL - 16 JUL      50          997 (N)
BERTHA        04 AUG - 09 AUG      35         1008 (N)
CRISTOBAL     05 AUG - 08 AUG      40          999 (N)
DOLLY         29 AUG - 04 SEP      55          994 (N)
EDOUARD       01 SEP - 06 SEP      55         1002 (N)
FAY           05 SEP - 08 SEP      50          998 (N)
TD7           07 SEP - 08 SEP      30         1009
GUSTAV        08 SEP - 12 SEP      80          960 (N,H)
HANNA         12 SEP - 14 SEP      45         1001 (N)
ISIDORE       14 SEP - 26 SEP     110          934 (N,H,M)
JOSEPHINE     17 SEP - 19 SEP      50         1004 (N)
KYLE          20 SEP - 12 OCT      75          980 (N,H)
LILI          21 SEP - 04 OCT     125          938 (N,H,M)
TD14          14 OCT - 16 OCT      30         1002

In the previous chart, the N, H, and M that follows some storms denote what statistic they contributed to; Named storm (TS+), Hurricane (CAT1+), Major hurricane (CAT3+).
The winds and pressures reflect the data as posted in the operational advisories, NOT the final “best-track” data that will be available from the NHC in the post-season timeframe.

3. Climatology and Statistics
The average annual number of tropical disturbances (for the period 1944-2000) is:
9.9 named storms
5.9 hurricanes
2.5 major hurricanes

This year, the numbers were near the average:
12 named storms (15 in 2001)
4 hurricanes (9 in 2001)
2 major hurricanes (4 in 2001)

A fairly unique aspect of the past season was the non-uniformity of the activity. No storms formed in June, October, or November, only one formed in July, three formed in August, and eight formed in September, making it the most active month ever in recorded Atlantic Basin history. More amazingly, the first hurricane of the season [Gustav] only formed on September 11th, three days after the climatological peak of the season, and the last one [Lili] formed on Sep 30th. (Recall that last season also had a late start: Erin became a hurricane on Sep 8th.)

Kyle was also worthy of mention in this section. He lasted for 22 days (with 15.25 Named Storm Days), the 3rd longest-lasting storm in the Atlantic (behind Ginger ’71 and Inga ’69). In the process, he strengthened and weakened repeatedly, reaching TS-status four separate times and hurricane-status just once. Of the four hurricanes, half of them [Isidore, Lili] formed (i.e., became hurricanes) in the tropics… the other half formed in the subtropics or mid-latitudes.

For the fourth year in a row, the Atlantic Basin has not experienced a CAT5 hurricane (the last one was Mitch in October 1998).

There were a total of 54.00 “named storm days” (days during which a named storm was present). 10.75 of those days were "hurricane days", and 2.50 of those days were "intense hurricane days". This is 80.5% of the climatological mean, i.e., this season was about 1/5 less active than the "normal" season (last time there was a season below 100% was 1997). The average numbers (for the period 1944-2000) are 46.6 named storm days, 23.9 hurricane days, and 4.7 intense hurricane days.

Here is a summary of the season (VERY brief):

Unlike last season’s A-storm (Allison), Arthur was rather benign. He formed a few miles south of Cape Hatteras, NC on the trailing end of a cold front and headed east-northeast from there over open ocean, eventually losing tropical characteristics just south of eastern Newfoundland.

Bertha formed just off the Mississippi Delta in LA and headed inland. She dissipated over land, but exited over western LA, re-intensified to a Tropical Depression, then drifted into Padre Island, TX… same place that Bret hit in 1999 as major hurricane. She quickly dissipated over TX.

The third storm of the season also had non-tropical origins. Cristobal formed about 4° east of Savannah, GA from the same trough that spawned Bertha the day before. Three days later, he was absorbed by a mid-latitude trough.

Dolly was the first purely tropical system… forming near 10N 32W (deep tropics in the eastern Atlantic) from a tropical wave. She headed WNW, then curved northward at about 54W shortly after which she dissipated due to strong vertical shear.

Edouard formed about 2° east of Daytona Beach, FL, completed a small cyclonic loop, then moved over the northern FL peninsula. He exited and headed into the northern Gulf of Mexico, but strong vertical shear kept the storm from re-developing.

Fay, another short-lived mid-latitude storm, formed about 3.5° east of Corpus Christi, TX and drifted inland, dissipating over TX.

Gustav, who was destined to become the first hurricane of the season, formed between the Bahamas and Bermuda as a Subtropical Depression. He moved northwest toward the NC coast, but was picked up by a trough, just shy of reaching the coast. Through baroclinic enhancement, he became the first hurricane about 5° east of southern MD and was rapidly whisked northeastward by the trough and subsequently absorbed into it.

Hanna formed in the central Gulf of Mexico from a broad area of disturbed weather that was allowed to “brew” for a while. She was in weak steering flow, but meandered westward, then northward into the MS/AL border, doing little more than producing heavy rain.

Isidore had meager beginnings… having formed almost over land (near Trinidad) from a disorganized tropical wave. As the depression moved along the northern coast of South America, it weakened and dissipated into an open wave. However, the remnant disturbance headed northwest and reorganized a day later south of Haiti. Just two days later, he became the second hurricane of the season and rapidly intensified south of Cuba to become the first major hurricane of the season. He then headed west toward the Yucatán Peninsula, then dove south into a fairly populated area. To make matters worse, he ended up completing a small loop over land, exiting the Yucatán just a few miles west of where he entered, then proceeded north into the Gulf of Mexico. At this point, the size of Isidore was incredible, his circulation at least the diameter of the entire Gulf. Most likely, it was the immense size that prevented him from strengthening over the Gulf before hitting the eastern LA coast (for those who are interested, this is called inertial stability).

Josephine was a weak storm that formed in the central north Atlantic, moved northeast, and dissipated… perhaps only noticed by a few ships in popular shipping lanes.

Kyle is a storm for the record books. He formed in the central Atlantic from an extratropical Low. His first move was a cyclonic loop that lasted four days, then headed west, reaching hurricane strength shortly after completing the loop. This westward motion would last for five days, then the next nine days were spent drifting and meandering south and southwest of Bermuda, followed by four days of a more “normal” track along the southeast U.S. seaboard. What makes the longevity more amazing is how little area the storm covered in that time; the other record-holders traversed much longer distances.

Lili was the third and final storm to come from the deep tropics (she would become the 4th hurricane and 2nd major hurricane of the season). Her origins and track were classic, forming from a tropical wave near 10N 45W, passing through the central Caribbean, rapidly intensifying over the Gulf of Mexico, and making landfall on the southern U.S. coast. By far, the most interesting aspect of Lili was the rapid intensity changes that occurred in the Gulf. In the central Gulf, the central pressure fell 32mb in the 24 hours prior to peak intensity, then immediately rose 44mb in the following 24 hours. It will be some time until we fully understand what happened in those couple days… events such as this prove how little we know about rapid intensity changes.

On September 23-26, there were three active named storms... Isidore, Kyle, and Lili. There were many examples of two named storms being present at the same time.

4. Landfall
There were nine landfalling storms this year... seven of which made landfall on the U.S. It is also interesting to note that those seven landfalling storms hit five different states (TX, LA, MS, FL, and SC)… leaving thirteen coastal (Gulf, Atlantic) US states relatively unscathed. Also, we saw the first hurricane landfall (Lili) on the U.S. since Irene in 1999. There had been 21 non-U.S.-landfalling hurricanes between Irene and Lili.

The first column is the storm name, second column is the date of landfall, third column is the approximate time of landfall (UTC or Zulu), fourth column is maximum sustained winds (kts) at landfall, and the fifth column is the nearest location to landfall (preliminary storm-related deaths and damages are shown in parentheses).

--------- ------------ ----- ---------------
BERTHA      8/5  0400    35   Port Sulfur, LA, USA (1 death, minor damage)
            8/9  0900    25   Padre Island, TX, USA
EDOUARD     9/5  0000    35   Daytona Beach, FL, USA
FAY         9/7  1200    45   Palacios, TX, USA
GUSTAV     9/12  0400    70   Fourchu, Nova Scotia, Canada (1 death, minor damage)
           9/12  0900    65   Margaree, Newfoundland, Canada
HANNA      9/14  1400    45   Moss Point, MS, USA (3 deaths, $230 million)
ISIDORE    9/20  2100    85   Las Martinas, Pinar del Río, Cuba
           9/22  2200   110   Telchac Puerto, Yucatán, Mexico (6 deaths)
           9/26  0700    55   Grand Isle, LA, USA
KYLE      10/11  1600    40   Charleston, SC, USA
LILI       9/23  2200    50   Union Island, Grenadines (4 deaths)
           10/1  1600    85   San Juan y Martínez, Pinar del Río, Cuba (4 deaths)
           10/3  1400    85   Pecan Island, LA, USA ($700 million)
TD14      10/16  1300    30   Cienfuegos, Cienfuegos, Cuba

Hurricane Season 2003 begins June 1; the first names in the lineup are Ana, Bill, and Claudette. There are no new names in the upcoming season compared to the last rotation of these names in 1997 (i.e., no names were retired in 1997).

