Dr. Jeff Masters' WunderBlog

Tropical Depression Six forms; Henrietta lashes the Pacific coast of Mexico

By: JeffMasters, 08:57 PM GMT am 31. August 2007

The Hurricane Hunters have reported back from the tropical wave about 200 miles east of the southern Lesser Antilles Islands and found a closed circulation and 35 mph winds, good enough to justify naming this system Tropical Depression Six. This storm has the potential to become a hurricane in the Caribbean by Sunday or Monday. Visible satellite loops show a considerable improvement in organization occurred today, with low level spiral bands wrapping around the center of circulation, and upper level outflow now visible on both the north and south sides.

Water vapor satellite loops show the presence of a large amount of dry air to the north of TD 6, but the storm has generated enough thunderstorm activity to moisten the surrounding environment, which should speed development. Wind shear is a favorable 10 knots over TD 6, and is expected to remain below 10 knots for the next four days over the storm's expected path through the Caribbean, according to the 18Z SHIPS model. As seen in the latest microwave satellite image from 12:49pm EDT (Figure 1), the heaviest thunderstorm activity is on the south side of the system. Trinidad and Tobago will get the heaviest rains from TD 6, although the northern coast of South America plus the island of Grenada could also see some heavy rains.


Figure 1. Microwave satellite image of TD 6 taken at 12:49pm EDT 8/31/07. The heaviest rain (red colors) is just east of the island of Tobago. Image credit: Navy Research Lab.

Both the 12Z GFDL and the 18Z SHIPS intensity models forecast that TD 6 will intensify into a hurricane by 72 hours from now, as it tracks through the Central Caribbean. The model consensus has a track just north of the ABC islands (Aruba, Bonaire, and Curacao), then west-northwestward through the Caribbean. By Monday night, most of the models have TD 6 approaching the Honduras/Nicaragua border. The GFDL is further north, taking TD 6 into Belize on Wednesday as a Category 2 hurricane. The future strength of TD 6 depends critically upon how close it passes to the South American coast over the next two days. The Southeastern Caribbean just north of the South American coast is a climatologically unfavorable region for tropical cyclones, as they tend to pull dry continental air off of South America into their circulations. Many tropical cyclones passing through this region of the Caribbean die or become severely weakened. If TD 6 does survive and pass into the Central Caribbean, a more northwesterly turn to threaten Haiti, Jamaica, and Cuba is not out of the question. However, a strong ridge of high pressure is expected to build in over the region, and I don't see any troughs of low pressure capable of swinging TD 6 to the north coming.

The next Hurricane Hunter mission will be 8am EDT Saturday. The NOAA jet's first flight will be Sunday morning.

Links to follow today:
Martinique radar
Barbados weather
Trinidad Crown Point weather
Piarco, Trinidad weather
Grenada weather

96L
An area of low pressure ("96L") a few hundred miles south of Nantucket, Massachusetts got close to tropical depression status today. However, wind shear from strong upper level westerly winds has almost completely removed all heavy thunderstorm activity from the storm, and 96L has missed its chance to become a tropical depression.

Pacific coast of Mexico at risk from Henriette
Tropical Storm Henriette formed off the Pacific coast of Mexico this morning, and poses a threat to Mexican coast from Acapulco to Puerto Vallarta, as well as the Baja Peninsula. Acapulco radar shows that spiral banding around the center has increased today, and visible satellite loops show a rather disorganized system, with the beginnings of some upper-level outflow to the south. Henriette has brought heavy rains and sustained winds of 25 mph to Acapulco today, but the weather there will improve tonight as the storm tracks west-northwest, parallel to the coast. Wind shear is a rather stiff 20 knots over Henriette today, which should keep any development slow. By tomorrow, wind shear should decrease to 10-15 knots, allowing more rapid strengthening. Residents of Manzanillo should be prepared for tropical storm force winds Saturday evening, and the airport there will probably close for a time late Saturday. Puerto Vallarta is further from Henrietta's projected path and more sheltered, and will probably not get tropical storm force winds. Those planning on being in Baja Monday and Tuesday should keep a close eye on Henriette, as hurricane conditions may arrive late Monday.

Links to watch for Henriette
Acapulco radar
Manzanillo, Mexico observations

My next update will be Saturday by noon EDT.

Jeff Masters

Permalink

Hurricane Hunters find a tropical depression forming; Henriette lashes Pacific coast of Mexico

By: JeffMasters, 08:08 PM GMT am 31. August 2007

The Hurricane Hunters have reported back from the tropical wave about 200 miles east of the southern Lesser Antilles Islands, 94L, and found a closed circulation with top winds of 30 mph. This qualifies 94L as a tropical depression, and NHC will likely upgrade it to Tropical Depression Six in their 5pm update. This storm has the potential to become a hurricane in the Caribbean by Sunday or Monday. Visible satellite loops show a considerable improvement in organization occurred today, with low level spiral bands wrapping around the center of circulation, and upper level outflow now visible on both the north and south sides.

Water vapor satellite loops show the presence of a large amount of dry air to the north of 94L, but the storm has generated enough thunderstorm activity to moisten the surrounding environment, which should speed development. Wind shear is a favorable 10 knots over 94L, and is expected to remain below 10 knots for the next four days over the storm's expected path through the Caribbean, according to the 18Z SHIPS model. As seen in the latest microwave satellite image from 12:49pm EDT (Figure 1), the heaviest thunderstorm activity is on the south side of the system. Trinidad and Tobago will get the heaviest rains from 94L, although the northern coast of South America plus the island of Grenada could also see some heavy rains.


Figure 1. Microwave satellite image of 94L taken at 12:49pm EDT 8/31/07. The heaviest rain (red colors) is just east of the island of Tobago. Image credit: Navy Research Lab.

Both the 12Z GFDL and the 18Z SHIPS intensity models forecast that 94L will intensify into a hurricane by 72 hours from now, as it tracks through the Central Caribbean. The model consensus has a track just north of the ABC islands (Aruba, Bonaire, and Curacao), then west-northwestward through the Caribbean. By Monday night, most of the models have 94L approaching the Honduras/Nicaragua border. The GFDL is further north, taking 94L into Belize on Wednesday as a Category 2 hurricane. The future strength of 94L depends critically upon how close it passes to the South American coast over the next two days. The Southeastern Caribbean just north of the South American coast is a climatologically unfavorable region for tropical cyclones, as they tend to pull dry continental air off of South America into their circulations. Many tropical cyclones passing through this region of the Caribbean die or become severely weakened. If 94L does survive and pass into the Central Caribbean, a more northwesterly turn to threaten Haiti, Jamaica, and Cuba is not out of the question. However, a strong ridge of high pressure is expected to build in over the region, and I don't see any troughs of low pressure capable of swinging 94L to the north coming.

The next Hurricane Hunter mission will be 8am EDT Saturday. The NOAA jet's first flight will be Sunday morning.

Links to follow today:
Martinique radar
Barbados weather
Trinidad Crown Point weather
Piarco, Trinidad weather
Grenada weather

96L
An area of low pressure ("96L") a few hundred miles south of Nantucket, Massachusetts got close to tropical depression status today. However, wind shear from strong upper level westerly winds has almost completely removed all heavy thunderstorm activity from the storm, and 96L has missed its chance to become a tropical depression.

Pacific coast of Mexico at risk from Henriette
Tropical Storm Henriette formed off the Pacific coast of Mexico this morning, and poses a threat to Mexican coast from Acapulco to Puerto Vallarta, as well as the Baja Peninsula. Acapulco radar shows that spiral banding around the center has increased today, and visible satellite loops show a rather disorganized system, with the beginnings of some upper-level outflow to the south. Henriette has brought heavy rains and sustained winds of 25 mph to Acapulco today, but the weather there will improve tonight as the storm tracks west-northwest, parallel to the coast. Wind shear is a rather stiff 20 knots over Henriette today, which should keep any development slow. By tomorrow, wind shear should decrease to 10-15 knots, allowing more rapid strengthening. Residents of Manzanillo should be prepared for tropical storm force winds Saturday evening, and the airport there will probably close for a time late Saturday. Puerto Vallarta is further from Henrietta's projected path and more sheltered, and will probably not get tropical storm force winds. Those planning on being in Baja Monday and Tuesday should keep a close eye on Henriette, as hurricane conditions may arrive late Monday.

Links to watch for Henriette
Acapulco radar
Manzanillo, Mexico observations

I'll edit this blog when the 5pm advisory comes out. Next full update will be Saturday by noon EDT.

Jeff Masters

Updated: 08:10 PM GMT am 31. August 2007

Permalink

Two Atlantic storms are near tropical depression strength

By: JeffMasters, 02:23 PM GMT am 31. August 2007

The tropical wave about 200 miles east of the southern Lesser Antilles Islands, 94L, continues to become better organized, and will probably become a tropical depression today. This system has the potential to become a hurricane in the Caribbean Sea early next week. Visible satellite loops show a steadily organizing system, with low level spiral bands trying to form, and upper level outflow to the north, visible as cirrus clouds wafting from west to east. QuikSCAT winds from 5:36am EDT this morning showed that 94L's circulation was not well-defined yet. Sustained winds of 25-35 knots were observed, and some wind gusts to 55 mph are probably occurring on the south side of 94L's circulation. Water vapor satellite loops show the presence of a large amount of dry air to the north of 94L, but the storm has generated enough thunderstorm activity to moisten the surrounding environment, which should speed development. Wind shear is a favorable 5-10 knots over 94L, and may decrease some by Saturday, allowing further development. As seen in the latest microwave satellite image of 94L (Figure 1), the heaviest thunderstorm activity is on the south side of the system. Trinidad and Tobago will get the heaviest rains from 94L, although the northern coast of South America plus the island of Grenada could also see some heavy rains.


Figure 1. Microwave satellite image of 94L taken at 6:07am EDT 8/31/07. Image credit: N avy NRL Research Lab.

None of the reliable computer models forecast that 94L will develop into a tropical depression, but it look like the models will be wrong on this. The usually unreliable Canadian model appears to have the best idea on what 94L will do. This model gradually strengthens 94L as it passes along the northern coast of South America. By Monday night, the Canadian model has 94L approaching the Honduras/Nicaragua border. Yesterday afternoon's run of the GFDL model had a similar solution, and predicted 94L would be a Category 2 hurricane at that time. The future strength of 94L depends critically upon how close it passes to the South American coast over the next two days. The Southeastern Caribbean just north of the South American coast is a climatologically unfavorable region for tropical cyclones, as they tend to pull dry continental air off of South America into their circulations. Many tropical cyclones passing through this region of the Caribbean die or become severely weakened. If 94L does survive and pass into the Central Caribbean, a more northwesterly turn to threaten Haiti, Jamaica, and Cuba is not out of the question.

The Hurricane Hunters will be in the storm this afternoon. The NOAA jet's first flight will be Sunday morning.

Links to follow today:
Guadaloupe radar
Barbados weather
Trinidad Crown Point weather
Piarco, Trinidad weather
Grenada weather

96L
An area of low pressure ("96L") a few hundred miles south of Nantucket, Massachusetts is close to tropical depression status. The storm has a well-defined surface circulation, but strong upper-level winds from the southwest have kept heavy thunderstorm activity confined to 96L's northeast side. Low-level spiral bands on its northeast side are impacting Cape Cod and Nantucket today, as seen on long-range radar out of Boston. QuikSCAT saw winds as high as 30 knots (35 mph) in this morning's pass. The combined wind and pressure plot from buoy 44004, 200 nm east of Cape May, NJ, shows the passage of 96L nicely.

96L has about a day before increasing wind shear and cooler sea surface temperatures convert the storm into an extratropical system. The eastern tip of Nova Scotia and the south coast of Newfoundland could experience tropical storm-force winds from this extratropical system over the weekend.

Coast of Africa
None of the computer models are forecasting the development of a tropical depression during the coming week off the coast of Africa.

I'll post an update later today.
Jeff Masters

Updated: 02:36 PM GMT am 31. August 2007

Permalink

Lesser Antilles wave gets better organized; weather radio recall ordered

By: JeffMasters, 04:45 PM GMT am 30. August 2007

The tropical wave about 500 miles east of the Lesser Antilles Islands, 94L, has become better organized today, and has the potential to become a tropical depression on Friday. Visible satellite loops show a limited amount of thunderstorm activity, but the circulation has gotten much better defined in the past few hours. Low level cumulus clouds are now starting to spiral into the center, and there is the beginning of a spiral band with heavy rain trying to form on the southeast side of 94L. This process is being slowed by the presence of a large, dusty area of dry air on its northwest side, and thunderstorm activity will be slow to build on this northwest side over the next day or two.

Water vapor satellite loops of the region show that 94L has not significantly moistened its environment. However, now that it is better organized, it should be able to build more thunderstorm activity and moisten its surrounding environment. Wind shear is a favorable 5-10 knots over 94L, and may decrease some by Saturday, allowing further development. It looks like 94L will stay well south of the shearing winds of the upper-level low spinning north of Puerto Rico.

94L has slowed down its forward speed to 10-15 mph, so should be entering the Lesser Antilles Islands on Friday night, bringing gusty winds and heavy rain. None of the reliable computer models forecast that 94L will develop into a tropical depression, but it now appears that it has a legitimate chance of becoming one Friday or Saturday. Due to the dry air, I think it unlikely 94L will be stronger than a 50-mph tropical storm when it passes through the Lesser Antilles Islands. The southern portion of the islands will get the most rain, due to the dry air on the north side of 94L. The Hurricane Hunters are scheduled to investigate on Friday afternoon.


Figure 1. Infrared satellite image showing the various disturbed area of weather in the Atlantic today.

95L and 96L
An area of low pressure ("95L") about 400 miles east of Jacksonville, Florida has an elongated, closed circulation, but very little heavy thunderstorm activity. A separate low ("96L") is about 400 miles east of Cape Hatteras, North Carolina, and has a well defined closed circulation. Heavy thunderstorm activity is limited to a small glob of convection on the southeast side of the circulation, due to strong northwesterly winds aloft creating 10-20 knots of wind shear. QuikSCAT data from 6:08am EDT this morning shows this circulation nicely. Top winds seen by QuikSCAT were 30 knots (35 mph).

The North Carolina disturbance, 96L, is becoming the dominant system, which will make 95L's development slow, since the two systems are so close together. We can pretty much throw out the model predictions for these disturbances, since none of the models properly initialized two low pressure systems in this region. 96L (the North Carolina low) has developed some upper level outflow, and may become a tropical depression tomorrow if the wind shear relaxes a little and allows thunderstorm activity to build on more than just the southeast side of the storm. This system should get caught up in the jet stream and whisked out to sea to the northeast, and at present does not appear to be a threat to any land areas except the Canadian Maritime provinces.

It is possible that outflow from 96L will choke off and kill 95L. If 95L survives, it may have a chance to develop into a tropical depression once 96L clears out. It is less clear what the track of 95L might be, since it is further south where steering currents are weak. I expect 95L may wander erratically for a few days.

There is a Hurricane Hunter mission scheduled for this afternoon to investigate 96L. No missions are scheduled for Friday into this system.

Mexico is watching 97L
An area of low pressure ("97L") is bringing heavy rains to the same part of the Mexican coast inundated by Hurricane Dean. 97L has only about six more hours over water before it moves ashore, and probably does not have enough time to become a tropical depression before doing so.

Coast of Africa
None of the computer models are forecasting the development of a tropical depression during the coming week off the coast of Africa. There is a large surge of moisture with tropical waves embedded in it coming off the coast of Africa this week, so we will need to continue to monitor this area for development.

Weather radio recall
There's a recall of 66,000 Oregon Scientific Weather Radios that was announced today by the U.S. Consumer Product Safety Commission. The radios could fail to receive National Weather Service alert signals in certain areas of the country. In the event of severe weather, this failure could put a consumer's life and property at risk.

Jeff Masters

Updated: 06:06 PM GMT am 30. August 2007

Permalink

Lots of disturbances, but not much threat

By: JeffMasters, 02:12 PM GMT am 30. August 2007

You can tell it's getting near peak hurricane season, since we are tracking four separate areas of disturbed weather that NHC has labeled as "Invests." None of these disturbances are a major concern at present, but we will need to watch them closely.

Not much has changed from yesterday with 94L, the tropical wave about 600 miles east of the Lesser Antilles Islands. The wave has slowed down its forward speed to 10-15 mph, so should be entering the islands on Saturday, bringing gusty winds and heavy rain. QuikSCAT data from last night showed that the storm became less organized yesterday, with a weak, elongated circulation. Top winds were 20 knots (23 mph). Visible satellite loops show a limited amount of disorganized thunderstorm activity. Wind shear is still a favorable 5-10 knots over 94L, and may decrease some by Saturday, allowing a better chance for development. It looks like 94L will stay well south of the shearing winds of the upper-level low spinning north of Puerto Rico.

The presence of a large, dusty area of dry air surrounding its north side continues to cause major problems for 94L. Once again today, this dry air is being sucked into 94L's thunderstorms. This dry air is creating strong downdrafts visible as arc-shaped surface cumulus clouds along the ocean surface. The presence of these arc-shaped surface clouds is usually a good sign that a storm is struggling with dry air and will not intensify significantly for at least the next 12 hours.

Water vapor satellite loops of the region show that 94L has not significantly moistened its environment. As the storm continues further west, it may be able to gradually do so, allowing it more of a chance to get organized. None of the reliable computer models forecast that 94L will develop into a tropical depression. The Hurricane Hunters are scheduled to investigate this system on Friday afternoon.


Figure 1. Infrared satellite image showing the various disturbed area of weather in the Atlantic today.

95L and 96L
An area of low pressure ("95L") about 400 miles east of Jacksonville, Florida has an elongated, closed circulation, but very little heavy thunderstorm activity. A separate low ("96L") is about 400 miles east of Cape Hatteras, North Carolina, and has a well defined closed circulation with plenty of heavy thunderstorm activity. QuikSCAT data from 6:08am EDT this morning shows this circulation nicely. Top winds seen by QuikSCAT were 30 knots (35 mph). Wind shear over both disturbances was 10-20 knots, and some development of both systems is possible today. It appears now that the North Carolina disturbance, 96L, is becoming the dominant system, which will make 95L's development slow, since the two systems are so close together. We can pretty much throw out the model predictions for these disturbances, since none of the models properly initialized two low pressure systems in this region. 96L (the North Carolina low) has developed some upper level outflow and low-level spiral bands, and this storm may become a tropical depression today or tomorrow. It should get caught up in the jet stream and whisked out to sea to the northeast, and at present does not appear to be a threat to any land areas except the Canadian Maritime provinces.

It is possible 96L will grow into a strong tropical storm who's outflow will choke off and kill 95L. If 95L survives, it may have a chance to develop into a tropical depression once 96L vamooses. It is less clear what the track of 95L might be, since it is further south where steering currents are weak. I expect 95L may wander erratically for a few days.

There is a Hurricane Hunter mission scheduled for this afternoon to investigate 96L. No missions are scheduled for Friday into this system.

Mexico is watching 97L
An area of low pressure ("97L") is bringing heavy rains to the same part of the Mexican coast inundated by Hurricane Dean. 97L has only about 12 more hours over water before it moves ashore, but may become organized enough to become a tropical depression before doing so.

Coast of Africa
None of the computer models are forecasting the development of a tropical depression during the coming week off the coast of Africa. There is a large surge of moisture with tropical waves embedded in it coming off the coast of Africa this week, so we will need to continue to monitor this area for development.

Jeff Masters

Updated: 02:45 PM GMT am 30. August 2007

Permalink

Katrina's 2nd anniversary, and the tropical update

By: JeffMasters, 02:31 PM GMT am 29. August 2007

A tropical wave midway between Africa and the Lesser Antilles (94L) has changed little since yesterday. QuikSCAT data from 4:47am EDT this morning shows a poorly organized system with a weak, elongated circulation. Top winds were 25 knots (29 mph). Visible satellite loops show a limited amount of disorganized thunderstorm activity. Wind shear is a favorable 5-10 knots over 94L, and should not be a problem for it until Friday or Saturday. By then, 94L will be moving through the Lesser Antilles Islands, and may encounter high wind shear if it is far enough north to feel the winds of an upper-level low pressure system that will be just north of Puerto Rico.

The presence of a large, dusty area of dry air surrounding its north side is the main thing holding back 94L. This dry air is being sucked into the circulation and is interfering with the storm's organization. When the dry air encounters a thunderstorm inside 94L, this denser dry air gets incorporated into the thunderstorm's downdraft, accelerating the downdraft, and creating arc-shaped surface cumulus clouds that mark the downdraft's position as it spreads out along the ocean surface (Figure 1). The presence of these arc-shaped surface clouds is usually a good sign that a storm is struggling with dry air and will not intensify significantly for at least the next 12 hours.

Water vapor satellite loops of the region show that 94L has not significantly moistened its environment. As the storm continues further west, it should be able to gradually do so, allowing it more of a chance to get organized. The system should track through the Lesser Antilles Islands on Friday, which is the earliest day I expect it could become a tropical depression. None of the reliable computer models make a believable forecast showing 94L developing into a tropical depression before it reaches the Lesser Antilles. The GFDL develops 94L into a tropical storm once it makes it into the central Caribbean south of the Domincan Republic, and this is a believable forecast, if 94L hangs together and makes it into the central Caribbean. The Hurricane Hunters are on call to investigate 94L on Friday.


Figure 1. Visible satellite image of 94L, show arc-shaped outflow boundaries from thunderstorm downdrafts.

South Carolina low
An area of low pressure has developed a few hundred miles off the South Carolina coast, along an old frontal boundary. This disturbance has been designated "95L" by NHC this morning. QuikSCAT showed a sharp wind shift but no closed circulation around 95L this morning at 6:34am EDT, and measured winds as high as 50 mph. Wind shear is about 15 knots over the disturbance, which is drifting south into a region where wind shear is expected to remain low enough to allow some development this week. I do think 95L will become a tropical depression, and most of the computer models also agree on this. The models disagree substantially on 95L's track, though. Steering currents will be weak in its vicinity, and 95L may spend a number of days wandering erratically. The Hurricane Hunters will investigate 95L Thursday afternoon.

Coast of Africa
The UKMET model is indicating the possible development of a tropical depression by Friday off the coast of Africa. There is a large surge of moisture with at least one strong tropical wave embedded in it coming off the coast of Africa this week, and it would be no surprise to see this wave develop into a tropical depression.

Katrina, two years later
Two years ago today, on August 29, 2005, Hurricane Katrina smashed into the Gulf coast with Category 3 winds and an incredible storm surge up to 27.8 feet high. Wunderblogger Mike Theiss was at ground zero in Gulfport, Mississippi during Katrina, and has posted a blog this morning on his experience, complete with some very compelling photos. His video of the storm surge washing into the hotel he was at is the most amazing storm video I've ever seen.

Margie Kieper's Katrina's Storm Surge feature on our tropical page provides an extraordinarily detailed 16-part examination of each portion of the coast devastated by Katrina. Margie is scheduled to be a guest on the Talking Tropics Internet radio show Thursday night to talk about Katrina's storm surge. Check the listings to see if there are any last minute changes.


The photo above was taken from Part 8: Lakeshore to Waveland, MS of Katrina's storm surge. An excerpt from the text:
I found an astonishing photo, of the peak of the surge in Waveland, which didn't appear to be faked, but I'm pretty much of a skeptic. The photo had this caption, "Photo taken in Waveland, MS, just North of the Railroad Tracks during Katrina around 9 AM by Judith Bradford." Note that it is being taken from the second floor window of a home, and that the water is close to the roof line of the first floor. There is a man perched on what is left of a home across the street, wearing a tiny life jacket and clutching a neon green pool noodle. There are electric lines running down from a pole to a home from left to right. In the distance on the right is a home with water up to the roof line. It is likely after 9am, as the bulk of the surge came between 9 and 10 am (that is when most of the fatalities occurred along the Mississippi coast), and probably the eye is already overhead, as the water is relatively calm and there appears to be little wind or rain, even though the pine trees are bent from the recent force of the eyewall winds.

The information provided by the Bradfords regarding the surge was very specific. The power went out at around 6:30am at their Waveland home on the morning of the 29th. They were staying in the home for a couple of reasons; first, because the home had not received any water at all from Camille, and, secondly, because both work in the medical field and needed to be available after the storm. At almost exactly 8:30am, water started coming over the railroad track embankment, from the coast, and into their yard.

Their home is 18 inches off the ground, and the first floor has 8-foot ceilings. There is an 18-inch truss between the 1st and 2nd floors, and this is what saved their 2nd floor from being flooded. In a matter of only five to ten minutes the water came up six feet, and quickly filled the first floor after that. Judith said that is why they saved so little from the first floor; they had no time to get anything. She first tried to shut the living room front door, but the force of the water burst the door open. She grabbed a camera and the Bradfords and their children ran upstairs. They marked the high water mark (HWM) on the inner stairwell showing how high the water came ? a little more than six more inches into the truss, which is a total of 10 feet of surge.

They saved two other people besides the man who was floating by on the roof in the photo. He was a chef named Glen, holding a four month old dachshund named Pinky, in the surge. He had lost his other dog and three cockatiels when his mother's home collapsed. The roof wedged against their van, underwater, and stopped, so they were able to save him. Bill Bradford told me when he swam out to rescue that man, that the water was so warm it seemed almost hot. He said the current was nothing like white water, but was a gentle continuous flow.

Because their home is right by the railroad tracks, it is not as high in elevation as I had thought. It is around 17 feet elevation. That is close to the HWM observed in Pass Christian, 27 feet.

With such a good quality HWM, I wondered why their house was not surveyed. Judith Bradford told me that no one from the federal government seemed to realize their house was there. The road leading up to Jeff Davis (they own 6 acres and raise miniature horses, which were drowned in their stables when the surge came) was filled with debris. The teams doing Search and Recovery for bodies didn't even check the house because they didn't know it was there; it was a good thing the family survived!

The water started to go down sometime after 11am, and by noon was about chest high, and by 2pm about waist-deep. The water finally left the house completely by about 4 or 5 pm that evening. She believes the railroad track embankment kept the water from receding faster. "


I'll have an update Thursday morning, unless there's a major change in 94L or 95L. My thoughts and prayers are with all those affected by Hurricane Katrina today. Let us not forget what happened two years ago.

Jeff Masters

Updated: 02:37 PM GMT am 29. August 2007

Permalink

Disturbance 94L in Central Atlantic; NHC management changes

By: JeffMasters, 01:58 PM GMT am 28. August 2007

A tropical wave midway between Africa and the Lesser Antilles (labeled "invest 94L" by NHC this morning) has become a little better organized this morning, and does have the potential to develop into a tropical depression later this week. Wind shear is a favorable 5-10 knots over 94L, and shear should not be a problem for it over the next few days. The system is not well-organized, with a sloppy, elongated circulation, as seen on last night's QuikSCAT pass. This morning's QuikSCAT pass missed the storm. The system does have a major impediment to development--the presence of a large, dusty area of dry air surrounding its north side (Figure 1). Water vapor satellite loops of the region don't show any significant moistening of the region around 94L occurring, and this will have to happen before the system can develop into a tropical depression. If the storm can develop a better-organized circulation, it will be able to pump more moisture into the surrounding atmosphere and help itself out. Current visible satellite loops shows that this is not happening at present--the thunderstorm activity associated with 94L is rather weak.

None of the reliable computer models develop 94L. The system should track through the Lesser Antilles Islands on Friday, bringing gusty winds and heavy rain. I expect that the earliest 94L could become a tropical depression would be Thursday, and it is unlikely the Lesser Antilles would experience anything worse than a 50 mph tropical storm. It is more likely that 94L will still be a tropical disturbance when it passes through the Lesser Antilles.


Figure 1. Water vapor satellite image of the Central Atlantic from 8:15am EDT 8/28/07. The brown colors denote very dry and dusty air from the Saharan Desert.

Honduras disturbance
A westward-moving tropical wave is bringing heavy rains to Honduras and Nicaragua today. Due to its close proximity to land, development of this wave into a tropical depression is not expected. The wave is under 10-15 knots of wind shear. None of the reliable models are predicting that this system will develop.

Computer model update
The UKMET and GFS models are indicating the possible development of a tropical depression by Thursday or Friday off the coast of Africa. There is a large surge of moisture with at least one strong tropical wave embedded in it coming off the coast of Africa this week. This moister air should make a more favorable environment for a tropical depression to form in than the one 94L finds itself in.

Most of the models also predict a low pressure system will develop off the North Carolina coast along an old frontal boundary on Thursday. Such a development may be an ordinary extratropical low pressure system, but could make the transition to a tropical system if the shear drops low enough.