Please visit my tropical Atlantic headquarters.

16 October 2002

TD14 losing the battle.

If you recall Monday's update, things looked favorable for TD14 to
quickly become a TS.  However, the strong vertical shear (presently on
the order of 35kts westerly) has been relentless due to the deep trough
scooping though the Gulf of Mexico, and the storm continues to look more
and more DISorganized.  Recent satellite imagery reveals what seems to
be a frontal structure rather than spiral bands, indicating the
Depression is already becoming extratropical.

At 15Z today, TD14 was located at 22.3N 80.2W (inland over central Cuba;
their third landfall of the season) and tracking NE at 15kts.  A
Tropical Storm Warning is in effect for central Cuba and the Bahamas. 
It is expected to continue to the NE, picking up forward speed and
losing what tropical characteristics it still has as it's absorbed into
the mid-latitude trough.  It looks like the storm will not get named...
although the winds may reach TS intensity, it might not be tropical in
nature by that time.

The tropical wave SW of the Cape Verdes referenced yesterday is
encountering much higher shear now, and attention can shift away from it
for the next couple days, at which point we'll see if there's anything
left of it.

Please visit my tropical Atlantic headquarters.

15 October 2002

TD14 fighting strong shear.

TD14 has been struggling against strong (20kts) westerly shear since
yesterday, which is why it still has not reached TS intensity.  The
low-level center is removed from the convection, although the convection
has been deep and persistent.  The future intensity will be a delicate
battle between shear, land, SSTs, and baroclinicity, but in general, I
don't see it becoming much more than a weak TS.  The IR satellite
imagery reveals no real banding features, no classic CDO, just
convection scattered around a mid-level center.

At 15Z, TD14 was located at 17.9N 82.7W (130 miles southwest of Grand
Cayman) and tracking NE at 4kts.  Maxiumum sustained winds are 30kts and
the MSLP is 1004mb.  Some slight strengthening is likely (to perhaps
50kts according to the latest NHC forecast) due to the latter two
factors I mentioned above [or major strengthening according to the GFDL
model, which makes it a minimal hurricane before hitting Cuba, then a
strong CAT2 as it passes over the Bahamas].  Note the motion... it's
already NE, so that will remove FL from the primary threat zone.  A
Tropical Storm Warning is in effect for the Cayman Islands, Isle of
Youth, and central Cuba, and a Tropical Storm Watch is in effect for the
Bahamas.  The advancing trough will continue forcing the track to
accelerate to the NE, crossing over Cuba on Wednesday morning, the
Bahamas Wednesday night, and be up by Newfoundland by midday Friday. 
Again, if/when this gets named, it will be Marco.

The tropical wave I had mentioned a few days ago near the Cape Verde
Islands is re-organizing somewhat.  It's located at approximately 7N 31W
(1100km SW of the Cape Verdes) and heading W at 15kts.  It is also in a
region of high shear, but that MAY relax over the next couple days.  

Please visit my tropical Atlantic headquarters.

14 October 2002

TD14 forms in northwest Caribbean.

The area of disturbed weather that was in the southwest Caribbean late
last week and during the weekend has drifted north and become much
better organized, exactly how the models had predicted five days ago
(they're on a roll this season).  As I watched the system over the
weekend, it was remarkable to watch in real-time a tropical cyclone
develop out of nothing more than a broad area of convection.  Hour by
hour, the convection became more centralized, a mid-level circulation
developed, deeper convection, and finally a low-level circulation was
obvious on visible imagery this morning (I'm sure I'm not the ONLY one
who woke up excitedly to check the first available VIS images for a
low-level circulation...).

At 15Z today, TD14 was at 17.8N 83.0W and tracking N at 10kts. 
Intensity is 25kts and 1009mb, but is expected to quickly reach TS
status today (and would become the thirteenth named storm of the season,
Marco).  A Tropical Storm Warning is in effect for the Cayman Islands,
Isle of Youth, and central Cuba, and a Tropical Storm Watch is in effect
for southern FL from Golden Beach to Key West.

The forecast track is generally N-NNE, allowing the center to pass east
of the Florida peninsula, but admitedly, the distance at which it's
forecast to pass by at is within the forecast error (hence the TS Watch
for southern FL).  There is a very deep trough digging into the Gulf of
Mexico which will be the primary steering mechanism for this storm, so
it's movement and amplitude will play a huge role in track forecasting. 
The GFDL model is very insistent on making this a strong storm, hitting
Cuba as a 92kt CAT2 hurricane, then the Bahamas as a 103kt CAT3
hurricane.  The SSTs in its path would support a major hurricane, it
will just be a matter of how Cuba affects it and how much shear the
trough will introduce.

Please visit my tropical Atlantic headquarters.

12 October 2002

Kyle makes landfall, and one more intensity cycle; eyes point to SW Caribbean.

Although landfall technically occurred yesterday, it's tough to say
precisely WHERE it was.  Kyle followed the contour of the southeast US
with amazing detail, from the GA/SC border all the way north to Cape
Hatteras.  There were six damaging tornadoes in eastern NC and some
fairly serious coastal flooding associated with his passage.  Of course,
something strange or unusual had to happen... he was downdraded to a TD
at the 00Z Intermediate advisory, then once back out over the Atlantic,
re-strengthened into a TS at 09Z -the 5th time doing this.  Now at 15.5
Named Storm Days and nearly 22 total days, he is (I believe) the 3rd
longest-living tropical cyclone in Atlantic history.

At 15Z, TS Kyle was at 36.0N 75.0W (35 miles ENE of Nag's Head, NC) and
tracking ENE at 20kts.  Max winds as indicated by buoys, satellite, and
aircraft are 40kts and the MSLP is 1009mb.  He is in the process of
becoming extratropical, and the baroclinic assist he's getting is what
led to the intensification (as one would have expected).  However, in
the near future, his circulation will be indistinguishable from that of
the trough and mid-latitude Low.  And so ends the life of a storm we've
been talking about since September 20.

Meanwhile... the wave near the Cape Verde Islands is looking much less
organized and is not of great concern at this time.  However, of much
greater concern is the brewing situation in the southwest Caribbean Sea
(correction to yesterday's update: the area of interest is not the Bay
of Campeche, but rather the Caribbean down by Panama).  Although
disorganized and broad now, the area of disturbed weather down there is
still favored by virtually every model for development.  The track is
what makes it a concern:  northward over western or central Cuba, then
into the southern FL peninsula, then perhaps up along the US east
coast.  The timing and exact positions are not agreed upon, but that
general scenario is.  Although presently in very strong westerly shear,
if organization occurs fairly soon and the models therefore prove their
skill, landfall on FL could be as soon as Tuesday morning, but could
certainly be later.  Seeing that this year has been the year of U.S.
landfalls, I'd be cautious!  

Please visit my tropical Atlantic headquarters.

11 October 2002

Kyle becomes a TS, for the 4th time; two other areas are becoming interesting.

Yesterday at roughly 22Z, a HUGE explosion of convection occurred
directly over Kyle's low-level center, leading to a large and cold CDO
just four hours later.  This burst allowed him to concentrate enough
vorticity to spin the storm back up to a Tropical Storm by 09Z this
morning.  Still under a more ragged -70C CDO, the storm has not gotten
convectively organized, as determined from an SSM/I microwave satellite
pass a few hours ago.  For anyone interested, the Wilmington, NC (KLTX)
radar shows the passing of the storm as he makes landfall not far from

At 15Z today, Kyle was located at 32.4N 80.1W (20 miles SW of
Charleston, SC) and tracking NE at 13kts.  Intensity is 40kts and
1008mb.  He is expected to make landfall on Charleston, SC as a 40kt TS
at about noon Eastern time and remain a minimal Tropical Storm for the
next three days, and will therefore keep adding to his total of 15 Named
Storm Days and nearly 21 days of existence.  A Tropical Storm Warning is
in effect for all of the SC and NC coasts, a Tornado Watch is in effect
for northern SC and most of eastern NC, Flash Flood Warnings are in
effect for parts of SC, NC, and VA.  

The tropical wave off of Africa's coast I mentioned in yesterday's
update has become better organized. Presently at about 11N 25W (just
south of the Cape Verdes), it's tracking west, and vertical shear is
decreasing as an upper-level anticyclone develops over it.  There is
healthy outflow and the low-level vorticity is becoming more
centralized.  The models that do develop this system recurve it by 40W
or so, so it's seemingly no threat to land.

Now, here's something that strict observationalists should bypass... the
models are beginning to agree on a solution of a rather impressive
tropical cyclone developing in the southwest Caribbean by Sunday, and
taking it northward into western Cuba then perhaps the FL peninsula
(poor folks in western Cuba have earned a multi-year vacation from
landfalls after this season).  There is already a large pool of
disorganized vorticity and disturbed weather in extreme southern Bay of
Campeche.  This needs to be watched very closely, as this scenario is
climatologically favored for this time of the season (recall Mitch '98
and Michelle '01?  Maybe Marco '02 will follow in their footsteps??).

FYI, the next numbers/names on deck are 14,15/Marco,Nana... just in case
we should need them in the coming days.  If the past several years are
any help in forecasting activity, the season won't shut down until the
very end.  The past 4 years we've reached the N (11/24-12/1), L
(11/13-11/21), N (10/19-10/22), and O (11/24-12/4) storms.  So far we're
at L this season and it's only 10/11!

Please visit my tropical Atlantic headquarters.

10 October 2002

Kyle still out there...

Since the last update on Monday, Kyle had experienced yet another bout
with high vertical shear.  It stripped away all convection leaving just
a well-defined low-level vortex.  But alas, since we'd expect nothing
less from him, convection re-fired in Kyle on Wednesday evening.  Well,
James Franklin at NHC put it best in the 03Z duscussion today:
"OH...OH...OH...OH...STAYIN ALIVE...STAYIN ALIVE".  Julian Heming from
the UK Met Office then commented that "I suppose we should be thankful
it is not heading for Massachusetts, to save us further quotations from
the Brothers Gibb!"  Nothing wrong with a little humor in the
forecasting business.  For those keeping track, Kyle's now the 6th
longest-lasting tropical cyclone in the Atlantic basin since records
began.  He's been around for 19.75 days, 14.75 of which were Named Storm
days (35kts+).

At 15Z today, TD Kyle was at 28.5N 78.6W (about 2.5 degrees east of Cape
Canaveral) and tracking W at 10kts.  Intensity is 25kts and 1010mb...
quite weak, but convection is becoming better organized, albeit slowly. 
A Tropical Storm Watch is in effect for the northern half of eastern FL
and parts of the southern GA coast.  He's expected to pass VERY close
the FL/GA border Friday morning as a 30kt TD, then head NE along the
coastline and over SC, NC, VA, if he's even trackable by that point.

For lack of something else to talk about, a fairly large area of
disturbed weather has exited the African coast at about 11N 15W.  Shear
is presently less than 10kts, and there's weak vorticity associated with
it, but such favorable conditions should not persist long enough to
allow cyclogenesis to occur.

Please visit my tropical Atlantic headquarters.

07 October 2002

You guessed it, Kyle's still out there.

Little has transpired over the weekend: TS Kyle is still at about the
same place and same intensity as he was on Friday morning.  As of 15Z
today, he has racked up 13.75 Named Storm Days, 2.5 Hurricane Days, and
0 Intense Hurricane Days, and has still traversed a mere 20 degrees
longitude in that time (with only 2 real eras of movement, the rest of
the time he's been meandering or stationary).  This morning, the center
has apparently split again, making the position and motion highly

The 15Z advisory places TS Kyle at 32.6N 70.8W and drifting SW at 3kts. 
Intensity is 35kts and 1006mb.  He did survive another episode of high
shear during the weekend, during which he weakened to a TD, again; but
has since made a comeback, again.  There is presently deep convection
near the southern main center, so shear is most decidedly weaker now. 
The forecast is for very gradual and slight strengthening over the next
day or so, then level off.  In terms of track, he is expected to start
drifting more southerly, then perhaps SSW through Thursday, but I can
say that NHC's skill with his track in the past has been minimal
(likewise with the computer models).  We'll probably still be talking
about Kyle next weekend too.

Please visit my tropical Atlantic headquarters.

04 October 2002

Kyle becoming a permanent fixture, Lili transitions to extratropical.

For those of you keeping track, this is the second full week of talking
about Kyle, and it's not over yet.  Kyle has confounded all
climatological wisdom about Atlantic storms with his longevity, mediocre
intensity, and stagnation.  As a named storm, his intensity has had a
range of 40kts, and in terms of stagnation, in two weeks his entire
track would fit within a box 20 degrees E-W by 7 degrees N-S.  At 15Z
today, TS Kyle was located at 30.2N 69.8W and is tracking NW at 5kts. 
Intensity is 35kts and 1004mb.  The forecast is for a turn to the north,
slight acceleration, and slight strengthening as the shear should lessen
by Friday night.  He's expected to be about 400km (250 miles) east of
Cape Hatteras on Monday morning.

After landfall on Thursday morning, Lili continued up through LA, then
along the AR/TN border and is still tracking NE across the lower
Mississippi Valley, but is no longer tropical in nature.  The last
advisory was written on Lili at 09Z today.  I know of 14 tornadoes that
she spawned in LA and MS yesterday after landfall, and more could occur
today across northern AL, central TN, KY, IN, OH, and southeast MI as
the vorticity associated with the remnants could still be focused into
smaller-scale features... namely tornadoes.  Other landfall effects
include half a million people without power, many structures collapsed
from wind or floods, and flooded coastal areas; but since 900,000+
people were evacuated in preparation for landfall, the human toll was
greatly minimized (several injuries, but no known deaths so far).

Please visit my tropical Atlantic headquarters.

03 October 2002

Kyle STILL a TS, Lili made landfall near Pecan Island at 0900 CDT.

Kyle has now racked up 11 Named Storm Days (days during which the storm
was at TS strength or above), and has barely traversed the Atlantic. 
Only as a hurricane was he moving appreciably, the rest of the time was
spent meandering slowly or completely stationary.  At 09Z today, TS Kyle
was located at 29.0N 66.6W and stationary.  Intensity was 55kts and
994mb, so it appears that his effort to regain hurricane status
yesterday was short-lived (but he's surprised us before).  The forecast
is for persistence once again... roughly same intensity and a start of a
NW movement over the next three days.  

Hurricane Lili had intensified very rapidly late Wednesday, with the
MSLP dropping 32mb in 24 hours -ending at 21Z- or 15mb in 6 hours
-ending 21Z.  Then, in hindsight, it seems that she was undergoing an
eyewall replacement cycle, which is quite normal for intense hurricanes,
but land (among other factors including SSTs, shear, trough, etc)
perhaps disrupted the normal completion of that cycle.  The inner
eyewall collapsed, and although a new larger, outer eyewall eventually
formed (as evident on the New Orleans and Lake Charles radars), it was
incomplete and the inner core did not fully [re]mature.  As a result of
the disruption, the storm weakened almost as fast as it strengthened
just hours before.  The pressure rose 19mb in 12 hours -ending at 09Z

Lili made landfall between Pecan Island and Marsh Island, LA (in
Vermilion Parish and Iberia Parish, respectively) at about 1400Z today
as an 85kt CAT2 hurricane... this is 80 miles SE of Lake Charles and 80
miles SW of Baton Rouge.  Due to the rapid weakening, the storm surge
was not as bad as expected, perhaps 6-10' instead of 15-20'.  Although
flooding from the rain will be a concern, it won't be as bad what we saw
from Isidore last week at this time.  So far I know of one tornado in
Terrabonne Parish, but that threat will still be quite large today,
especially to the east of the path.  This scenario was very much like
Bret in 1999; he was a CAT4 just off the coast but weakened just prior
to landfall, and hit halfway between Brownsville and Corpus Christi. 
Lili had the same intensity trend, and made landfall between Houston and
New Orleans.  Even Andrew in 1992 would have been magnitudes more
devastating if he hit downtown Miami, just 15-20 miles north of
Homestead.  The U.S. has been very fortunate in the exact locations of
these impacts. 
As of 13Z today, Lili was located at 29.4N 92.2W and moving NNW at
13kts... this is directly over the LA coastline, so weakening will be
accelerated now as friction with land causes the Low to fill and the
fuel source (Gulf) is gone.  The maximum sustained winds are 85kts and
the central pressure is up to 962mb.

The watches and warnings are almost too numerous to mention, but I'll
highlight: a Hurricane Warning is in effect for extreme eastern TX and
LA as far east as the Mississippi delta, a Tropical Storm Warning is in
effect for the rest of LA and MS and AL, Coastal Flood Watches and
Warnings extend all along the LA, MS, AL, and FL panhandle coasts, a
Tornado Watch is in effect for all of southern LA and MS (this watch
will shift northward during the day to include northern LA, MS, AR, and
perhaps AL), and Tornado Warnings and Flash Flood Warnings will
doubtlessly be posted throughout the day, so stay tuned to local TV or
radio if you live any of the states mentioned above.

Please visit my tropical Atlantic headquarters.

02 October 2002

Lili rapidly intensifies to CAT4 storm...

Since the primary update was mailed out earlier today, Lili has
strengthened significantly.  She is now at 120kts sustained (gusting to
145kts) and has a central pressure of 938mb.  This makes her a CAT4
hurricane.  The pressure has fallen 15mb in the past 6 hours, and the
temperature in the eye is continuously rising.

As of 21Z, Lili was at 25.9N 90.0W and tracking NW at 14kts.  Landfall
is expected to be very near Pecan Island at 18Z tomorrow (1300 CDT) as a
125kt storm (11kts shy of CAT5 status).  Besides the severe sustained
winds, gusts could reach 155kts (180mph), tornadoes will be likely in
the rainbands and eyewall, storm surge could be 18-20' and travel 25
miles inland, and up to a foot of rain could fall along her path.  Over
200,000 people have already been forced to evacuate, and more are still
in progress.

I'll take one sentence to summarize recent happenings with Kyle... he's
now at 60kts and 991mb, so is almost a hurricane again.

Please visit my tropical Atlantic headquarters.

Kyle making a comeback, Lili a major hurricane and heading for LA.