NHC management changes
Acting director Dr. Ed Rappaport of The National Hurricane Center (NHC) will continue on the job until the end of this hurricane season, NOAA management revealed yesterday. Rappaport replaced director Bill Proenza on July 9, following an extraordinary few months of tumult at NHC. Joining the NHC staff on September 4 will be a new interim deputy director--Bill Read, who currently serves as the meteorologist in charge of the National Weather Service's Houston-Galveston office. Deputy director was the post Rappaport held prior to July 9. Bill Read applied for the directorship of NHC when Max Mayfield retired, but was not awarded the job. His arrival at NHC during the peak of hurricane season will be a welcome addition, as he has considerable experience dealing with hurricane emergencies.

In an apparent effort to follow some of the management changes recommended by the independent review team that performed the snap inspection of NHC in early July (see Attachment 9 of the Senate testimony from July), a new manager of the ten hurricane forecasters has been appointed as well. Rick Knabb, who is one of the six senior hurricane specialists, will be the new manager, and will be Rappaport's backup for TV interviews during hurricane emergencies.

It is still undecided where former director Proenza will wind up, but the chairs of two Senate subcommittees investigating the matter have recommended that Proenza be returned to his former job as head the NWS Southern Region.

Next update
Tomorrow is the 2-year anniversary of Hurricane Katrina. I'll discuss my experience with blogging about the storm, and give an update on 94L and the rest of the tropics.

Jeff Masters

Updated: 01:59 PM GMT am 28. August 2007

Permalink

The cold water wake of Dean

By: JeffMasters, 02:39 PM GMT am 27. August 2007

A westward-moving tropical wave is kicking up some disorganized thunderstorm activity in the southern Caribbean, between Colombia and Jamaica. This area has a circulation at middle levels of the atmosphere, but no circulation at the surface, as seen in the 7:21am EDT QuikSCAT pass. The wave is under 10-20 knots of wind shear, and could undergo some slow development. None of the reliable models are predicting that this system will develop into a tropical depression, but it could bring heavy rains to Nicaragua and Honduras on Tuesday and Wednesday.

Of the four reliable forecast models for forecasting the genesis of tropical cyclones, only one--the UKMET--is now calling for formation of a tropical depression late this week off the coast of Africa. There are some active tropical waves that will start pushing off the African coast later this week, and we'll have to watch this region for development.

Heavy thunderstorm activity off the Virginia coast is associated with an old frontal boundary. A non-tropical area of low pressure is forming here and is expected to move off to the northeast. This low could become tropical if the shear relaxes from its current 20-30 knots.

First snow of the season
Fall is fast approaching, as evidenced by the first snow of the season in the Colorado Rockies. Snow fell Friday on mountain peaks above 12,000 feet. Late August snow is not unusual for the Rockies. Ski season is still 70 days away, though!


Figure 1. Sea Surface Temperature (SST) departure from normal for Sunday, August 26, 2007. Image credit: U.S. Navy.

Hurricane Dean's wake
Hurricane Dean's passage will be remembered not only by the people it affected, but by the ocean itself. A large, powerful storm like Dean generates a tremendous amount of mixing of the ocean, which brings up deep, cold water to the surface. Dean's passage cooled off the Gulf of Mexico waters near the Yucatan Peninsula by up to 3 degrees C (5.6 degrees F). The western Caribbean was not as strongly affected, since there is a much deeper layer of warm water there, thanks to the presence of the warm Loop Current. The cold water anomaly left by Dean will take several weeks to dissipate.

I'll have an update on Tuesday.

Jeff Masters

Updated: 05:21 PM GMT am 27. August 2007

Permalink

Quiet in the tropics

By: JeffMasters, 03:18 PM GMT am 26. August 2007

Disorganized thunderstorm activity continues over the southwestern Gulf of Mexico. This activity is probably too close to land to develop, as it is expected to move westwards over Mexico over the next day. A westward-moving tropical wave is kicking up some disorganized thunderstorm activity in the southern Caribbean near the coast of Columbia. This area is under 10-20 knots of wind shear. The NOGAPS model predicts that this disturbance could develop into a tropical depression on Tuesday as it approaches the Nicaragua/Honduras coast. However, the NOGAPS model has been overly aggressive developing tropical cyclones in this portion of the Caribbean this year, and I don't expect this system will develop.

The four reliable forecast models for forecasting the genesis of tropical cyclones are all indicating possibility of a tropical depression forming off the North Carolina coast or off the coast of Africa late this week. Anything that does develop off the Carolina coast is likely to move northeastwards, out to sea. The greater threat to land would be a development off the coast of Africa. While it is currently quiet in the ITCZ region between Africa and the Lesser Antilles, I expect this activity to pick up late this week as some strong tropical waves move off the coast of Africa. A new tropical depression in this region 5-8 days from now has about a 40% chance of happening.

Wunderground for dial-up users
I did not post a blog yesterday as promised, since I found myself on dial-up as the result of a particularly nasty line of tornadic thunderstorms that swept through Southeast Michigan on Friday night. I posted a few photos of the damage from an EF-2 tornado that hit Fenton, about 8 miles north of my house. The storms knocked out power and high-speed Internet connectivity to tens of thousands of customers, and I am still without my beloved cable modem two days later. If you're in a similar predicament, or simply aren't interested in seeing all the comments for my blog, you can visit the "wunderground lite" version of my blog at http://www.wund.com/blog/JeffMasters/show.html. There is only one ad on the page, and the content is highly stripped down for modem users.

Hurricane Dean damage update
Hurricane Dean hit the Costa Maya Cruise Ship Terminal with full force, and the photos show damage consistent with at least a Category 4 hurricane. Huge chunks of the concrete dock are missing, and it is estimated that this second-busiest cruise ship destination in Mexico will be out of commission for at least six to eight months.

The death toll from Dean now stands at 37, with at least ten deaths now reported from Dean's second Mexican landfall, as a Category 2 hurricane. Remarkably, no deaths have been reported from Dean's landfall on the Yucatan Peninsula as a Category 5 hurricane.

I'll have an update on Monday.

Jeff Masters

Updated: 03:22 PM GMT am 26. August 2007

Permalink

One third of Arctic ice cap now missing; Midwestern floods; tropical update

By: JeffMasters, 12:46 PM GMT am 24. August 2007

Sea ice in the Arctic continues its record decline, thanks to unusually cloud-free conditions and above-average temperatures. For August 21, the National Snow and Ice Data Center estimated that fully one third of the Arctic ice cap was missing, compared to the average levels observed on that date from 1979-2000. Sea ice extent was 4.92 million square kilometers on August 21, and the 1979-2000 average for the date was about 7.3 million square kilometers. Arctic sea ice has fallen below the record low absolute minimum of 4.92 million square kilometers set in 2005 by about 8%, with another 3-5 weeks of the melting season still remaining. Reliable records of sea ice coverage go back to 1979.


Figure 1. Extent of the polar sea ice on August 21, compared to the average for the date from the 1979-2000 period (pink line). Image credit: National Snow and Ice Data Center.

With one third of the Arctic ice cap already gone, and another month of melting to go, we need to consider what effect this will have on weather, climate, and sea level rise. Well, we don't need to worry about sea level rise, since the polar sea ice is already in the ocean, and won't appreciably change sea level when it melts. However, the remarkable melting of the ice cap will likely lead to unusual weather patterns this fall and winter. The lack of sea ice will put much more heat and moisture into the polar atmosphere, affecting the path of the jet stream and the resultant storm tracks. Expect a much-delayed arrival of winter to the Northern Hemisphere again this year, which may lead to further accelerated melting of the ice cap in future years.

Last week, I remarked that the most recent images from the North Pole webcam show plenty of melt water and rainy conditions near the Pole. It turns out that was misleading, since the webcam is on a ship that was headed towards the pole, but had not reached it. There have been rainy conditions at the Pole this summer, and there is some open water there, but this is not uncommon in summer. Shifting ice frequently opens up leads (cracks) with open sea water at the Pole. It was one of these open leads that British swimmer Lewis Gordon Pugh swam in for 18 minutes this July to draw attention to global climate change.


Figure 2. Total rainfall from August 10-22 as estimated by NASA's TRMM satellite.

Midwest flooding
To get an idea of the magnitude of the flooding that has hit the Midwestern U.S. during the past ten days, take a look at the total amount of rain from August 10-22 (Figure 2). We can blame Tropical Storm Erin for the rain in Texas and Oklahoma (up to 11 inches), and for the nine flooding deaths that occurred in those states. However, the unbelievable rain amounts in excess of 20 inches in Minnesota and Wisconsin were primarily due to a frontal system--with the help of some copious moisture pumped northwards by the counter-clockwise circulation around Erin while it spun over Oklahoma.

Tropical update
There are no threat areas in the Atlantic to discuss. Two of our four reliable forecast models, the NOGAPS and ECMWF, are predicting that a tropical depression could form off the coast of Nicaragua on Sunday. The models forecast that this system would move inland over Nicaragua and Honduras by Monday.

I'll have an update on Saturday morning.
Jeff Masters

Climate Change Flood

Updated: 08:01 PM GMT am 16. August 2011

Permalink

The aftermath of Dean

By: JeffMasters, 03:20 PM GMT am 23. August 2007

Dean's rampage across the Caribbean is history. Dean made its final landfall yesterday as a 100-mph Category 2 hurricane near the tourism and fishing town of Tecolutla, Mexico. A slogan one could have used throughout Dean's tour of the Caribbean is, "it could have been much worse". The storm hit halfway between the most populous cities in the region--Tampico, population 300,000, and Veracruz, population 444,000. The region Dean hit is known as Mexico's Emerald coast, and is dotted by villages, cattle ranches, and uncrowded beaches. The storm weakened rapidly as it moved inland, and passed about 75 miles north of Mexico city, dropping heavy rains along its path. The remains of Dean are expected to make it to the Pacific ocean this weekend, then get pulled northwards in to Arizona, potentially bringing extra rainfall there, but not flooding. Wunderblogger Randy Bynon has a blog with some great photos of his flight into Dean with the Hurricane Hunters yesterday.

Insured damage from Dean to Mexico's Yucatan Peninsula was estimated at $400 million by one insurance company. Using the the usual rule of thumb that total damage is double the insured damage, the Yucatan suffered $800 million in damage. The total bill to Mexico from Dean will likely exceed $1 billion, when the damages from the storm's second landfall is factored in. Dean fortunately did little damage to the oil rigs in the Gulf of Mexico that provide about one third of Mexico's cash.

Once again, Mexico came out of a major hurricane will no deaths reported. I am very impressed with how well Mexico's civil defense system has performed during the past three hurricanes. Mexico also had no deaths from Hurricane Emily, which hit the country twice--once as a Category 4 storm near Cozumel, then as a Category 3 hurricane south of the Texas border. Hurricane Wilma, which clobbered Cancun and the northern tip of the Yucatan for three days as a Category 4 hurricane, killed only four people. The low loss of life from these three major hurricanes is something Mexicans can be truly proud of.


Figure 1. Flooding on the island of Dominica from Hurricane Dean. Image credit: Mike Theiss.

How some of the other countries on Dean's list fared:

Belize
About 5% of the buildings in northern Belize were damaged, and there was some destruction to the papaya crop. Electricity is nearly restored, and water was never lost.

Jamaica
It could have been very, very much worse on Jamaica. Dean missed the island, bringing Category 1 and 2 hurricane conditions to just the southern portion of Jamaica. According to articles in the Jamaica Observer and Jamaica Gleaner, Jamaica is making progress in the wake of the estimated $1.5 - $3 billion in damage left by Dean--the second most expensive hurricane in Jamaican history, next to the $4 billion in damage wrought by Hurricane Gilbert. Hurricane Dean cut water to 80% of the island, but by Wednesday, 48 hours after storm, water had been restored to 45% of the island. Half of the 248 roads blocked by the storm had been cleared, by Wednesday, and another 89 roads wad been opened for one lane traffic. Cruise ships had returned to the mostly undamaged northern part of the island. All of the hotels on the island are open except one. Most of the island is still without power, but 50% of Jamaica should have power by the weekend. There is significant damage on the South Coast 69 Kv transmission line and severe damage to the power transmission infrastructure in the east and south, and it may be many weeks before power is restored to the entire island. Cell phone communication is available on 70% of the island.

Only three deaths were reported on Jamaica, which is far fewer than the 17 deaths suffered during Hurricane Ivan and the 45 deaths from Hurricane Gilbert. Better building codes and better hurricane awareness and planning are to credit for this low death toll. Jamaica has done a great job preparing for and recovering from this storm.

There were media reports of a 114 mph sustained wind measured in Kingston during Dean, which I though sounded unreasonably high. The Kingston airport measured top sustained winds of 81 mph. I asked Jeff Meeks, who weathered the storm in Kingston, about this. He said his suburb of Kingston--Barbican--had sustained winds of 30 mph, gusting to 63 mph, with 5.24 inches of rain. He further commented,

There was no possibility of 114mph sustained winds in any part of Kingston. The damage is just not there. Further I have another friend who also had a high gust of just 65mph. He is also located on the outskirts of Kingston. Norman Manley International is at the southern most extent of Kingston and would have been closest to the nearest approach of Dean and may have had higher winds but the damage there is also minimal. In truth Ivan from 2004 gave Kingston much more damage.

Haiti
Haiti suffered the highest death toll from Dean, eleven. Several hundred houses were damaged or destroyed on the south coast, and there was some moderate damage to agriculture. It could have been much worse. Haiti was lucky Dean moved by so quickly, and was not able to dump devastating amounts of rain on the country.

Lesser Antilles
The tourist industry in the Lesser Antilles was not significantly affected by Dean. All of the hotels on the affected islands are now open, and little damage occurred to the hotels. However, agriculture suffered tremendously. The banana crop was wiped out on St. Lucia, Martinique, and Dominica, and was 80% destroyed on Guadaloupe. The hardest hit island, Martinique, is estimating storm costs of $270 million. Phone service was knocked out to 50% of the island, and was still out to 35% of the island on Wednesday. St.Lucia is reporting $18 million in total damage, and Dominica is reporting $98 million in damage to infrastructure (agricultural damage may be another $100 million).

Wunderground Ultimate Chase blogger Mike Theiss was in Dominica for Hurricane Dean, and he's written a detailed account of what is was like to go through he hurricane, complete with some great photos.