Kyle, who during the weekend was severely sheared, is now making a
comeback.  Convection is returning to the center of circulation and
banding features are starting to form (again).  Recall that advisories
began on Kyle on 9/20!  As of 15Z today, he was at 28.9N 67.3W and
heading ENE at 4 kts.  Intensity is 50kts and 997mb.  He is expected to
strengthen to nearly a hurricane again on Thursday, then weaken as
northeasterly shear picks up.  As one might expect for this storm known
for his longevity and permanence, the forecast track is to remain nearly
stationary for the next three days or so.

Yesterday afternoon, as expected, Hurricane Watches were issued for the
LA and eastern TX coasts.  The storm rather rapidly got much better
organized; starting at about 21Z the inner core symmetrized and a small
defined eye formed just a couple hours later.  Since then, the hurricane
has strengthened continuously (pressure has fallen 13mb in the past 12
hours, and 18mb in the past 24 hours).  According to another early
morning chat with Hucky Purpera at the LA Office of Emergency
Preparedness, mandatory evacuations were ordered for Iberia Parish
yesterday afternoon, and for Cameron and Terrabonne Parishes this
morning.  Vermilion and Calcasieu Parishes should be issuing evacuation
orders this morning as well. Lafayette Parish is very closely watching
the situation but no evacuations have been ordered as of this writing. 
Their goal is to have everyone out of the warned areas by sundown
tonight.  FEMA and Red Cross personel are already in place, along with
several university teams who will be collecting data in the eye and
eyewall (hopefully).

Presently, a Tropical Storm Warning is in effect for extreme eastern TX
and then from the Mississippi Delta to the AL/FL border.  A Hurricane
Warning is in effect for nearly all of the LA coast.  Residents along or
near the path can expect 10-20" of rain, sustained winds of 125mph,
gusts to 155mph, 15-18' storm surge, and tornadoes mostly to the east of
the center.  Considering that we're talking about coastal LA, the
biggest threat will be the storm surge, a swell of water caused by the
very low central pressure (the ocean actually bulges up under the center
of the storm) and the bulldozing of the water by the storm's relentless
winds.  This is above and beyond the normal daily tides.

At 15Z today, Lili was located at 24.8N 88.9W and tracking NW at 13kts. 
The intensity has increased to 105kts and 953mb, making her a Category 3
hurricane and the second major hurricane of the season.  Her eye is very
tiny at only 27km in diameter.  The forecast is for only slight
strengthening before landfall near Pecan Island, LA (about 62 miles
southeast of Lake Charles, or 85 miles east of the TX/LA border) midday
Thursday.  The exact landfall position is uncertain though, and with an
eyewall this intense and small, it DOES matter exactly where she hits...
unlike with Isidore who didn't have an eyewall to speak of when he made
U.S. landfall on eastern LA last Thursday.  Residents in these parts may
recall Hurricane Audrey (1957) who made landfall as a CAT4 storm,
killing nearly 400 people... stories of that event should be enough to
scare people to evacuate!

Please visit my tropical Atlantic headquarters.

01 October 2002

Kyle still a TS and meandering, Lili nearly a CAT2 hurricane over Cuba.

At 21Z yesterday, Kyle was downgraded to a TD, experiencing fairly
strong vertical shear and undergoing what appears to have been
circulation splitting.  However, as of 15Z today, he was re-upgraded to
a TS based on satellite estimates.  The location at 15Z was 28.0N 67.3W
and motion was stationary.  Winds are 35kts and MSLP is 1005mb.  The
forecast is for persistence... same place and same intensity for the
next three days.  A Tropical Storm Watch is still in effect for Bermuda.

The real concern is Hurricane Lili, who continues to intensify on her
track over Cuba and toward the U.S. Gulf coast.  She has passed directly
over the Isle of Youth and is over extreme western Cuba now.  Aircraft
flying into the storm have reported that the eyewall has been damaged by
the passage over land, and IR satellite imagery confirms that the
central convection is slightly weaker.  However, this is almost
certainly a temporary disruption in the intensification process... and
an expected one.  Lili will soon be over the open Gulf waters, which are
warm enough to support a major hurricane.  Shear is very low through the
storm and is expected to remain low.

At 15Z today, Lili was at 21.8N 83.7W and tracking WNW at 11kts (note
the increase in speed compared to during the weekend).  Maximum
sustained winds are 80kts (nearly CAT2) and the minimum central pressure
is down to 971mb.  A Tropical Storm Watch is in effect for the northern
Yucatan Peninsula, and a Hurricane Warning is in effect for the Isle of
Youth and western Cuba.  I strongly suspect that a Hurricane Watch will
be issued for LA and eastern TX by later today.  

Early this morning, I spoke with Hucky Purpera, the Natural and
Technological Hazards Division Chief at the Louisiana Office of
Emergency Preparedness, and although nothing has been ordered yet, he
believes their meeting this afternoon will initiate evacuations. 
Precautionary evacuations need to begin very soon, since landfall is
only about 48 hours away (volunteer evacuations are always appreciated
for those who wish to do so).  Recommended evacuations should begin
tonight or early Wednesday morning, and mandatory evacuations should
begin midday Wednesday at the latest.  The official forecast is for
landfall as a moderate CAT3 hurricane on Pecan Island, LA on late
Thursday morning.  This is about halfway between New Orleans and
Galveston, so a track deviation could put either of these major cities
at risk.  A storm such as this will force everyone south of I-10 and in
her path to leave; this could include Cameron, Lake Charles, Baton
Rouge, and/or New Orleans.  See
http://www.dotd.state.la.us/maps/lacoastlandsurf.pdf for a very good
coastal map of LA and evacuation routes.  LA has only experienced 12
major hurricane landfalls in the past 103 years, so it's not an
every-year occurence!

Elsewhere, a tropical wave and associated 1011mb Low is located at 10N
55W and moving W at 10kts.  There is a well-defined surface circulation,
but convection is sparse and the system is disorganized.
Please visit my tropical Atlantic headquarters.

30 September 2002

Kyle still hanging on, Lili now a hurricane... and a serious threat to the U.S. coast.

TS Kyle is still far from the U.S. mainland and presents no threat. 
Although this morning's satellite imagery indicates that he is
reorganizing slightly, the shear is still causing the convection to be
well-separated from the deepest convection.  At 15Z today, Kyle was
located at 29.8N 64.8W (345km south of Bermuda) and stationary.  Only
slight strengthening to 45kts is forecast as he curves back to the
east.  A Tropical Storm Watch is in effect for Bermuda.  Kyle will be
remembered at the end of the season for contributing to so many Named
Storm Days with so little movement... 9.25 NSD so far with another 3-4+
to add on and has only covered 15 degrees longitude and very little
latitude in that time (for reference, Isidore contributed 8.5 NSD).

At 15Z today, Lili was upgraded to a hurricane, the fourth of the
season.  Since yesterday, the satellite presentation has continually
improved, with a cold CDO and classic bands.  Her center has managed to
avoid passing over land, successfully weaving between Jamaica, Haiti,
and Cuba (look at a map, that's no easy task!).  So although the outer
bands had been disturbed by the 7,000'+ mountains on those islands, the
inner core was always over the warm Caribbean.  As of 15Z, Hurricane
Lili was at 19.8N 80.0W (over Cayman Brac) and tracking WNW at 9kts. 
Intensity has reached 65kts and 986mb.  Now that real organization is
taking place and land is becoming less of a player, we should begin
seeing more rapid intensification (although true "rapid" intensification
is not forecast).  

The NHC forecast calls for her to be 105kts on Thursday morning at 8am,
just 90 miles south of the western LA coast.  Today's 06Z run of GFDL
shows landfall on the TX/LA border on Thursday morning at 7am.  The 00Z
AVN shows landfall very near Lake Charles, LA at the same time.  The 00Z
NOGAPS run is bit slower and goes for landfall on the LA/TX border on
Friday morning at about 1am.  Keep in mind that times are still
approximate, but the location is being narrowed down to western LA or
extreme eastern TX as soon as early Thursday morning.

A Tropical Storm Warning is in effect for eastern Cuba and Jamaica; a
Hurricane Warning is in effect for western and central Cuba, the Isle of
Youth, and the Cayman Islands.  I would not be surprised if a Hurricane
Watch would be issued for the eastern TX and all of LA coasts today to
expedite evacuations.  As pointed out earlier, Lili is forecast by
virtually every model to make landfall on the LA coast as a major
hurricane on Thursday morning (exactly one week after Isidore made
landfall at the same location), so prudent emergency management would
call for optional evacuations beginning today and mandatory evacuations
beginning Tuesday.

The tropical wave in the central Atlantic mentioned in the past two
updates has become less organized and should be beaten by shear in the
near future.

Please visit my tropical Atlantic headquarters.

29 September 2002

Kyle heavily sheared, Lili encountering land, eastern Atlantic still interesting.

TS Kyle continued to weaken in a high shear environment on Saturday, but
is forecast to make a comeback.  The big story with him is the motion...
there has been very little his whole existence!  Recall that he formed
on 9/20, and is still around, and probably will be for another 5+ days,
still no threat to land in that time.  Although the low-level center has
been exposed for a while, centralized convection is making a comeback
this morning, indicating that perhaps the shear is lessening.  As of
15Z, he was at 27.7N 64.8W (520km S of Bermuda) and stationary.  Max
winds are 40kts and the MSLP is 1002mb.  The forecast is for gradual
strengthening and an eastward drift.

Lili has kept a reasonable appearance on satellite, but is passing over
waters surrounded by Jamaica, Hispaniola, and eastern Cuba, all very
mountainous, and that has inhibited her intensification (the center has
managed to weave between all of these islands, a remarkable feat in
itself).  However, all signs point to her being able to regain strength
once she clears the terrain.  At 15Z today, TS Lili was at 18.7N 77.6W
(between Cuba and western Jamaica) and heading W at 6kts.  Intensity is
45kts and 994mb.  A Tropical Storm Watch is in effect for Grand Cayman
and central Cuba and a Tropical Storm Warning is in effect for Cayman
Brac, Little Cayman, Jamaica and eastern Cuba.  She is forecast to
become the 4th hurricane of the season on Tuesday morning.

The GFDL model has quite the ironic track forecast for her: landfall
near Lake Charles, LA on Thursday morning as a CAT3 hurricane.  This
would put New Orleans just to the east of the center which would be
catastrophic, not to mention that they just had a TS landfall last
Thursday morning at the same place.  We'll see how the next couple of
days pan out, but by Tuesday, evacuations need to begin, so hopefully
the models will continue to converge on a solution (so far they do). 
AVN, the most reliable track forecaster this season, goes for landfall
on New Orleans on Thursday morning.

The tropical wave that was near the Cape Verde Islands on Friday is now
1300 miles WSW of those islands (about 12N 35W) and moving W at 12kts. 
It has a 1012mb Low, Vertical shear is fairly high, the vorticity center
is elongated, but the SSTs are warm and as it heads west, it may take
advantage of improving conditions.

Please visit my tropical Atlantic headquarters.

27 September 2002

Isidore still a rain maker, Kyle still a small hurricane, Lili regains TS status.

Isidore continues to dump copious amounts of rain over his path, now
with the heaviest over OH and PA, but will spread into NY and New
England during the day. Flooding, gusty winds, and isolated tornadoes
can be expected today in KY, VA, WV, PA, NJ, CT, and southern NY.

Hurricane Kyle continues to spin out in the central Atlantic southeast
of Bermuda. He has lost the mature appearance (eye, defined spiral
bands, etc) and is now a rather ragged hurricane.  At 15Z, Kyle was at
26.9N 60.7W and drifting WSW at 4kts.  Maximum sustained winds are still
75kts and the central pressue is 980mb.  The long-term forecast is
curious.  The ridge that has been over him -and responsible for the lack
of steering flow- may end up eventually nudging him along its periphery
and toward the U.S. east coast.  Even if this is the case, it's MANY
days off, because over the next few days, he's expected to stall with
little change in intensity (fluctuations here and there, but no real

At 15Z yesterday, NHC had ceased writing advisories on Lili, as she had
apparently degenerated into an open wave.  However, aircraft flights
into the system had shown that by 03Z today, a low-level circulation had
reformed and she was upgraded to TD Lili.  Then at 15Z today, she was
upgraded to TS Lili and shows signs of continued organization.  She has
a very cold CDO and the microwave imagery is showing an established
circulation as well.  As of 15Z, she was located at 16.3N 74.8W and
tracking WNW at 6kts.  Intensity is 35kts and 1005mb.  A Tropical Storm
Warning is in effect for Jamaica.  The forecast is for gradual
strengthening, reaching minimal hurricane by Monday morning just south
of central Cuba.  Given the forecast track and history of storms this
season, I would tend to delay that by a bit, to perhaps Tuesday
morning.  She is expected to enter the Gulf, and if the long-range
models are at all correct, she should continue the WNW track toward the
TX coast.

Elsewhere, a tropical wave that exited the African coast yesterday is
now at about 9N 25W (south of the Cape Verde Islands) and shows some
hint of organization, but this area is not very climatologically favored
for development this late in the season.  However, shear is low and SSTs
are warm, so if the convection can organize, it could become a TD in the
next couple days.
Please visit my tropical Atlantic headquarters.

26 September 2002

Isidore makes landfall, Kyle makes hurricane, Lili downgraded to TD.

Isidore made landfall today at about 2am CDT on Port Fourchon, LA (about
75 miles south of New Orleans) as a 55kt Tropical Storm.  Since there
was no eyewall, the exact location doesn't matter too much; the storm is
huge and all surrounding states will feel roughly the same effects
(something I emphasized during a -unexpected- radio interview with
Denver's KOA News Radio yesterday morning, along with the inland
flooding potential).  At 15Z, TS Isidore was located at 30.7N 89.7W and
tracking NNE at 12kts.  This motion will accelerate with time as it
merges with a trough.  Intensity is 50kts and 985mb.  Further weakening
is expected as the storm is now over land and being sheared by that

The storm surge on the east side of the center was moderate, 3-6' above
tides in most areas, then about 10' above tides in the more shallow
bays.  Keep in mind that a good portion of southern LA is at or below
sea level in the first place. In addition to the storm surge, places
along the central Gulf coast will receive on the order of 2 feet of rain
from Isidore.  Also, there have already been several tornadoes
associated with the Tropical Storm, though I believe there have not yet
been any injuries or deaths.

Tornado Watches and Warnings cover parts of LA, MS, AL, FL, and GA and I
suspect the watches will be shifted north as the rainbands move further
inland.  There are Coastal Flood Watches and Warnings along those same
states; Tropical Storm Warnings for eastern LA, MS, AL, and the FL
panhandle; and Flood Watches for every state from the Gulf coast up to
New England, including LA, MS, AL, FL, GA, TN, KY, IN, OH, PA, and NY. 
People at highest risk for tropical cyclone-spawned tornadoes are those
living in eastern MS, southern TN, all of AL, western GA, and the
western FL panhandle.  This would be a good day for everyone in the
states mentioned in this section to keep a local radio station or TV
channel on just in case action is required, whether it's involving a
tornado or a flash flood.  http://www.spc.noaa.gov/products/wwa/ is a
good source for monitoring the Watches and Warnings, but should not
replace local broadcasts from your NWS office.

Yesterday at 21Z, Kyle was upgraded to a hurricane, the third of the
season.  The satellite signature is fairly good; there's a small ragged
eye that microwave, visible, and infrared imagery all pick up, but the
convection is not very deep (cloud tops aren't very cold).  The drift to
the SW over warmer SSTs has made all the difference, but he exists in a
small patch of the Atlantic where the vertical shear is favorable for
the development of a small storm.  At 15Z today, he was at 27.8N 58.7W
(770km SE of Bermuda) and heading WSW at 8kts.  Maximum sustained winds
are 75kts and the MSLP is 980mb.  Only slight strengthening is possible
in the short term before an advancing trough introduces higher shear. 
Of course, as you're probably envisioning, he's already completed one
loop, and now he's heading SW, but a trough is coming, so he'll turn
back NE ahead of it.  By the time he's done, it will be an interesting
track to look at!

Lili was downgraded to a TD today at 15Z and the last advisory has been
issued, unless regeneration occurs (which could very well happen).  An
aircraft in the storm was unable to find a defined low-level center, and
the convection has been somewhat disorganized.  Her position at 15Z was
15.2N 72.5W (roughly halfway between Colombia's Guajira Peninsula and
Haiti), winds were reported as 30kts with 1006mb central pressure. 
Redevelopment is forecast on Sunday morning just south of the western
tip of Cuba as she heads into the same area that Isidore did when he
became a hurricane... between the Yucatan Peninsula and Cuba, perhaps
entering the Gulf by Monday afternoon.

Please visit my tropical Atlantic headquarters.

25 September 2002

Isidore still huge but weak, Kyle nearly a hurricane, Lili losing to shear.

Very little has changed with Isidore since he exited the Yucatan
Peninsula.  Without question, the most notable feature of the storm is
its size.  It has lost all central convection, but the low and mid-level
swirl (including an area of tropical-storm-force winds 800km in
diameter) is now perhaps 2000km across in the E-W direction and 3000km
across in the N-S direction.  The outflow, assisted by a mid-latitude
trough to the north and the ITCZ to the south extends from just west of
Costa Rica and up to central NY.  See
for an impressive basin-wide VIS shot.  At 15Z TS Isidore was located at
26.0N 90.2W (430km south of New Orleans) and tracking N at 11kts. 
Intensity is 50kts and 990mb, indicating that the Low is filling
(pressure is rising).  However, a Tropical Storm Warning is in effect
from approximately Galveston, TX to Apalachicola, FL and a Hurricane
Watch is in effect for all of the LA and MS coasts.  

Landall is expected to be during the early morning hours on Thursday in
eastern LA.  The forecast is for gradual strengthening before landfall
(perhaps not to a hurricane though), then to track over the southeast
U.S. and up to the northeast, passing over TN, PA, and ME as landmarks
along the way.  This journey will be part of his merging with a
mid-latitude trough... forcing him to become extratropical.  By the time
the remnants are over Pennsylvania, New York, and New England on Friday,
there could potentially be a very big rain event, not to mention wind. 
The convective patterns and wind structure already indicate that Isidore
may be losing tropical characteristics... however, this does certainty
not mean he is no longer a threat.