After Dean, what next?
An area of disturbed area has developed just west of Jamaica in the western Caribbean, associated with a tropical wave moving west-northwest at 15-20 mph. The wave is under 20 knots of wind shear, and I don't expect any development to occur. None of the reliable hurricane forecast models are calling for anything to develop in the next seven days.

I'll have a new blog Friday morning.

Jeff Masters

Updated: 07:31 PM GMT am 23. August 2007

Permalink

Dean's final landfall

By: JeffMasters, 02:35 PM GMT am 22. August 2007

Hurricane Dean is hours away from its final landfall--a strike as a Category 1 or 2 hurricane along Mexico's Gulf Coast, near Tuxpan. Dean lost its eyewall and became rather disorganized from its passage over the Yucatan Peninsula, and is only now beginning to increase its winds. With only a few hours left before landfall, Mexico is fortunate that Dean is moving so quickly--20 mph--and does not have time to organize further.


Figure 1. Dean over the Yucatan Peninsula at 3:05pm EDT 8/21/07. Image credit: NASA.

The Mexican coast near the landfall point is the most densely populated area Dean will affect, and damage there will probably exceed $1 billion. Risk Management Solutions has estimated the insured damage to the Yucatan was between $750 million and $1.5 billion. Total damage is typically double the insured damage, so the price tag for Dean will be very steep for Mexico. Oakland Calif.-based EQECAT, Inc. issued a preliminary assessment of the insured losses from Hurricane Dean of between $1.5 and $3 billion. The estimate covers damages incurred by the hurricane's passage through the Lesser Antilles as far as Jamaica.

Dean's core passed just north of Chetumal, Mexico during its landfall as a Category 5 hurricane yesterday, and that city of 150,000 suffered mostly Category 1 and 2 hurricane damage. Power has already been restored, and the governor of the province estimated that the city would be back to normal in two weeks--except for the 3-6 months needed to repair some of the roads washed out. Farther north, about a third of the hotels and cabins in Tulum, and strip of coastal development just south of Cozumel, received damage. Beach erosion was significant all along the Yucatan. The worst damage was reported in Majahual, on the coast 30 miles northeast of Chetumal, where the full Category 5 strength of the storm was felt. According to the Associated Press, "Hundreds of homes were collapsed in Majahual when Dean's eye passed almost directly overhead, crumpling steel girders, splintering wooden structures and washing away about half of the immense concrete dock that transformed the sleepy fishing village into Mexico's second-busiest cruise ship destination on the peninsula. The storm surge covered almost the entire town in waist-deep sea water." For those interested in tracking the effects of the storm, the Hurricane Dean wikipedia page is an excellent source of information.

Once inland, Dean poses a threat to Mexico City due to flooding rains. However, Dean is moving quickly enough that I don't forsee a major flooding disaster, such as occurred in Mexico after Hurricane Gilbert made its final landfall in 1988. Gilbert stalled over the Mexican mountains after landfall, dumping huge quantities of rain that triggered flash floods, killing over 200 Mexicans.

Links to follow today:
Tuxpan, Mexico observations
Radar from Alvarado, Mexico.
Morphed microwave animation.


Figure 1. Flooding on Dominica captured by storm chaser Mike Theiss

Chasing Dean in the Lesser Antilles
Wunderground Ultimate Chase blogger Mike Theiss has just returned from a trip to the Lesser Antilles, where he encountered the fury of Hurricane Dean as it blasted through the islands as a strengthening Category 2 hurricane. He's written a detailed chase account on his experience, complete with some amazing photos.

Elsewhere in the tropics
An area of disturbed weather associated with a tropical wave, 92L, is moving through the Bahama Islands. The upper-level winds over it are not hostile, but this system shows little organization and is not a threat to develop before moving inland over Florida on Thursday. Elsewhere, the tropics are quiet, and none of the reliable computer models show anything developing in the next week.

I'll have an update Thursday morning.

Jeff Masters

Updated: 07:33 PM GMT am 23. August 2007

Permalink

Hurricane Dean--9th strongest hurricane on record

By: JeffMasters, 02:46 PM GMT am 21. August 2007

Hurricane Dean powered ashore in Mexico's Yucatan Peninsula this morning as a Category 5 hurricane with 165 mph winds. The pressure of 906 mb measured by the Hurricane Hunters shortly before landfall at 4:30am EDT makes Dean the ninth strongest hurricane on record in the Atlantic. This is the third lowest pressure at landfall behind the 1935 Labor Day hurricane in the Florida Keys and Hurricane Gilbert of 1988 in Cancun Mexico. Dean is also the first Category 5 hurricane to make landfall in the Atlantic Basin since Andrew of 1992.


Figure 1. Dean at landfall, as seen by the Cancun radar. Image credit: Meteorological Service of Mexico.

Radar images at landfall (Figure 1) show that Dean came ashore just north of Chetumal, Mexico, a city of 130,000 people. Dean's center passed about 15 miles north of the city, and Chetumal missed the strongest Category 5 winds of the storm. The strongest winds from Dean were in the right front quadrant on the northern side, since the forward speed of the storm adds to the rotational speed of the winds there. It appears Chetumal was just at the edge of the southern eyewall, and probably experienced sustained winds of Category 3 strength, 115 mph. We don't know, since the weather station stopped reporting data long before the storm arrived. However, a wind analysis done by NOAA's Hurricane Research Division (Figure 2) supports this estimate. The coastal area north of Chetumal where Dean's greatest fury was unleashed has a cruise ship port and a stretch of beach front development, and this region probably suffered near-total destruction.


Figure 2. Dean's winds one hour before landfall. Winds are in knots, multiply by 1.15 to convert to mph. Locations of Chetumal and the Costa Maya Cruise Ship Port are marked. Winds of Category 1 strength (65 kt) are colored yellow, and winds of minimal Category 3 strength (100 knots) are colored pink. Image credit: NOAA/Hurricane Research Division.

Further north, it appears that Cozumel probably got sustained winds near tropical storm force, 39 mph. The weather station there stopped transmitting data before the storm arrived. Cancun's winds topped out at 29 mph, gusting to 54 mph. To the south, Belize City has had top winds of 23 mph, gusting to 35 mph, so far this morning. On the western side of the Yucatan Peninsula, the winds are starting to rise at Campeche. Dean's center will pass south of Campeche, and bring tropical storm or Category 1 hurricane-force winds to the city.

Dean is powerful enough to survive the crossing of the Yucatan as a hurricane, and I expect it will be a Category 2 hurricane with 100 mph winds when it pops out into the Gulf of Mexico later today. Hurricane Janet of 1955, which hit near Chetumal as a Category 5 storm with 170 mph winds, weakened to a Category 2 storm with 105 mph winds when it popped out into the Gulf of Mexico south of Campeche. Janet was moving at about the same speed Dean is, so I expect Dean will behave similarly. Once out over the warm waters of the Gulf of Mexico, Dean has time to intensify by perhaps 15 mph before it makes a second landfall near Poza Rica. Dean will finally dissipate in the mountains about 100 miles north of Mexico City, and could bring heavy rains to the Mexican capital. No hurricane has ever survived the crossing from the Gulf of Mexico to the Pacific over the wide, mountainous portion of Mexico.

Links to follow today:
Campeche, Mexico observations.
Radar from Cancun, Mexico.
Belize City observations.
Morphed microwave animation.

Disturbance 92L
An area of disturbed weather associated with a tropical wave, 92L, is a few hundred miles north of Puerto Rico. This disturbance is less organized than yesterday, despite some rather favorable upper-level winds. There is some dry air to the north that may be interfering with organization, and there is probably not enough turning motion available from this tropical wave to get 92L spinning. I don't expect significant development today given its current state of disorganization, but 92L deserves close scrutiny over the next few days. None of the reliable computer models develop the system.

My live appearance tonight on Internet Partnership Radio
I'll be the guest tonight on the Internet Partnership Radio (http://www.ipr365.com). Tonight's show is called "Center of Circulation", and consists of global severe, winter, and tropical weather news/topics with up to the minute advisories, watches and warnings, safety & preparedness info, and periodic special guests. The host is Charlie Wilson. I hope you can listen in!

I may do a short update this afternoon, and the next full update will be Wednesday morning around 10am EDT.

Jeff Masters

Updated: 02:33 PM GMT am 22. August 2007

Permalink

Dean hits Category 5

By: JeffMasters, 02:10 AM GMT am 21. August 2007

Hurricane Dean has intensified into the first Category 5 storm in the Atlantic since Hurricane Wilma of 2005. The latest Hurricane Hunter fix at 8:34pm EDT found 185 mph winds at their flight level of 10,000 feet, which corresponds to surface winds of 160 mph. The pressure had dropped to 914 mb, and I expect Dean will strengthen right up until landfall. Landfall is expected near Chetumal, Mexico, just after midnight local time. Dean will be a tremendously destructive storm for southern Mexico. Dean is powerful enough to be able to survive the crossing of the Yucatan as a Category 2 hurricane, and hurricane advisories have been posted for cities on the western coast of the Yucatan Peninsula. The only hurricanes on record that survived crossing this portion of the Yucatan and maintaining hurricane intensity were Hurricane Roxanne of 1995, which hit just south of Cozumel as a Category 3, and emerged near Campeche as a Category 1; and Hurricane Janet of 1955, which hit near Chetumal as a Category 5 storm, then weakened to a Category 2 storm when it popped out into the Gulf of Mexico south of Campeche. We can expect Dean will carve out a path of great destruction all the way across the Yucatan Peninsula, then potentially re-intensify before hitting Mexico again along the Gulf Coast in the Bay of Campeche.


Figure 1. Current satellite image of Dean.

Links to follow today:
Radar from Cancun, Mexico.
Chetumal, Mexico observations.
Cozumel, Mexico observations.
Belize City observations.
Campeche, Mexico observations.
Morphed microwave animation.

I'll have a full update Tuesday morning at about 10am EDT.

Jeff Masters

Permalink

Dean intensifying, near Category 5 strength

By: JeffMasters, 08:33 PM GMT am 20. August 2007

Hurricane Dean is intensifying. The latest Hurricane Hunter data and satellite intensity estimates both show an intensifying storm, and the 3:54pm EDT eye report showed a 6 mb pressure drop in less than two hours, which is a big fall. The pressure now stands at 918 mb, which is the lowest pressure Dean has attained thus far. The storm is over waters with very high heat content, and is under light wind shear, so continued intensification is probable. Landfall is expected near Chetumal, Mexico, just after midnight local time. Dean will be a tremendously destructive storm for southern Mexico. Dean is powerful enough to be able to survive the crossing of the Yucatan as a hurricane, and hurricane advisories have been posted for cities on the western coast of the Yucatan Peninsula. The only hurricanes on record that survived crossing this portion of the Yucatan and maintaining hurricane intensity were Hurricane Roxanne of 1995, which hit just south of Cozumel as a Category 3, and emerged near Campeche as a Category 1; and Hurricane Janet of 1955, which hit near Chetumal as a Category 5 storm, then weakened to a Category 2 storm when it popped out into the Gulf of Mexico south of Campeche. We can expect Dean will carve out a path of great destruction all the way across the Yucatan Peninsula, then potentially re-intensify before hitting Mexico again along the Gulf Coast in the Bay of Campeche.

Jamaica and Haiti
News is still slow to emerge from Jamaica due to blocked roads and lack of electrical power. It does appear that the north coast, including Montego Bay, did not suffer extensive damage. News reports now indicate four people died on Haiti, and 150 homes were destroyed.


Figure 1. Current satellite image of Dean.

Links to follow today:
GOES rapid scan satellite loop
Radar from Cancun, Mexico.
Chetumal, Mexico observations.
Cozumel, Mexico observations.
Belize City observations.
Campeche, Mexico observations.
Morphed microwave animation.

Disturbance 92L needs to be watched
An area of disturbed weather associated with a tropical wave a few hundred miles northeast of the Lesser Antilles Islands, "Invest 92L", has changed little in organization today. Wind shear is about 5-10 knots in this region, and an upper-level anticyclone has formed over 92L. This is a very favorable environment for intensification, should 92L start to get organized. The disturbance has the potential to develop into a tropical depression by Wednesday. It is moving west to west-northwest at 15-20 mph, and will be near the central Bahamas by Wednesday, and the east coast of Florida by Friday. It does not appear that any troughs strong enough to recurve 92L will swing by until Saturday at the earliest.

Tropical wave approaching the Lesser Antilles
A tropical wave approaching the southern and central Lesser Antilles islands will bring showers and gusty winds to the islands Tuesday. Wind shear has dropped to 5-10 knots over the wave, and there is some potential for it to develop as it moves west to west-northwest into the Caribbean Sea.

I'll have a full update Tuesday morning at about 10am EDT.

Jeff Masters

Updated: 09:12 PM GMT am 20. August 2007

Permalink

Dean roars towards the Yucatan; Erin's remains kill 6 in Oklahoma

By: JeffMasters, 02:03 PM GMT am 20. August 2007

Hurricane Dean is headed west towards the Yucatan, still packing Category 4 winds, with the potential to add even more punch. Dean is over waters with some of the highest heat content anywhere in the Atlantic, and could well take advantage of this energy source and become a Category 5 hurricane today. Landfall will occur late tonight or early Tuesday morning over the Yucatan Peninsula. Landfall should occur over 100 miles south of Cozumel, so the Mexican Riviera areas of Cancun and Cozumel will only experience tropical storm force winds. The region where Dean is expected to hit, just north of the border with Belize, has one large city, Chetumal. Chetumal is on a narrow bay, but the expected storm surge of up to 11 feet will have difficultly making it up the bay to flood the city. The surrounding region is sparsely populated near the coast. Damage from Dean in Mexico may be considerably higher for the storm's second landfall, if it hits the major city of Tampico in the western Gulf of Mexico.


Figure 1. Hurricane Dean close-up taken from the Space Shuttle Endeavour, Saturday August 18, 2007 at 1pm EDT. Image credit: NASA.

Jamaica
There haven't been many news reports out of Jamaica yet, but we know that electricity is still out, many roads are blocked, and numerous mudslides have been reported across the island. Wind analyses prepared by NOAA's Hurricane Research Division suggest that Category 1 and 2 hurricane conditions affected most of Jamaica, but the Category 3 and 4 winds from the eyewall stayed offshore. I expect the damage tally will reach at least $1 billion on the island, but fall well short of the $4 billion done by Hurricane Gilbert of 1988. Gilbert cut straight across Jamaica as a Category 3 hurricane with 125-130 mph winds. Kingston measured sustained winds of 116 mph during Gilbert; the top winds reported so far from Dean were in the 80-100 mph range.