TS Lili has been having problems dealing with vertical shear (only 10kts
SW-erly, but the convection is displaced from the low-level center
nevertheless).  The storm is poorly organized and even aircraft flying
through the system had difficulty finding the center.  But conditions
are expected to improve (depending on track), allowing her to once again
approach hurricane strength during the weekend.  The 15Z position was
14.1N 69.3W and movement was WNW at 10kts.  She is a weak TS with 40kt
sustained winds and a 1008mb MSLP.  A Tropical Storm Watch is in effect
for most of the southern coast of Hispaniola and I suspect that
southeast Cuba will also soon issue a watch.  The official forecast is
an outlyer in terms of track... NHC is going north of most of the model
tracks, which agree on taking her between Jamaica and Cuba and
eventually into the Gulf.  That version of the track forecast would be
less favorable for her development, but as I said, the models are in
pretty good agreement on that, so we'll see.

TS Kyle is just now moving over 28C+ SSTs, and the convective pattern
indicates that he's taking advantage of the increased fuel source. 
Since this morning, an eyewall has formed; for the first time,
convection has wrapped all the way around the center.  As of 15Z, Kyle
was located at 29.2N 54.8W and heading WSW at 7kts.  Intensity is 60kts
and 990mb, very much on course to maintain the eye and reach hurricane
status later today... the third of the season.  The forecast is for
continued strengthening in the low shear and warming SST environment,
then weakening as a trough approaches and shears him apart (tough world
out there!).

The tropical wave I'd been mentioning the past couple days in the
central Atlantic has dissipated.

Please visit my tropical Atlantic headquarters.

24 September 2002

Isidore back over the Gulf, Kyle heading for warmer waters, Lili getting better organized.

Isidore spent all of Monday over the northern Yucatan Peninsula, and did
indeed complete a small cyclonic loop there, exiting the coast just
miles west of where he entered.  The inner core convection is virtually
gone and the warm core aloft has collapsed, but there are still active
spiral bands.  The low-level circulation is immense, spanning over 2000
km (1400 miles) now... somewhat reminiscent of Floyd '99.  At 15Z today,
he was located at 22.1N 90.0W and tracking NNW at 7kts.  The intensity
is rather weak at 50kts and 987mb, but that will change now that he's
over the warm Gulf waters.  The forecast calls for strengthening up to a
moderate CAT1 hurricane, and landfall in central LA Thursday morning (a
bit east of my 9/20 forecast of Lake Charles, LA).  If the storm
intensifies more rapidly than expected, the situation would become much
more serious (recall Opal '95 at this same time of year).  A Tropical
Storm Warning is in effect for the northern and western Yucatan
Peninsula, the southern coast along the Bay of Campeche, and parts of
the U.S. coast from extreme western FL panhandle westward to near
Galveston, TX.  A Hurricane Watch is in effect for the MS and LA coasts.

At 15Z, TS Kyle was located at 30.3N 51.1W and moving SW at 7kts.  This
motion will take him over warmer SSTs, allowing for intensification. 
Winds are 45kts with a MSLP of 1000mb.  The convective patterns have
improved lately, vertical shear is moderate (not negligible, but not too
strong), and as mentioned, SSTs will be warming, perhaps to the 28-28.5C
mark.  Based on this reasoning, Kyle is expected to become a hurricane
by tomorrow night.  It might be close race to see if Isidore, Kyle, or
Lili makes it to hurricane status first!

TS Lili has maintained a cold CDO the past 24 hours and is getting
better organized.  At 15Z, she was at 13.2N 65.6W and tracking W at
14kts.  She is nearly at hurricane strength with 60kt winds and 1004mb
MSLP.  The storm is expected to turn more NW-erly in the near future,
taking it over Haiti and extreme eastern Cuba.  A Tropical Storm Watch
is in effect for the western half of the southern coast of the Dominican
Republic, and it's likely that Haiti will issue a Hurricane Watch for
its southern coast shortly.

Elsewhere, there's a small tropical wave at 18N 39W.  It has a 1015mb
Low associated with it but shear is quite high (30-40 kts), so it's not
an immediate concern.

Please visit my tropical Atlantic headquarters.

23 September 2002

Isidore, Kyle, and Lili all Tropical Storms now.

At 21Z yesterday, Hurricane Isidore made landfall on the northwest
Yucatan peninsula as a very strong CAT3 storm.  From what I've heard so
far, the devastation caused by this major hurricane has been severe (2+
deaths, widespread loss of power and communication, and washed-out
roads).  Since landfall, intensity has dwindled rapidly (due to lack of
fuel source and enhanced friction) and is now a Tropical Storm.  As of
15Z today, TS Isidore was located at 20.1N 89.5W and tracking E at
2kts... this is over the central Yucatan Peninsula.  Winds have
decreased to 55kts and the central pressure has increased to 960mb.  He
is beginning what appears to be a tiny cyclonic loop over the peninsula,
and should be heading back over the same areas he hit yesterday before
heading out into the Gulf where he can re-intensify prior to another

A Tropical Storm Warning is in effect for all of the northern and
western Yucatan Penisula, as well as the southern coast along the Bay of
Campeche.  The forecast is for re-intensification once he hits the Gulf
waters, perhaps making it back to CAT3+ status, and for a northward turn
toward eastern TX or Louisiana.  From what I'm seeing, somewhere between
Lake Charles and New Orleans, LA should be on highest alert for Friday
afternoon, but that's several days out and very unofficial.  Isidore is
a HUGE storm; the circulation is at least 1500km across (nearly 1000
miles), so the watches and warnings will cover many miles of coastline
as he approaches.

TS Kyle has managed to maintain intensity despite meandering over the
same location for a while and being in moderate vertical shear.  He is a
small storm, but convection has been firing up sporadically, enough to
maintain the vortex and surface Low.  At 15Z today, he was at 32.1N
49.6W (1400km east of Bermuda) and heading WSW at 5kts.  The motion is
expected to be SW over the next few days, brining him into warmer SSTs
which should allow for gradual intensification, perhaps reaching
hurricane intensity by mid-late week.

TD13 was recently upgraded to TS Lili, the 12th named storm of the
season, at 1530Z today based on an aircraft flight into the storm.  She
has looked consistently better on satellite over the past couple days,
so it's no surprise that the plane found what it did.  At 1530Z, TS Lili
was located at 12.4N 59.7W and tracking W at 17kts.  Maximum winds are
45kts and the MSLP is 1004mb.  She is expected to continue heading W or
WNW and reach hurricane intensity by Wednesday morning.  However, my two
cents is that it may not take that long... the shear is nearly zero
there, the ocean is very warm through a deep layer along the forecast
track, and there's been vigorous convection in the storm for quite a
while now (allowing the warm core aloft to become mature and to
concentrate more vorticity).  

A Tropical Storm Watch is in effect for the northern Windward Islands
from Guadeloupe to Martinique and a Tropical Storm Warning is in effect
for the southern Windward Islands from St. Lucia to Grenada.  Lili
should pose a threat to Dominican Republic, Haiti, Cuba, and perhaps
Jamaica by the Thursday-Friday timeframe.  Then, if the storm survives
those mountainous encounters in the Greater Antilles, the southeast U.S.
may need to be very watchful.

Please visit my tropical Atlantic headquarters.

22 September 2002

Isidore nearly a CAT4 hurricane, Kyle turning tropical, TD13 forms and heading for Lesser Antilles.

Isidore has been maintaining constant intensity recently, but did
undergo rapid intensification yesterday afternoon (18mb drop between 09Z
and 21Z) as anticipated.  The track since entering the Gulf has been
following the northern coast of the Yucatan Peninsula VERY closely,
which is undoubtedly limiting him from becoming even stronger.  At 18Z
today, Isidore was at 21.7N 88.9W and heading WSW at 7kts.  Winds have
reached 110kts (gusting to 135kts) and the MSLP is 934mb.  The eye has
grown a bit since yesterday, from 18km to 37km.  A Hurricane Warning is
in effect for all of the western and northern Yucatan Peninsula.  He is
forecast to resume intensification once he pulls away from the Yucatan
coast, and with only 5kts needed to reach CAT4, that landmark should be
achieved later today.  The offical track forecast is a continued
westward drift, followed by a turn to the north, but the models are
still unable to agree on a track... anything from continuing westward
into central Mexico to as far east as the Lake Charles, LA area.  Chaos
reigns yet again.