Dean's impact elsewhere
Eight deaths are being blamed so far on Dean--five in the Lesser Antilles, one in the Dominican Republic, and two on Haiti. Dean missed the Cayman Islands by over 100 miles, and the top winds on Grand Cayman never exceeded tropical storm force (39 mph).


Figure 2. Radar image of Dean at its closest approach to Jamaica. Dean had two eyewalls at the time, concentric with each other. The outer eyewall just grazed the sparsely populated southern tip of Jamaica. Image credit: Cuban Meteorological Institute.

Links to follow today:
Radar from Cancun, Mexico. (Overloaded today, good luck getting an image!)
Morphed microwave animation.
Chetumal, Mexico observations.
Cozumel, Mexico observations.
Belize City observations.

New disturbance 92L to watch
There is an area of disturbed weather associated with a tropical wave a few hundred miles northeast of the Lesser Antilles Islands. NHC has labeled this system "Invest 92L" this morning. Wind shear is about 10 knots in this region, which is low enough to allow some development over the next few days. This area is moving west-northwest, and will be near the U.S. East coast late this week. The next trough of low pressure strong enough to recurve this system is not due until Saturday, so this system will definitely be a threat to the U.S. if it develops. A QuikSCAT pass from 5:25am this morning shows no signs of a surface circulation, but plenty of strong straight-line winds from the thunderstorms.

Erin finally dissipates
Tropical Storm Erin finally died this morning over Missouri. Erin dramatically re-intensified Saturday night over Oklahoma, forming a tropical storm like-vortex that brought up to 11 inches of rain to Oklahoma, and helped feed disastrous rains of up to a foot over southeastern Minnesota. At least 13 deaths are being blamed on the resulting flooding, six of them in Oklahoma. The radar presentation of Erin's remains (Figure 3) looks remarkable tropical storm-like, and such re-intensifying tropical cyclones over land, complete with calm eye and spiral bands, have been observed in Australia, where they have been dubbed "landphoons". Hurricane David of 1979 performed a similar feat, generating severe weather over Washington D.C. 27 hours after it had made landfall in Georgia. What seems to be happening in these cases is that the circulation at upper levels of the atmosphere can remain intact if there is not a lot of wind shear to tear it apart. This circulation can then reach down to the surface again during its passage far inland. I've saved a long animation of this "landphoon", and Dr. Kevin Kloesel of the University of Oklahoma provided me a plot showing that the winds in Fort Cobb, Oklahoma were sustained at tropical storm force for over 10 minutes Sunday morning, with a peak wind gust of 75 mph. Winds at Watonga, OK, blew a sustained 55 mph, with gusts to 73 mph!


Figure 3. The remains of Tropical Storm Erin re-intensified into a remarkably tropical storm-like cyclone Sunday morning.

I'll have a short update this afternoon at about 4pm EDT, and full update Tuesday morning at about 10am EDT.

Jeff Masters

Updated: 05:30 PM GMT am 20. August 2007

Permalink

A very bad day in Jamaica

By: JeffMasters, 11:21 PM GMT am 19. August 2007

It could have been much worse, but it is very bad for Jamaica. Hurricane Dean's northern eyewall is just offshore the southern tip of Jamaica, bringing sustained Category 2 hurricane winds to southern Jamaica. A recent wind analysis prepared by NOAA's Hurricane Research Division (Figure 1) at 3:30pm EDT today shows winds of Category 1 strength (>65 knots, or 74 mph) already affecting the east end of the island. By extrapolating this wind field over the island to the west-northwest, in anticipation of Dean's track, it is apparent that perhaps 90% of the island will experience sustained winds of 74 mph or greater. At 4pm EDT, Kingston, on the southern side of the island, recorded sustained winds of 81 mph before the instrument failed. We can expect that the southern 1/3 of the island, including Kingston, will receive sustained winds of Category 2 strength--96 to 114 mph. Category 3 and higher winds will be confined to the southernmost 5% of the island, and it appears that the Category 4 winds will stay offshore. The portion of the island affected by the Category 3 winds is very sparsely populated.

Jamaica will probably suffer a billion dollars in damage from Dean, perhaps more. The high winds and rains of up to 20 inches will no doubt claim lives, though probably not nearly as many as the 45 who died during Hurricane Gilbert of 1988. Gilbert cut straight across Jamaica as a Category 3 hurricane with 125-130 mph winds, doing $4 billion in damage. Kingston measured sustained winds of 116 mph during Gilbert; I expect the top winds in Dean will be 10 mph slower than that.


Figure 1. Wind analysis of Dean at 3:30pm EDT 8/19/07. Areas in yellow mark winds of Category 1 and stronger (65kt, 74 mph). Dean was moving west-northwest, and the most intense winds (120kt, 140 mph) should stay barely south of the island. Image credit: NOAA's Hurricane Research Division.

I'll have a full update Monday morning at about 10am EDT.

Jeff Masters

Updated: 12:56 AM GMT am 20. August 2007

Permalink

Dean charges towards Jamaica; Erin returns

By: JeffMasters, 03:37 PM GMT am 19. August 2007

Hurricane Dean continues to pound Haiti and the Dominican Republic with high winds and heavy rain, and is headed for a very close encounter with Jamaica and the Cayman Islands. Early this morning, winds at Barahona on the southernmost point of land of the Dominican Republic hit 52 mph, gusting to 104 mph. Sustained hurricane force winds are expected to remain well south of both Haiti and the Dominican Republic, but a major spiral band has brought extremely heavy rain to the south portion of both countries.


Figure 1. Microwave satellite image of Dean taken at 7:17am EDT Sunday August 19. Think of this as a weather radar in space--the red areas show where the most intense thunderstorms in the spiral bands and eyewall are occurring. Note the incomplete double ring of echoes around the dark blue eye. Dean has two eyewalls, concentric around each other.

Jamaica and the Cayman Islands
Jamaica is already receiving high winds and heavy rain from an outer spiral band. How bad will it get? The big question is if the eyewall will move over the island. Unfortunately for Jamaica, Dean has two eyewalls, forming concentric rings (Figure 1). The inner eyewall is 15 miles in diameter, and the outer eyewall is 37 miles in diameter. Winds of Category 3 and 4 strength are blowing in both eyewalls, as seen in the latest data from the SFMR surface winds taken by the Hurricane Hunters. So, Dean's center has to pass more than 25 miles south of Jamaica for the island to be spared the worst of the hurricane. The nation's capital, Kingston, lies on the southern portion of the island, and will be the hardest-hit major city. The tourist city of Montego Bay is on the northern part of Jamaica, and will fare much better.

The same story holds true for the Cayman Islands. Grand Cayman, the southernmost of the islands, it at greatest risk. If Dean passes more than 30 miles south of the island, they will miss seeing the outer eyewall of Dean and will fare relatively well. It's going to be a close call, but it appears that both Jamaica and the Cayman will miss seeing the eyewall of Dean.

Mexico and Texas
Mexico will not be so lucky, and will receive a double beating. Dean is expected to make landfall twice, once near the tourist havens of Cozumel, and then again south of the Texas border. Mexico has to hope that the steering currents will be kind and take Dean south of the most heavily populated regions of the Yucatan. Hurricane Emily of 2005 grazed the southern tip of Cozumel Island as a Category 4 hurricane with 135 mph winds, and Dean may follow a similar path. Mexicans can take heart in the fact that Emily caused no deaths in Mexico, and damage was surprisingly light. Most of the tourist regions were relatively unaffected by Emily--it was Wilma two months later that really punished the Mexican Riviera.

As for Texas, it looks right now like only extreme southern Texas near Brownsville needs to worry about Dean. Hurricane Emily hit 90 miles south of Brownsville as a Category 3 hurricane in 2005, and I expect a similar story will unfold for Dean. Emily brought sustained winds of about 40 mph to extreme south Texas, a 4-5 foot storm surge, eight tornadoes, and heavy rains. Damage was minor.

Links to follow:
Radar in Piln, Cuba.
Radar from Guantanamo Bay, Cuba.
Morphed microwave animation.
Kingston, Jamaica observations.
Montego Bay, Jamaica observations.
Grand Cayman observations.

After Dean, what next?
There is an area of disturbed weather that has formed off the northeast coast of South America, 400 miles southest of Barbados. Wind shear is 20-25 knots in this region, and will stay too high to allow develoment for at least the next two days. None of the reliable computer models are suggesting anything will develop over the coming week. The ITCZ region between Africa and the Lesser Antilles is relatively quiet.

Erin returns
The remains of Tropical Storm Erin re-intensified this morning into a major storm that slammed central Oklahoma with rains up to seven inches and wind gusts of tropical storm strength. The radar presentation of Erin's remains (Figure 2) looks remarkable tropical storm-like. I've saved a long animation of this "landcane". Numerous flood watches, flood warnings, and severe thunderstorm warnings have been posted for Oklahoma today.


Figure 2. The remains of Tropical Storm Erin re-intensified into a remarkably tropical storm-like cyclone today.

Typhoon Sepat
Typhoon Sepat has moved inland over mainland China, after hitting as a Category 1 storm. Earlier, Sepat hit Taiwan as a Category 3 typhoon. No deaths occurred on Taiwan, but at least 15 died in China--11 of them in a tornado spawned by the typhoon.

I'll have a full update Monday morning, and may have a short update later today.

Jeff Masters

Updated: 03:44 PM GMT am 19. August 2007

Permalink

Can Jamaica pray away Hurricane Dean?

By: JeffMasters, 01:15 AM GMT am 19. August 2007

Can Jamaica pray away Hurricane Dean? The official forecast and nearly all of the computer models have put Jamaica in the bulls-eye for several days now. But hurricanes have a funny way of taking 11th-hour wobbles that spare the island a direct hit. Witness the remarkable turn Hurricane Ivan took in 2004, as it headed directly for the island with 145 mph winds. Ivan took a sudden turn 35 miles from the island, traced out an exact outline of the island's coast 35 miles offshore, then resumed its previous track. In the Jamaica Observer, Custos of Kingston, Reverend Carmen Stewart, contends that it was not the first time that prayers had influenced the turn of events when disaster faced Jamaica. "It has happened time and time again," Reverend Stewart says. "I know people have been praying and I don't see any other reason why it (the hurricane) would make such a drastic turn.... God hears prayer."


Figure 1. Hurricane Ivan as it miraculously skirted the island of Jamaica. Image credit: jamaicancaves.org.

Category 5 Hurricane Allen took an odd wobble around the island, too, but Hurricane Gilbert of 1988 did not. Can Jamaicans pray away Hurricane Dean? Well, the recent motion of Dean has been more due west, which may bring the storm just south of the island. Keep praying, Jamaica!

Links to follow over the next day:
Radar in Piln, Cuba.
Radar from Guantanamo Bay, Cuba.
Punta Cana, Dominican Republic observations (east tip of the island).
Santo Domingo, Dominican Republic observations.
Kingston, Jamaica observations.
Montego Bay, Jamaica observations.

I'll have an update by noon Sunday.
Jeff Masters

Updated: 01:30 AM GMT am 19. August 2007

Permalink

Extremely dangerous Dean heads for Jamaica

By: JeffMasters, 03:37 PM GMT am 18. August 2007

Hurricane Dean put on an impressive round of rapid intensification last night, deepening 49 millibars in just 24 hours. Dean is now a major Category 4 hurricane with 150 mph winds. Reports from Hurricane Hunter aircraft show that Dean has likely peaked in intensity, and may be undergoing an eyewall replacement cycle. The eye has shrunk from 20 miles in diameter yesterday down to 13 miles in diameter this morning. This inner eyewall will probably shrink even more and collapse sometime in the next day, to be replaced by a new outer eyewall 30-40 miles in diameter. Dean's winds may decrease to the lower end of the Category 4 scale, 135-140 mph, if that occurs. The inner eyewall and the new outer eyewall that is forming can be seen on a microwave satellite image from this morning (Figure 1). The 11:02am EDT eye report from the Hurricane Hunters said that the southern portion of the inner eyewall was missing, so the eyewall is probably collapsing now.


Figure 1. Microwave satellite image of Hurricane Dean at 7:31am EDT Saturday August 18. Two small partial rings of strong echoes, marking the boundaries of concentric inner and outer eyewalls, are visible. Image credit: Navy Research Lab.

Dean pounded Martinique, St. Lucia, and Dominica yesterday, and the storm's death toll now stands at three. A 62-year old man died on St. Lucia while trying to save his cow from raging flood waters, and a rain-triggered landslide killed a mother and child in their home in Dominica. Martinique suffered the worst damage, with 100% of the banana crop destroyed, 70% of the sugar cane crop gone, and considerable damage to buildings on the south end of the island. Lesser damage occurred on Dominica and St.Lucia, and overall, it appears that the Lesser Antilles islands were fortunate to get off so lightly.

Puerto Rico
Dean's eye is visible on long range radar out of Puerto Rico. A major spiral band of rain moved over Puerto Rico at about 11am EDT, and will bring up to four inches of rain to the island today. Radar estimated rainfall from the Puerto Rico radar shows up to two inches had fallen as of 1pm EDT. The wind/pressure plot from Buoy 42059 south of Puerto Rico shows that Dean passed just north of that location this morning, bringing wind gusts to 66 knots.

Dominican Republic
The tourist town of Punta Cana on the east tip of the island reported sustained winds of 34 mph, gusting to 46 mph this morning. The capital city of Santo Domingo can expect sustained winds of 40-45 mph today as Dean makes its closest pass to the south. Heavy rains of 1-3 inches in non-mountainous area will create some minor flooding problems in these areas. Mountainous area along the south coast of the Dominican Republic, particularly in the rugged Barahona Peninsula that juts farthest south into the Caribbean, will receive higher rain amounts and are at great risk of life-threatening flash floods.

Haiti
Dean will make a very close passage to the south of Haiti's mountainous southern Peninsula, and the capital city of Port-au-Prince could experience winds just below hurricane force. A Category 4 or 5 hurricane passing so close to Haiti is a serious threat. Deforestation has denuded the mountainsides of protective tree cover, and flood waters will wash down the mountains into populated areas. The only saving grace in this situation may be the relatively rapid forward speed of Dean, which will reduce the amount of rain that will fall, compared to other hurricanes that have affected Haiti.