STS Kyle continues to get better organized and is very close to being a
purely tropical system.  At 15Z today, he was located at 33.0N 49.9W
(1400km east of Bermuda) and stationary.  Max winds are 40kts and MSLP
is 1006mb.  The steering current is still very weak but he should drift
toward warmer SSTs and gradually intensify, perhaps reaching hurricane
strength by Wednesday.  There is no threat to land in the foreseeable

At 21Z yesterday, the more westward tropical wave I mentioned in the
9/21 update was upgraded to TD13 based on satellite estimates.  It has
been getting better organized and should become the 12th named storm
(Lili) in the next day or so.  As we've seen several times this season,
the low-level center is mis-aligned with the convection, so one of three
things will happen: 1) convection will blow up over the center, 2) the
center will reform under the existing convection, or 3) the system will
degenerate back to a tropical wave.  My hunch is option (1) and the
Depression will strengthen as it heads west through fairly low shear and
amply warm SSTs (next name is Lili, FYI).  Interestingly, if they can
all cooperate, by late Wednesday there could be three hurricanes in the
Atlantic: Isidore, Kyle, Lili.  As of 15Z today, TD13 was located at
12.0N 52.0W (about 850km east of southern Lesser Antilles) and tracking
W at 20kts.  Intensity is 30kts and 1007mb.  Microwave imagery shows
some band-ish features, but it is not well organized.  The track
forecast is to pass through the central Lesser Antilles on Monday night
and Hispaniola Wednesday night.

The more eastward tropical wave I mentioned in yesterday's update has
made very little progress except in location.  It has a 1011mb Low
associated with it and is moving west at 12kts.  It's presently at about
12N 33W.  Assuming TD13 becomes TS Lili, the next number/name on deck is

Please visit my tropical Atlantic headquarters.

21 September 2002

Isidore becomes first major hurricane, Kyle forms in central Atlantic, east Atlantic becomes active.

Hurricane Isidore was upgraded to a CAT3 storm today at 15Z, making him
the first major hurricane of the season.  He made landfall on western
Cuba yesterday (with devastating rains exceeding 2' at places), and is
now entering the southeastern Gulf.  Shear is almost zero, and the SSTs
are actually an interesting story (as relayed by the Air Force C-130
crew): Although he's moving slowly and the Gulf waters are much
shallower than the Caribbean, cold upwelling is not a concern because of
the loop current... an ocean current that runs between Cuba and Yucatan,
through the Gulf, then exits between Florida and Cuba. Very warm (30C)
SSTs are constantly being replenished so this may be a setup for rapid
intensification today.

The satellite imagery now reveals a mature storm, not only at microwave
frequencies, but also at VIS and IR.  The eye will fully develop today
and the intimidation factor goes up with that, knowing that he will make
landfall somewhere along the Gulf coast.

At 15Z, Hurricane Isidore was located at 21.9N 86.1W and tracking W at
6kts.  The maximum sustained winds have reached 100kts and the MSLP is
down to 955mb, as observed by aircraft recon.  A Tropical Storm Warning
is still in effect for extreme western Cuba and the Isle of Youth, and a
Hurricane Warning is in effect for northern and eastern Yucatan
Peninsula.  The forecast is for a continued westward heading through
Tuesday, with winds of 125kts (strong CAT4).  It's still VERY difficult
to say what's in store for him after Tuesday; the trough off the US West
coast now could be a big player in either leaving him behind again to
linger in the Gulf or to pick him up and steer the storm more
northward.  It's simply too early to tell.

At 21Z yesterday, the broad circulation I mentioned ESE of Bermuda was
upgraded to Subtropical Depression 12, then at 09Z today, it was
upgraded again to Subtropical Storm Kyle, the 11th named storm of the
season.  It should be switch over Tropical Storm today, as it's
continually taking on more and more tropical characteristics (mature
warm core, banding, anticyclone aloft, CDO, etc).  At 15Z today, STS
Kyle was located at 31.3N 51.6W and heading N at 6kts.  Current intesity
based on satellite estimates is 40kts and 1007mb.  Given the relatively
light shear and moderate SSTs in his environment, further
intensification is likely, and the offical NHC forecast brings him to
hurricane status by Tuesday morning.  Also, he's in very weak steering
flow, so the current position may well be close to the position on

There's a very healthy tropical wave at 12N 44W (about 1750km east of
the Windward Islands) and tracking W at 12kts.  It has a 1010mb Low
associated with it, has been getting better organized, and conditions
are favorable for further development.  It could become TD13 in the near
future (and the next name on deck is Lili).

Yet another interesting wave near the Cape Verde Islands is at 10N 24W
and tracking W at 13kts.  It also has a 1010mb Low with it, but
convection is not as vigorous as with the first wave, so if anything
would be slower to develop (if it does at all).

Please visit my tropical Atlantic headquarters.

20 September 2002

Isidore now a CAT2 hurricane.

At the time of yesterday's update, Isidore was a strong TS with a
strengthening trend.  At 21Z yesterday, he reached CAT1 hurricane
status, then at 09Z this morning, reached CAT2 status and low-lying
western Cuba (and perhaps the occassional eyewall cycle as the storm
re-organizes itself) is the only obstacle to further intensification. 
The central pressure fell 24mb in the past 24 hours, and 13mb in the
past 12 hours, so the storm is not yet qualified as a rapid intensifier
(some classic definitions can be found at
http://www.nrlmry.navy.mil/~chu/chap6img/tab605.jpg and a recent
definition is +31kts/day).  It will be a very well-observed storm, in
terms of aircraft flights into it, around it, and sensors dropped into
the ocean ahead of and under it to record the depth of the warm water
(latest data I know about: 29.5-30.0C running 50-70m deep, which is an
amazing amount of energy available to this storm... it's a 230'-deep
bathtub of 86F water!).

At 15Z, Hurricane Isidore was located at 21.7N 83.9W (just a few miles
south of the western Cuban coast) and tracking WNW at 7kts.  Maximum
sustained winds have reached 90kts and the MSLP has plummeted to 966mb. 
Despite this low pressure, an eye has still not formed, which from my
experience is very unusual.  However, the inner precipitation structure
-as revealed by microwave imagery- does contain the mature features one
would expect from a 966mb hurricane.  A Hurricane Watch and Tropical
Storm Warning are in effect for most of the northern Yucatan Peninsula,
and a Hurricane Warning is still in effect for all of western Cuba and
the Isle of Youth.  You can download and print a high-quality map from
http://www.mcwar.org/gallery/tropics/MESO_atl_blank.ps if you wish to
keep up with the track of this storm.

The intensity forecast is relatively simple for a change... it's in a
very favorable environment; both the ocean and the atmosphere will allow
it to continue to strengthen, seemingly well into the major hurricane
domain (CAT3, 4, 5).  However, just to keep the operational tropical
meteorologists on their toes, the track is far from trivial to
forecast.  It's hard to find even two models that agree on a scenario,
let alone a concensus.  The official NHC forecast takes Isidore past
western Cuba and into the southeastern Gulf, slows him down and turns
him slightly westward.  This seems quite reasonable, and after that, I
suspect some sort of a more northward turn, in advance of the next
trough.  At any rate, the US is not threatened by landfall through the
end of the weekend, but should be taking that time to prepare for one.

Elsewhere, there's a broad circulation with a 1010mb Low and scattered
deep convection at about 29N 52W (roughly 14 degrees ESE of Bermuda). 
Vertical shear is fairly low, the SST there is about 27.5C, but the
vorticity center is still somewhat elongated and disorganized.  Should
this develop, the next number/name on deck is 12/Kyle.

Please visit my tropical Atlantic headquarters.

19 September 2002

Isidore quickly organizing, Josephine becomes extratropical.

Tropical Storm Isidore is on the brink of being the second hurricane of
the season.  After a slow start during the weekend and the first half of
this week, the center is finally under the coldest part of the CDO, and
recent microwave imagery shows very clearly that an eye has formed in
the precipitation.  It will be just a matter of hours until that is
reflected in the VIS and IR imagery.

As of 15Z today, Isidore is located at 20.0N 81.2W and moving WNW at
7kts.  maximum sustained winds are 60kts and MSLP is 990mb (which
coincides nicely with the classic pressure for eye development in the
Atlantic).  Aircraft recon will investigate the storm later today, and
given the trend in organization and the microwave imagery showing an
eye/eyewall, they should find 65-70kt winds, making it a hurricane at
that time.  The outflow is expanding in all directions, showing that the
anticyclone aloft is maturing and assisting the "exhaust" process.

A Tropical Storm Warning is still in effect for the Cayman Islands, a
Hurricane Watch is in effect for some of central Cuba, and a Hurricane
Warning is in effect for the western half of Cuba.  It is possible that
a Tropical Storm Warning will be issued for the Florida Keys this
afternoon if the storm's wind field keeps expanding.

The forecast is very difficult, and I'd say that the track forecast is
more challenging than the intensity this time.  Cuba will be the only
hinderance to intensification, and western Cuba is fairly flat, but land
nonetheless.  All other signs point to a major hurricane in a few
days... low shear, very deep, warm ocean to travel over, and time.  The
models are in good agreement about taking it over the western tip of
Cuba, but beyond that, the concensus is to stall it in the southeast
Gulf (with several outliers).  Depending very strongly on what the
trough does that's draped over the central US now, Isidore could end up
hitting anywhere from Mexico to Florida, so everyone along the Gulf
coast needs to be be watchful and prepared.  So far, the track is
reminiscent of Helene '00 (and she hit the western end of the Florida
peninsula), just FYI.