Jamaica and the Cayman Islands
Jamaica is my greatest concern. A direct hit by Dean would make it the worst hurricane strike on Jamaica for over a century. Jamaica has not received a direct hit by a Category 4 or 5 hurricane since perhaps 1832. The worst strikes of the 20th century were Category 3 Hurricane Gilbert of 1988 and Category 2 Hurricane Charlie of 1951. The Cayman Islands also have much to fear from Dean. Dean could rival Ivan as the Cayman's worst hurricane strike of the past century.

Cuba
Mountainous regions in Cuba area also at risk of dangerous flash flooding from Dean. However, civil defense is so good in Cuba that I don't expect any loss of life. Portions of south Cuba will experience sustained tropical force winds.

Mexico
The models have come into much better agreement this morning on the longer term forecast, and it looks very bad for Cancun and Cozumel. Dean will give those resort areas a pounding like they received from Hurricane Emily (Category 4) in 2005, and Hurricane Gilbert (Category 5) in 1988. Dean will probably not be as bad as Wilma (Category 4) in 2005, since Wilma stalled out over Cancun for three days as a major hurricane. Dean is moving quickly, and will not linger long over any of the regions it strikes.

Texas and Louisiana
Things are looking much brighter for Louisiana, as the GFDL model has come in line with all of the other models in predicting a landfall in Southern Texas or Northern Mexico. It now appears likely that Mexico's Yucatan Peninsula will knock Dean down a category or two before it can approach the Texas coast. The upper level low that was forecast by the GFDL to potentially steer Dean northwards appears to be weakening and moving westwards, out of the way of Dean. You can watch this upper level low on water vapor satellite loops. It is the counter-clockwise spinning region that has moved west off the Florida coast into the eastern Gulf of Mexico. If this low continues to weaken and move westwards, it will not be able to swing Dean northwestwards towards northeast Texas and Louisiana. If Dean does manage to catch up to the upper level low, the counter-clockwise circulation around the upper low will bring some south-to-north winds over Dean that would steer it on a more northerly track into the Gulf of Mexico.

The NOAA jet is scheduled to fly tonight and Sunday morning, so we should have some excellent model runs available Sunday morning and afternoon.

I'll have a short update tonight by 9pm, and a full update Sunday morning by noon. Tonight's update will focus on Jamaica, and I'll post the relevant radar and current condition links to follow the storm's path. I'll talk about Typhoon Sepat, as well.

Jeff Masters

Updated: 04:49 PM GMT am 18. August 2007

Permalink

Which model do you trust? And, Arctic sea ice reaches a record minimum

By: JeffMasters, 07:33 PM GMT am 17. August 2007

Hurricane Dean, now a major Category 3 hurricane with 125 mph winds, continues to intensify and grow larger in size. Dean pounded Martinique and St. Lucia this morning, and claimed its first victim when a 62-year old man died on St. Lucia while trying to save his cow from raging flood waters.

Dean's eye is now visible on long range radar out of Puerto Rico. Buoy 42059 is in Dean's path, and should be interesting to watch.

We're fairly confident of the 1-2 day forecast, which has Dean headed west to west-northwest over the Central Caribbean, very close to Jamaica and the Cayman Islands, then into the Western Caribbean. After that, things become murkier. The latest 12Z runs of the NOGAPS, UKMET, GFS, and HWRF computer models all show Dean hitting the Yucatan Peninsula, and continuing on into the Gulf of Mexico towards a second landfall near or south of the Texas border. The HWRF run is slower, and does not take Dean to the coast at the end of its forecast period. The big outlier is the GFDL model, which now takes Dean northwest into central Louisiana. Which model is correct? The problem is that each model has a different solution for the behavior of an upper-level low pressure system expected to be over the Gulf of Mexico early next week. Which model should we trust?

In 2006, the official NHC forecast performed better than any of the individual computer forecast models. However, several "consensus" forecasts made using an average of the "big four" computer models (GFDL, GFS, UKMET, and NOGAPS), slightly outperformed the official forecast at some time periods (Figure 1). The Florida State Super-Ensemble (FSSE), for example, combines the "big four" models on the basis of past performance in an attempt to correct for biases in those models. (The FSSE is owned by a private company, which makes it available to NHC but not the general public). The Florida State Super Ensemble slightly out-performed the official NHC forecast at most forecast times.

The "big four" models are plotted on wunderground.com's computer model page for Dean, (along with the inferior BAMM model, which is plotted since it is always available quickly, and has done well at longer range forecasts in the past). We do not get tracking points for the ECMWF or HWRF models at this point, so you'll have to go the raw plots to see those forecasts. Note that three of the "big four" models performed well in 2006, with the GFDL and GFS performing the best. The UKMET had a very poor showing in the Atlantic in 2006. However, the UKMET was the best-performing model in the Eastern Pacific in 2006, along with the GFDL and BAMM models.

The European Center's model (ECMWF) outperformed the "big four" consensus models for 72, 96, and 120 hours forecasts in the Atlantic. However, the ECMWF model was generally not available in time to be used by forecasters. Efforts are being made to make the ECMWF available in a more timely fashion for the 2007 season, which would be a big help. We also have the new HWRF (Hurricane Weather Research Forecast) model this year. In tests done on a number of hurricanes for past years, the HWRF performed about as well as the GFDL (Figure 2).



Figure 1. Track forecast skill in 2006 of the official forecast and the various models, compared to a "zero skill" forecast using NHC's CLIPER5 model. The CLIPER model (short for CLImatology and PERsistence) is a model that makes a forecast based on historical paths hurricane have taken, along with the fact that hurricanes tend to keep moving in the direction they are going (i.e., their current motion persists). Note that many models had a negative skill for their 120 hour (5 day) forecast. The official NHC forecast had about 10% skill at 5 days. Image credit: NHC.

Figure 2. Track errors for 48-hour forecasts from the 2006 version of the GFDL model (black) and the new HWRF model (red). The HWRF model performed better on some hurricane than the GFDL, and worse on others. Overall, the two models had about the same performance on the cases tested. Image credit: Naomi Surgi, NOAA Environmental Modeling Center.

In conclusion, the official NHC forecast outperforms all the individual models, particularly at long ranges. Looking at the individual model plots can be helpful to determine the uncertainty in the forecast, but it's tough to beat the NHC. In the case of Dean, where one model is an outlier from the rest, it is usually better to believe the consensus of the other models.

If you want to look at plots of the individual models, I've written a description of the various models and where to find these plots on our tropical weather page.

Arctic sea ice shrinks to record low
The National Snow and Ice Data Center announced today that Arctic sea ice has just surpassed the previous single-day (absolute minimum) record for the lowest extent ever measured by satellite. Satellite measurements began in 1979. Sea ice extent has fallen below the 2005 record low absolute minimum and is still melting. Sea ice extent is currently tracking at 5.26 million square kilometers (2.02 million square miles), just below the 2005 record absolute minimum of 5.32 million square kilometers (2.05 million square miles). This new record was set a full five weeks before the usual late September minima in ice extent, so truly unprecedented melting is occurring in the Arctic. The most recent images from the North Pole webcam show plenty of melt water and rainy conditions near the Pole.


Figure 2. Current extent of the polar sea ice, compared to the normal for this time in August (pink line). Image credit: National Snow and Ice Data Center.

I'll have an update Saturday morning.
Jeff Masters

Climate Change

Updated: 08:02 PM GMT am 16. August 2011

Permalink

Dean pounds Martinique; Erin soaks Texas; Sepat zeroes in on Taiwan

By: JeffMasters, 02:07 PM GMT am 17. August 2007

Hurricane Dean plowed into Martinique this morning as a Category 2 hurricane with 100 mph winds. Radar imagery from Meteo France (Figure 1) shows that the eye of Dean grazed the southern tip of Martinique at about 5am EDT this morning. Winds at the Martinique airport gusted as high as 89 mph, but the airport was north of the eyewall, and missed seeing the calm at the center (Figure 2). An intermediate report, not sent out as part of the regular observations received by wunderground, showed sustained winds of 75 mph, gusting to 103 mph, at 4:42am EDT. The airport is still sending in observations, which is a good sign, and it is likely that only the southern 1/4 of the island suffered heavy wind damage. The northern part of St. Lucia also suffered heavy wind damage, and Dean's winds lifted the roof off the pediatric wing at Victoria Hospital in St. Lucia's capital, Castries. Dean underwent a brief period of weakening just before hitting Martinque, thanks to some dry air intruding into the circulation, so may have brought only Category 1 winds to Martinique. The rains continue on the island this morning, and current radar imagery shows that Dean has intensified since departing Martinique. This is backed up by the 9:12am EDT Hurricane Hunter data, which found the pressure had dropped to 965 mb and the surface winds had risen to 105 mph. Dean's rainbands are dumping torrential rains on Martinique and Dominica, and life-threatening flooding and mudslides will be a major hazard on those islands today.


Figure 1. Radar image of Dean as it passed over the southern tip of Martinique. Image credit: Meteo France.


Figure 2. Current conditions for Friday, August 17, 2007 at Le Lamentin, Martinique. This airport is on the sheltered west side of the island, by the capital city of Fort de France, and was far enough north that the eyewall missed.

Latest model runs
The NOAA jet flew last night, collecting a large amount of high-quality data around Dean's environment that was used to initialize the latest (06Z) model runs. These models continue to unanimously show a grave threat to Jamaica and the Cayman Islands two days from now. Even if Dean misses these islands, it won't be by much, and residents should expect hurricane conditions, as Dean should be an extremely dangerous Category 4 storm by then. Haiti and eastern Cuba also appear likely to suffer severe flooding problems from Dean's outer spiral bands, but a direct hit in these areas appears unlikely. Lesser flooding problems, but still potentially life-threatening, will occur in Puerto Rico and the Dominican Republic. Additionally, Dean may grow so large and powerful that its spiral bands will cause heavy rains and flash flooding in Venezuela, Nicaragua, and Honduras. Hurricane Gilbert of 1988, which traced out a similar path, was large enough to cause several flooding deaths in each of those nations.

The NOAA jet mission did not help at all with narrowing down the uncertainty in the computer forecasts for the 4-5 day period, which remain divergent. The 06Z run of the GFS model takes Dean over the center of the Yucatan, then into the Texas/Mexico border region on Thursday. The 06Z GFDL is much faster and further north, taking Dean through the Yucatan Channel and into northeastern Texas near Galveston on Wednesday. The 06Z HWRF is in between and much slower, taking Dean over the tip of the Yucatan Peninsula by Wednesday. The problem is that each model has a different solution for the behavior of an upper-level low pressure system expected to be over the Gulf of Mexico early next week, and there is no is currently no way to guess which model will be right. Which model should you trust? I'll present the statistics for how these models performed last year in my update this afternoon.

Erin leaves a Texas-sized mess
Tropical Depression Erin continues to dump heavy rain on Texas, with another 2-5 inches expected today in west Texas. Erin dropped over 10 inches of rain near San Antonio, and areas near Houston received nine inches (Figure 3). Over 70 high water rescues were performed in the Houston area from Erin, and four people died in Texas due to flooding. The region encompassing Corpus Christi, Austin and Houston has already surpassed its average yearly rainfall total, with 40-48 inches having fallen. Erin's rains could be setting the stage for a major flooding disaster next week in Texas, should Hurricane Dean hit the state and dump heavy rain of its own. With Erin's rains leaving the soil saturated, Dean's rains will have nowhere to go.


Figure 3. Radar estimated precipitation from Tropical Storm Erin.

Super Typhoon Sepat
In the Western Pacific, residents of Taiwan are preparing for the arrival of Typhoon Sepat, which is expected to hit the island Saturday morning local time. Sepat is no longer a Super Typhoon, thanks to an eyewall replacement cycle it underwent yesterday. Still, the storm has a huge eye about 70 miles in diameter that will bring Category 3 or 4 winds to a large area of Taiwan. An impressive microwave image of Sepat's eyewall replacement is available from the Monterey NRL web site.

I'll have an update around 4pm EDT this afternoon, when I'll discuss which long-range computers model forecasts of Dean we should believe.

Jeff Masters

Updated: 03:21 PM GMT am 17. August 2007

Permalink

Dean's likely impact on the Caribbean; Super Typhoon Sepat update

By: JeffMasters, 08:17 PM GMT am 16. August 2007

The Hurricane Hunters made their first penetration of Hurricane Dean this afternoon, and found a strong Category 1 hurricane--100 mph winds, and surface pressure of 974-979 mb. Dean is now a Category 2 hurricane. Once the winds rise to 115 mph, it will be a Category 3 storm--a major hurricane. Wind shear remains near 5 knots, and is expected to remain low for the next five days. Recent satellite loops and reports from the Hurricane Hunters show that an eye has appeared. The eye is not fully formed, and has a gap on the west side. This gap is probably due to the presence of dry air on the storm's northwest side, which is getting wrapped into the storm. This dry air will persist through at least Friday, and should act to prevent Dean from undergoing rapid intensification until it clears the Lesser Antilles Islands. Dean is steadily moistening the environment around it, and may be able to overcome the dry air on Friday and put on a burst of rapid intensification. I expect Dean will become a large and extremely dangerous major hurricane by Saturday.

Latest model runs
The latest (12Z) model runs from this morning don't show much change from yesterday's runs for the 1-3 day period, but have a wider spread for the 4-5 day period. All the models show Dean moving through the Caribbean, passing over or just south of Jamaica and the Cayman Islands on Sunday or Monday, then into the western Caribbean. At that point, the models diverge. The NOGAPS model has the southernmost solution, taking Dean into northern Belize/Southern Mexico. The GFDL takes Dean through the Yucatan Channel and northwestward, towards western Louisiana. The other models are in between, with both the HWRF and UKMET nudging their tracks more to the north, grazing the tip of the Yucatan Peninsula. The GFDL and HWRF intensity forecasts both project Dean will be a Category 5 hurricane when it nears the Yucatan Peninsula. Tonight marks the first flight of the NOAA jet, and we'll have a much more reliable set of model runs Friday morning. Hopefully, this will narrow down the uncertainty of what will happen when Dean reaches the Gulf of Mexico.

Impacts on the Caribbean
Two storms in the historical record with a similar tracks and intensities to what we might expect for Dean in the Caribbean were Hurricane Gilbert of 1988 and Hurricane Ivan of 2004. Gilbert intensified to a Category 3 hurricane as it passed south of Haiti, and made a direct hit on Jamaica, passing the entire length of the island. Gilbert then began a remarkable rapid intensification spurt as it moved over the Cayman islands into the Western Caribbean, reaching an all-time record low pressure of 888 mb before it slammed into the Yucatan Peninsula. Ivan tracked a bit further south in the Caribbean, but was also a Category 5 storm after it passed Jamaica.

Lesser Antilles Islands of Martinique, Dominica, Guadaloupe, and St. Lucia
Dean will pass through the central Lesser Antilles Islands Friday morning. Martinique and Dominica will likely receive the harshest blow, although damage may also be significant on Guadaloupe and St. Lucia. Heavy wind damage will be the primary threat on these four islands, although torrential rains of 2-7 inches may cause flash flooding problems as well. Storm surge is generally not a problem in the Lesser Antilles, since the surge tends to flow around islands surrounded by deep water.

Surrounding Lesser Antilles islands from Grenada to Antigua
These islands will experience tropical storm force winds and heavy rains, but Dean's rapid forward speed will keep these rains below four inches. Puerto Rico can expect 1-3 inches of rain from the outer rainbands of Dean, but tropical storm force winds should stay just south of the island.

The Dominican Republic
The Dominican Republic suffered four deaths from flash floods due to heavy rains along the south side of the country during Ivan's passage. The northern part of the country was relatively unaffected. Five people died from Hurricane Gilbert. I expect similar effects from Dean, which will bring bands of very heavy rain over Hispaniola, leading to isolated life-threatening flash floods on Saturday. I don't think there will be any airport closures or major impact to tourist areas. The Barahona Peninsula, which juts out to the south, will be at greatest risk.

Haiti
Haiti is at major risk from heavy loss of life any time a hurricane brushes the island of Hispaniola. Hurricane Ivan did not pass close enough to the island to trigger major flash flooding, and did not kill anyone. However, thirty people died in Hurricane Gilbert. Dean will take a path similar to Gilbert's and will have a similar strength, so I expect severe flash flooding in the southern part of Haiti may cause many deaths. The airport in the capital of Port-au-Prince will likely close for a time Saturday and Sunday.

Jamaica
If you have travel plans to go to Jamaica, plan on spending a lot of time praying for the hurricane to miss, because that is what the locals will be doing. This seemed to be what spared Jamaica in 2004, when Hurricane Ivan made a beeline for the island, then suddenly turned and wobbled around the island. Ivan still killed 17 people in Jamaica and left 18,000 people homeless. Most of the major resorts and hotels fared well, and reopened a few days after Ivan passed. Damage on Jamaica totaled $360 million. Jamaica did not fare as well in Hurricane Gilbert, which made a direct hit as a Category 3 hurricane, killing 45. Gilbert dumped up to 27 inches of rain in the mountainous areas of Jamaica, causing severe flash flooding. Gilbert was the worst hurricane to hit Jamaica since Hurricane Charlie in 1951. Gilbert left $4 billion dollars in damage, and it was difficult to leave the island for over a week due to blocked roads and closed airports. If Dean makes a direct hit on Jamaica, expect to be stranded on the island for many days, with no power. If Dean makes a close pass but misses, as is more likely, expect a few days of hassle. All Jamaica airports will likely close on Sunday when Dean will begin to batter the island.

Cayman Islands
The poor Caymans got drilled by Ivan at Category 5 strength, and have still not fully recovered. However, the islands did a great job protecting the people there, and only suffered two deaths. 95% percent of the homes and other buildings (which generally follow South Florida's building codes) were damaged or destroyed. Expect Dean to perform a similar feat if it makes a direct hit as a Category 5. If Dean passes close but misses, the islands will fare much better--Gilbert passed 30 miles to the south of the Cayman Islands, and didn't kill anyone. There was very severe damage to crops, trees, and homes, but nothing near the level of the destruction wrought by Ivan.

Cancun and Cozumel
Mexico's Yucatan Peninsula got hammered by Gilbert, which hit as a Category 5 storm with 160 mph winds and a 15-20 foot storm surge. Ivan passed to the east of the area, largely sparing it. It's a little too early to speculate on what Dean might do, but I expect the Cancun and Cozumel airports will close on Monday. This will not be a repeat of Wilma, which hung around the Yucatan for three days. Dean is a fast moving storm that will bring about a day of bad weather to the affected locations. I'll talk more about Dean's likely impact on Mexico in a later blog. If you have plans to be in Cancun or Cozumel Monday, be prepared to endure a major hurricane.

Super Typhoon Sepat
In the Western Pacific, residents of Taiwan are anxiously watching Super Typhoon Sepat, which is expected to hit the island as Category 4 or 5 storm this weekend. Wunderground meteorologist Elaine Yang, who is from Taiwan, dug up this information for me: The last time Taiwan was hit by a super typhoon was back in 2005 when Category 4 typhoon Longwang (Dragon King in Chinese) made landfall at 0515 local time on October 2nd, just south of Hualien City. Over the last 13 years, there were two typhoons, July 1994 Tim and August 2000 Super Typhoon Bilis that had similar tracks to Sepat.

South and southeast of Taiwan are under the first level of alerts. The residents have been preparing for the arrival of Sepat. Retailer stores are taking down the advertisement boards, residents are nailing their windows, and farmers are in a hurry to harvest agriculture products. The tall trees over the major roads are being cut.

Links to watch:
Martinique current conditions.
St. Lucia current conditions.
Dominica current conditions.
Guadaloupe current conditions.
Martinique radar.

I'll have an update Friday morning.

Jeff Masters

Updated: 12:29 AM GMT am 17. August 2007

Permalink

Hurricane Dean, TD Erin, and Super Typhoon Sepat

By: JeffMasters, 12:38 PM GMT am 16. August 2007

Hurricane Dean continues to intensify as it continues westwards towards the central Lesser Antilles Islands. Wind shear has fallen to 5 knots this morning, and recent satellite loops show an eye has appeared. Dean's intensification is being slowed by dry air on it northwest side. This dry air is being wrapped into its circulation, and is keeping the storm relatively cloud-free on its northwest side.

Impact on the Lesser Antilles
Dean will likely pass through the central Lesser Antilles Islands Friday morning as a strong Category 1 or weak weak Category 2 hurricane. Dominica will likely receive the harshest blow, although damage may also be significant on Martinique and Guadaloupe. Heavy wind damage will be the primary threat on these three islands, although torrential rains of 2-7 inches may cause flash flooding problems as well. Storm surge is generally not a problem in the Lesser Antilles, since the surge tends to flow around islands surrounded by deep water. Storm surge values of 2-4 feet are expected with Dean.

Surrounding islands from Grenada to Antigua will experience tropical storm force winds and heavy rains, but Dean's rapid forward speed will keep these rains below four inches. Puerto Rico can expect 1-3 inches of rain from the outer rainbands of Dean, but tropical storm force winds should stay just south of the island. I'll discuss the likely impacts on the rest of the Caribbean this afternoon, in my 4pm EDT update.

What the models say
The latest (00Z or 06Z) model runs from last night and early this morning don't show much change from yesterday's runs. All the models show Dean moving through the Caribbean, passing very near Jamaica and the Cayman Islands on Sunday or Monday, then into the western Caribbean. None of the models show Dean moving northwards into Florida, and I don't see any feature in the steering currents that could potentially lead to a northern excursion by Dean into Puerto Rico, the Bahamas, or the East Coast of Florida. Landfall in the Yucatan are the preferred solutions, followed by an emergence into the Gulf of Mexico, with a second landfall near the Mexico/Texas border. I'd be surprised to see Dean make a turn northwards in the Gulf of Mexico towards Louisiana or points further east, as there are no strong troughs of low pressure coming across the U.S. until late next week.

The GFDL and HWRF models have been mysteriously weakening Dean and keeping the storm weak during its passage over the Lesser Antilles Islands, and through the eastern Caribbean. These models have not been correct, and a continued slow strengthening of the storm as indicated by the SHIPS intensity model seems in order. Dry air will continue to slow the intensification process down until the storm gets in the central Caribbean, at which time Dean should be able to reach Category 3 or 4 status. Once in the western Caribbean, where the ocean heat content is near the record levels observed during 2005 (Figure 1), Dean could reach Category 5 status.


Figure 1. Tropical Cyclone Heat Potential (TCHP) for August 14 2005 (top) and August 14 2007 (bottom). TCHP is a measure of the total heat energy available in the ocean. Record high values of TCHP were observed in 2005, and the TCHP is at similar levels in the Western Caribbean this year. Values of ocean heat content greater than 50 kJ/cm^2 (the light green regions in the plot above) have been shown to promote greater rates of intensity change for storms in moist air with low wind shear. Values greater than 90 kJ/cm^2 can lead to rapid intensification. Image credit: NOAA/AOML.

Erin
Tropical Depression Erin has been downgraded from a tropical storm after making landfall this morning over South Texas. Erin brought a storm surge of up to 2 feet to the coast, and some gusty winds up to 40 mph. Heavy rain will be the main threat from this storm, and rainfall amounts up to 3 inches have already been observed along the coast north of Corpus Christi (Figure 2). Although conditions were favorable for Erin to intensify, it never had time to consolidate its circulation and was not able to take advantage of these favorable conditions. Texas says to Erin, "Mahalo!" (Hawaiian for "thank you").


Figure 2. Current estimated rainfall for Erin from the Corpus Christi radar.

Super Typhoon Sepat
The Western Pacific has its own tropical cyclone drama unfolding, as residents of Taiwan anxiously watch Super Typhoon Sepat. A Super Typhoon is a tropical cyclone in the Western Pacific that has sustained winds of 150 mph or greater, and Sepat's maximum winds of 160 mph also make it a Category 5 storm (winds of 156 mph or greater characterize a Cat 5). Sepat is headed northwest towards Taiwan, and is expected to make landfall Sunday morning local time. There's not much I can see that will weaken Sepat before it makes landfall. Wind shear is 10 knots, and the ocean heat content is an extremely high 100 kilojoules per square centimeter along the forecast track. The typhoon may suffer a drop in intensity down to a Category 4 storm due to an eyewall replacement cycle, but would still strike a devastating blow to Taiwan at that intensity.

I'll have an update by 4pm EDT this afternoon, when data from the first Hurricane Hunter mission into Dean will be available. I will project the likely impact of Dean on the rest of the Caribbean, and discuss the latest model runs.

Jeff Masters

Updated: 12:39 PM GMT am 16. August 2007

Permalink

Afternoon update on Dean, Erin, and Flossie

By: JeffMasters, 07:47 PM GMT am 15. August 2007

Tropical Storm Dean continues to strengthen as it continues westwards towards the Lesser Antilles Islands. Wind shear has remained constant at 10 knots today, which has allowed the storm to consolidate its heavy thunderstorm activity, and recent microwave satellite imagery (Figure 1) shows that Dean has made a partial eyewall on the east side. Recent satellite loops show a small Central Dense Overcast (CDO) trying to form over the center. A CDO is a high canopy of cirrus clouds that forms over the center, and is the hallmark of a storm nearing hurricane strength. However, satellite loops also show that the storm is wrapping some dry air into its circulation on the west side, and this will slow down Dean's progress towards becoming a hurricane.


Figure 1. Microwave satellite image of Dean from the AMSU instrument on the polar-orbiting NOAA-18 spacecraft. Photo taken at 12:13pm EDT Wed Aug 15. Think of this image as a weather radar in space. Note the comma-shaped half-eyewall present on the east side of Dean.

Impact on the Lesser Antilles
All of the computer models forecast that Dean will hit the Lesser Antilles Islands. The cone of uncertainty Friday morning, when Dean is expected to pass through the islands, covers a wide stretch of ocean from Grenada to Antigua. We can expect that this uncertainty cone is too wide, since the steering currents are relatively stable right now and Dean is well-formed. This puts the islands of Martinique, Dominica, and Guadaloupe in the bulls-eye. If Dean strengthens to a Category 1 hurricane, one of these islands will suffer considerable damage from hurricane-force winds, and surrounding islands will have minor damage due to flooding rains and tropical storm force winds. I expect Dean will be a Category 2 storm, as forecast by the HWRF model (Figure 2). We can expect two of these islands will have heavy damage. Dean is rather small in size at present, so this wind damage will likely be confined to a relatively small area 30 miles wide. Flooding rains of 5-10 inches can be expected over a wider swath, perhaps 100 miles in diameter. The latest run of the GFDL model unrealistically projects Dean to be a weak tropical storm with a 1004 mb central pressure on Friday morning, when it crosses the Lesser Antilles near Dominica.

The latest (12Z) model runs from this morning don't show much change from last night's. All the models show Dean moving through the Caribbean, passing very near Jamaica on Sunday or Monday, then into the western Caribbean. None of the models show Dean moving northwards into Florida. Landfall in the Yucatan or western Cuba are the preferred solutions.


Figure 2. Forecast pressure 6-hourly rainfall for Dean at 8am EDT Friday, from the 12Z Wednesday run of the HWRF model. HWRF forecasts that Dean will hit Dominica as a Category 2 hurricane (976 mb pressure).

Erin
Tropical Storm Erin formed this morning over the Gulf of Mexico. A QuikSCAT satellite pass at 7:38am EDT today showed a weak circulation with winds of 25-35 mph on the northeast side. Since then, the Hurricane Hunters found 50 mph winds at flight level (1500 feet), which corresponds to about 40 mph winds at the surface. Winds at a buoy 70 miles south of Freeport, TX have risen to 21 kt, gusting to 27 kt. Long range radar out of Corpus Christi shows intensifying bands of rain starting to come ashore. Visible satellite loops show the storm is rapidly organizing, and it may have time to reach sustained winds of 55 mph before making landfall over the Texas coast on Thursday. Wind shear over Erin is only 5-10 knots, and an upper-level high pressure system has parked itself directly over the storm. This is an ideal situation for intensification, since the upper-level high provides very favorable outflow at the top of the storm, venting all the air forced up at the center of the storm.

The Big Island says Mahalo to Flossie
"Mahalo" in Hawaiian means "thank you", and residents of the Big Island have plenty to be thankful for today. Tropical Storm Flossie spared the island, as wind shear ripped up the storm just as it was about to make its closest approach to the island. Flossie decayed from a Category 2 hurricane to a tropical storm in just six hours, and her flooding rains never materialized. Radar estimates of rainfall from Flossie over the Big Island (Figure 3) are no more than an inch, and it is unlikely that the island will get as much as four inches from the storm the remainder of today. Some coastal flooding occurred because of waves up to 20 feet, and sustained winds of 40 mph gusting to 48 mph were observed at South Point, the extreme southern tip of the island.