TS Josephine was absorbed by the mid-latitude trough this morning and
advisories have ceased as of 15Z (11am EDT).  The last advisory placed
Josephine at 42.5N 42.5W and tracking NE at 27kts, basically with the
front.  Intensity was 50kts and 1004mb (definite baroclinic

Please visit my tropical Atlantic headquarters.

18 September 2002

TD10 upgraded to Isidore, TD11 forms and is upgraded to Josephine.

At 06Z today, TD10 was upgraded to TS Isidore based on aircraft recon
into the storm.  The convection with Isidore has been amazing, even as
an open tropical wave a couple days ago, but the CDO has been
persistently cold during the revived TD state and now the TS state. 
Although the low-level center is still somewhat disorganized, the middle
and upper levels are certainly well-developed, and I suspect that if the
low-level center shifts under the center of the CDO, the storm could
rapidly become strong.  

Sadly, the TMI, AMSU, and SSM/I microwave sensors on various satellites
all missed the center of the system as of the latest passes, each
catching it on the edge of a swath.  This CDO-dominated stage of
development in a Tropical Storm is best observed with microwave imagery,
because it can "see" through the CDO and into precipitating regions
below... able to pick up spiral bands or even an eyewall long before VIS
or IR can.

The 15Z advisory positions TS Isidore at 17.9N 78.7W (just 50km away
from Jamaica) and moving NW at 6kts.  Maximum sustained winds are still
fairly weak at 40kts, and the MSLP is 1003mb.  Given the present
appearance of the storm, one might expect it to gain intensity a bit
more quickly than it has been.

A Tropical Storm Watch is in effect for Grand Cayman, a Tropical Storm
Warning is in effect for Little Cayman, Cayman Brac, and Jamaica, and a
Hurricane Watch is in effect for western Cuba and the Isle of Youth. 
Isidore is forecast to reach hurricane strength by the time he reaches
Cuba early Friday morning.  Once crossing Cuba and entering the Gulf of
Mexico, more rapid intensification is likely, as the shear is already
moderately low and expected to lessen, and the SSTs there are very

As a long-range possibility, one model (GFDL) has the storm hitting the
Gulfport, MS area on Sunday evening as a strong CAT2 hurricane.  NOGAPS
hints more at a curve toward Brownsville, TX, while UKMET and MM5
virtually stall it in the central Gulf and intensify it substantially. 
A Canadian model has it hitting the Florida panhandle on Saturday
night.  Unfortunately, the latest AVN run is not in at the time of
writing, and they've done well with tracks this season.  So, the point
is, Cuba should certainly be prepared for rough weather, but the entire
Gulf coast should be watching this very closely because the future track
is so uncertain.

The small vortex I mentioned in yesterday's update was upgraded to TD11
at 21Z yesterday based on satellite estimates.  At 09Z today, a
fortuitous (for us, not them) ship observation very near the center of
TD11 found 37kt sustained winds, and in combination with satellite
estimates, was enough to warrant upgrading it to TS Josephine, the 10th
named storm of the season.  It is very small, weak, and no threat to
land.  As of 15Z today, she was located at 36.0N 51.4W and tracking NNE
at 10kts.  Intensity is 35kts and 1009mb, and it should maintain current
intensity or weaken as a trough advances on it and it becomes

Please visit my tropical Atlantic headquarters.

17 September 2002

TD10 regenerates in central Caribbean Sea, could have a big future.

At 15Z today, advisories resumed on what was TD10 during the weekend. 
It had degenerated into an open wave on 9/15 at 21Z (as determined by
visible satellite imagery and aircraft recon).  Those same tools
assisted in the recent upgrade.  Deep convection has been remarkably
persistent with this system, and there are now hints of what appear to
be spiral bands trying to form.  It is in 15kts of deep (850mb-200mb) SW
shear, and the convection being displaced to the east of the center is
one visible sign of that.  The SST there is slightly warmer than 29C,
which no doubt has been the cause for the copious convection (a lot of
heat energy being supplied to it). 

The 15Z advisory places TD10 at 15.9N 77.2W (about 200km south of
Jamaica) and tracking W at 8kts.  Intensity is 30kts and 1009mb.  The
shear is marginal for development now, but it is expected to lessen, and
combined with the warm SSTs, the storm should be able to intensify in
the near future.  Interaction with the mountains islands in the Greater
Antilles may keep intensity down for the next couple days, but the
forecast track is one that could give the storm plenty of time to get
its act together.  NHC's forcast is for gradual strengthening, just shy
of being a hurricane by Friday morning, but admitedly, a lot can change
in a short time.  If named, it will become Isidore.

Nearly all models keep the center south of the Greater Antilles (which
include Cuba, Jamaica, Hispaniola, and Puerto Rico) and take the system
between the Yucatan Peninsula and Cuba by Friday morning... in perfect
position to enter the Gulf.  GFDL is perhaps the quickest landfall model
of the bunch, bringing the storm into the Tampa area on Saturday
afternoon as an 80kt hurricane.  If the storm is allowed to enter the
Gulf, it would spend 2-3 days there (basically the weekend) and perhaps
strengthen quite a bit.  Remember the key to hurricanes in the Gulf...
they almost have to make landfall somewhere.

Elsewhere, there is a vortex with moderate convection at 33N 53W. 
Although formed from an upper-level Low, a surface circulation has
developed and is taking on tropical characteristics.  If further
development occurs, it could be upgraded to TD11.

Please visit my tropical Atlantic headquarters.

16 September 2002

Hanna made landfall, TD10 came and went, but is making a comeback.

Tropical Storm Hanna made landfall on the MS/AL border at 14Z on 9/14
with about 40kt winds and disorganized rain (not an eyewall or
well-formed spiral bands).  The only serious effect besides some minor
localized flooding was a wind damage report from Mobile County, AL where
"several trees were blown down".

TD10 formed at 21Z on 9/14 just a few miles off the eastern Venezuela
coast (very low latitude), tracked WNW along the coast and dissipated
back to an open wave on the 15th at 21Z.  However, it is still a key
feature in the Caribbean, presently located at about 17N 73W (just south
of Hispaniola) and moving WNW at 20kts.  Conditions are now marginal for
development, but will be improving, so re-development is possible, or
even likely.  It's still got a lot of centralized relative vorticity
(curvature of the wind field), moderate upper-level divergence (decent
outflow), and healthy convection.  As evident in the satellite imagery,
there is westerly shear, caused by an upper-level anticyclone to its

The computer models are largely in agreement on strengthening it, and
also that Cuba and then Florida will be the target areas.  Some keep it
fairly weak, while some have it reaching CAT3 status by landfall (for
those of you who check the models closely, the 06Z run of GFDL was
fascinating, ask for details if interested).  Should this reform and
reach TS strength, the next name on deck is Isidore.  Just FYI, the last
time we had Isidore was in late September 1996, and she reached CAT3
intensity in the central Atlantic.  The time before that was early-mid
September 1990, and she reached CAT2 intensity, also tracking northward
across the central Atlantic. 

Please visit my tropical Atlantic headquarters.

13 September 2002

Gustav becomes extratropical, Hanna forms, central and eastern Atlantic still interesting.

Following the forecast remarkably well, Gustav continued on his rapid
northeasterly course on Wednesday afternoon and Thursday, making
landfall on extreme eastern Nova Scotia then southwest Newfoundland as a
weak CAT 1 hurricane early Thursday morning.  He transitioned to
extratropical and was absorbed into the mid-latitude trough just hours
after the landfalls.

TD9 formed late Wednesday night from that large area of disturbed
weather in the central Gulf that I had been mentioning for a few days. 
Despite poor convective patterns, pressures had dropped persistently,
reaching 1001mb by the time it was classified as a TD (very low for that
point in its lifetime!).  At 09Z today, aircraft recon, buoys, and
nearby ships reported that the sustained winds were strong enough to
upgrade it to TS Hanna, the 8th named storm of the season.  The 12Z
advisory places Hanna at 27.0N 88.6W and tracking NW at 7kts (after
having been meandering for a while).  Intensity is observed to be 40kts
and 1002mb.  A Tropical Storm Watch is in effect for the eastern half of
the Florida peninsula, and a Tropical Storm Warning is in effect from
the central FL panhandle over to near New Orleans, LA... see
http://www.nhc.noaa.gov/ftp/graphics/AT09/AL0902W.GIF for details.  

Currently, the convection is all southeast of the low-level center (due
to 10-15kt NW vertical shear), which has been exposed during much of
Hanna's existence.  The forecast is for further strengthening, and for
accelerated motion.  As the intensity picks up to 50kts, the track
should bend more northerly, and the NHC forecast is for landfall near
the FL/AL border (although many computer models predict landfall a bit
further west, like Gulfport) Saturday around noon local time. 

The tropical wave that I positioned at 19N 31W in the 9/11 update is now
at about 14N 52W (at least that's where the convection is occuring now,
the wave itself is much larger and harder to pin down) and becoming
better organized as it clears the high-shear zone.  It's tracking W at
20kts and has the potential to become TD10 during the weekend.

Lastly, there's a tropical wave just exiting the African coast with
fairly high, concentrated vorticity that is only in moderate shear. 
Though it lacks deep convection now, it will be closely watched the next
few days.
Please visit my tropical Atlantic headquarters.