Figure 3. Latest radar-estimated rainfall from Flossie.

I'll have an update Thursday morning.

Jeff Masters

Updated: 08:36 PM GMT am 15. August 2007

Permalink

Dean, Erin, and Flossie

By: JeffMasters, 02:14 PM GMT am 15. August 2007

Tropical Storm Dean is strengthening as it continues westwards towards the Lesser Antilles Islands. Wind shear has fallen to 10 knots this morning, which has allowed the storm to consolidate its heavy thunderstorm activity into a symmetrical pattern around the center. Estimates of intensity from both traditional satellite images and microwave images have shown a steady intensification this morning, and some well-formed spiral bands are now visible on satellite loops. The major impediment to intensification is the large amount of dry air (Figure 1) to the storm's north, and this will continue to be a problem for it until it reaches a moister environment near the Lesser Antilles Islands. This morning's QuikSCAT satellite pass missed Dean, but last night's pass showed a large and well-formed circulation.


Figure 1. Current water vapor satellite image of Dean, showing very dry air (brown colors) to the north of the storm. Image credit: NOAA.

Track forecast
None of the computer models are forecasting that Dean will miss the Lesser Antilles Islands. The trough of low pressure that will pass north of the islands on Saturday is now expected to be a bit weaker than earlier forecast, which should allow Dean to pass into the Caribbean on a west to west-northwest track. The trough is no longer forecast to spawn an upper-level low pressure system, which means that the danger to the U.S. East Coast north of the Carolinas is minimal. A ridge of high pressure is expected to build in after the first trough passes on Saturday, which should keep Dean on a west to west-northwest path into the middle of next week. The southernmost model solutions (GFDL, Canadian) take Dean into Honduras early next week. The more northerly solutions of the GFS and HWRF take Dean over Jamaica, then into the Gulf of Mexico. No models call for a threat to the east coast of Florida at present, but that could change once we see how strong Saturday's trough of low pressure really will be. The NOAA jet is scheduled to make its first flight Thursday night, and by Friday morning we should have a good set of model runs that will give us a more reliable idea of Dean's likely track. At present, it appears that Dean's main threat to the U.S. will be to the Gulf of Mexico coast.

Intensity forecast
With low shear and warm water ahead of it, Dean's intensification over the next few days will only be limited by the dry air to the north. I expect that this dry air will impede Dean enough so that the storm passes through the Lesser Antilles as a Category 1 hurricane. After that, the environment moistens, shear stays low, and the heat content of the ocean greatly increases. The 06Z run of new HWRF model is again very aggressive intensifying Dean after it crosses into the Caribbean, bringing the storm to 928 mb (Category 4) on Monday morning near Jamaica. The GFDL model is not nearly as aggressive, putting Dean at 964 mb (Category 2) Monday morning. I can't see any reason why Dean wouldn't become a Category 3 or 4 hurricane by the time it reaches the Cuba/Jamaica region, unless it passes very close to the mountainous island of Hispaniola.


Figure 2. Ocean Heat Content (OHC) in kilojoules per square centimeter along the forecast track of Dean. The left side of the image marks where Cuba. Values of ocean heat content greater than 50 kJ/cm^2 (the yellow regions in the plot above) have been shown to promote greater rates of intensity change for storms in moist air with low wind shear. Image credit: NOAA/CIRA/RAMMB.

Erin
Tropical Storm Erin formed this morning over the Gulf of Mexico. A QuikSCAT satellite pass at 7:38am EDT today showed a weak circulation with winds of 25-35 mph on the northeast side, where a major blow-up of thunderstorms is visible on the visible satellite loop. Wind shear over Erin is only 5-10 knots, and an upper-level high pressure system has parked itself directly over the storm. This is an ideal situation for intensification, since the upper-level high provides very favorable outflow at the top of the storm, venting all the air forced up at the center of the storm. The only thing holding back Erin from strengthening is its initial disorganization, and the latest satellite loops show that the storm appears to have overcome that problem. Spiral banding is starting to occur, along with good upper level outflow. Erin could grow in strength rapidly. Fortunately, the storm only has about 24 hours over water, so it should not be able to become more than a 55 mph tropical storm. Heavy rains will be the main threat from TD 5. A NOAA Hurricane Hunter aircraft is currently in the storm.

The Big Island says Mahalo to Flossie
"Mahalo" in Hawaiian means "thank you", and residents of the Big Island have plenty to be thankful for today. Tropical Storm Flossie spared the island, as wind shear ripped up the storm just as it was about to make its closest approach to the island. Flossie decayed from a Category 2 hurricane to a tropical storm in just six hours, and her flooding rains never materialized. Radar estimates of rainfall from Flossie over the Big Island (Figure 3) are no more than an inch, and it is unlikely that the island will get as much as four inches from the storm the remainder of today. Some coastal flooding occurred because of waves up to 20 feet, and sustained winds of 40 mph gusting to 48 mph were observed at South Point, the extreme southern tip of the island.


Figure 2. Latest radar-estimated rainfall from Flossie.

I may make some minor edits to this blog today as new information arrives. The next full update will probably be this afternoon.

Jeff Masters

Updated: 04:14 PM GMT am 15. August 2007

Permalink

Tropical Storm Dean, Gulf of Mexico development, and Hurricane Flossie

By: JeffMasters, 02:02 PM GMT am 14. August 2007

Tropical Storm Dean continues to struggle with wind shear of 15-20 knots this morning, but improved enough in organization since yesterday to be upgraded from a tropical depression. A QuikSCAT satellite pass at 6:17am EDT captured the large circulation nicely, and found top winds of 35-40 mph. Visible satellite loops show a little improvement in organization this morning, with one decent-looking spiral band to the north, and a small amount of upper level outflow to the south and west.


Figure 1. Ocean Heat Content (OHC) in kilojoules per square centimeter along the forecast track of Dean. The left side of the image marks where Puerto Rico is. Values of ocean heat content greater than 50 kJ/cm^2 (the yellow regions in the plot above) have been shown to promote greater rates of intensity change for storms in moist air with low wind shear. Image credit: NOAA/CIRA/RAMMB.

Track forecast
Two major models--the ECMWF and NOGAPS--stubbornly refuse to develop Dean at all. These models are in great likelihood wrong. The GFS, UKMET, and new HWRF model all develop Dean into a hurricane that threatens the central and northern Lesser Antilles Islands Friday or Saturday. The big question is how strong the trough of low pressure predicted to pass north of TD 4 on Saturday will be. If the trough is strong enough, it may be able to pull TD 4 far enough north so that it misses the Lesser Antilles. Another big question is, will the trough spawn a cut-off upper-level low pressure system off the Southeast coast of the U.S.? If so, this feature could act to steer TD 4 on a more northerly track early next week, increasing the threat to New England and the mid-Atlantic coast. If not, a high pressure ridge is expected to build in, forcing TD 4 westward into Florida. It's far too early to know which of these scenarios might occur. In any case, there is a significant possibility that TD 4 will hit the U.S. as a hurricane, and possibly a major hurricane. This could happen as early as Tuesday August 21. Of course, Dean could also stay farther south in the Caribbean, as forecast by the UKMET model, and eventually track into the Gulf of Mexico.

Intensity forecast
The 06Z run of new HWRF model indicates that Dean will not start intensifying until Wednesday afternoon. HWRF is then very aggressive intensifying the storm, bringing it to 956 mb (Category 3) on Saturday morning as it passes through the Lesser Antilles, then 920 mb (Category 4 or 5) Sunday morning near Puerto Rico. I'd be surprised to see Dean get that strong that fast, and HWRF is likely overdoing the intensification. The 06Z run of the GFDL model also shows a slow intensification starting Wednesday morning, followed by a steady increase to a Category 3 storm (956 mb) Sunday night as it passes through the northern Lesser Antilles. Given the upper-level high pressure system forecast to develop over Dean beginning Wednesday, combined with steadily increasing Sea Surface Temperatures and total oceanic heat content under the storm (Figure 1), intensification to a major Category 3 hurricane by Sunday is a reasonable forecast.

Gulf of Mexico disturbance a threat to Texas and Mexico
Thunderstorm activity over the Gulf or Mexico near the Yucatan Peninsula has gotten better organized this morning. Visible satellite loops show that a low level circulation has formed just northwest of the Yucatan Peninsula, and heavy thunderstorm activity is on the increase on the northeast side of this circulation. This disturbance has been labeled "Invest 91L" by NHC, and the preliminary model tracks (Figure 2) show that 91L is expected to move west-northwest into the south coast of Texas Wednesday afternoon. Wind shear over the disturbance is only 5-10 knots, and the upper-level low that was keeping high shear over the Gulf has now exited to the west. I expect a tropical depression will form here today, and An Air Force Hurricane Hunter airplane is scheduled to investigate 91L this afternoon at 1pm EDT.

The storm has the potential to organize quickly and become a tropical storm. Last night's run of the GFDL model brought 91L to the south Texas coast as a 55 mph tropical storm by Wednesday afternoon. Heavy rain will be the main threat from this storm.


Figure 2. Model tracks for TD 5.

Hurricane Flossie
Hurricane Flossie in the Eastern Pacific has finally succumbed to the twin effects of cooler SSTs and increased wind shear, and will pass 50-100 miles south of the southern tip of Hawaii today as a much weakened Category 1 or 2 hurricane. Sustained tropical storm force winds of 39 mph or greater currently extend out about 150 miles from the storm's center, so most of the southern part of the Big Island will be at risk of wind damage from Flossie. The biggest problem will be the rains of 10 inches of more could inundate the high mountain flanks of Mauna Loa volcano on the south part of the island. Fortunately, this part of the island is sparsely populated, and flash flooding will probably cause only limited damage. Storm surge will not be a big deal, since islands surrounded by deep water like Hawaii tend to have the storm surge flow around them, instead of up onto shore.

Long range radar animations from the Big Island show Flossie's approach nicely.


Figure 3. Latest long-range radar out of the Big Island.

I may make some minor edits to this blog today. The next full update will be late this afternoon or Wednesday morning.

Jeff Masters

Updated: 12:21 PM GMT am 15. August 2007

Permalink

TD 4 forms; Gulf of Mexico distubance a threat to Texas; Flossie powers towards Hawaii

By: JeffMasters, 06:22 PM GMT am 13. August 2007

The strong tropical wave that exited the coast of Africa Friday has strengthened into Tropical Depression Four this morning. This morning's QuikSCAT satellite pass at 3:21am EDT found a closed circulation centered near 12.5N 29W, with top winds of 35 mph. The pass captured only half of the circulation of TD 4, so the errors in these wind estimates may be high. Wind shear, thanks to strong upper-level winds from the east, has dropped from 20 knots to 15 knots today. This has allowed heavy thunderstorm activity to begin forming on the east side of TD 4's circulation, where it was previously absent (see NOAA's visible satellite loop.) This shear is expected to gradually decrease to 10 knots by Wednesday, which should allow the storm to grow and consolidate. SSTs will gradually warm as the storm heads westward, further aiding the potential for intensification. The total heat content of the ocean stays relatively low through the next 48 hours (Figure 1), so no rapid intensification is likely until Thursday, when the storm approaches the Lesser Antilles Islands. The 12Z (8am EDT) run of the GFDL model intensifies TD 4 into a Category 2 hurricane five days from now.


Figure 1. Ocean Heat Content (OHC) in kilojoules per square centimeter along the forecast track of TD 4. The left side of the image marks where the Lesser Antilles Islands are. Values of ocean heat content greater than 50 kJ/cm^2 (the yellow regions in the plot above) have been shown to promote greater rates of intensity change for storms in moist air with low wind shear. Image credit: NOAA/CIRA/RAMMB.

What the models say
Several major models--the ECMWF and NOGAPS--do not develop TD 4 at all. The GFS, UKMET, and new HWRF model all develop the system, and bring it to a strong tropical storm or hurricane by the time it reaches the Lesser Antilles Islands Friday. The big question is how strong the trough of low pressure predicted to pass north of TD 4 on Friday will be. If the trough is stronger than currently forecast, it may be able to pull TD 4 far enough north so that it misses the Lesser Antilles. This would be good for the islands, but potentially bad for the U.S. East Coast. The trough will likely not be strong enough to recurve TD 4 harmlessly out to sea, and the storm would then be forced westwards again by the next ridge of high pressure. A landfall along the U.S. East Coast as a hurricane--possibly a major hurricane--could result, unless the next trough of low pressure is strong enough to recurve the storm out to sea. The latest (12Z) run of the UKMET model has TD 4 passing through the very northernmost Lesser Antilles Islands. The 12Z GFDL, GFS, and HWRF have TD 4 jogging just far enough north that it would likely miss the islands. Some of these runs are considerably slower, delaying the possible impact to the islands to Saturday or Sunday. What may happen after 5 days is highly uncertain. Last night's GFS model run had TD 4 eventually making landfall south of Brownsville, Texas. This morning's run had it eventually hitting New England--a difference of about 2000 miles in landfall location!

Gulf of Mexico disturbance a threat to Texas and Mexico
Thunderstorm activity in association with a tropical wave interacting with a upper-level low pressure system over the Western Caribbean has gotten better organized this morning. This disturbance has been labeled "Invest 91L" by NHC this morning, and the preliminary model tracks (Figure 2) show that 91L is expected to move west-northwest into the Gulf of Mexico Tuesday and Wednesday. An upper-level low currently spinning over the Gulf of Mexico is bringing 10-20 knots of shear over the Gulf. This upper-level low is moving steadily westward, and should exit the Gulf by Wednesday, allowing wind shear over the Gulf to drop below 10 knots. An Air Force Hurricane Hunter airplane is scheduled to investigate 91L Tuesday afternoon at 1pm EDT. The 12Z (8am EDT) runs of the GFDL and HWRF models are predicting that a tropical depression could form, and there is plenty of time for one to form. The GFDL predicts landfall on Wednesday afternoon near Corpus Christi as a 50 mph tropical storm. The storm won't have enough time over water to become anything stronger than a 50 mph tropical storm, it appears. Residents along Mexican and Texas Gulf Coast should be prepared for heavy rains from 91L as early as Tuesday night.

What is an "Invest?
When a National Hurricane Center forecaster sees a tropical disturbance that may be a threat to develop into a tropical depression, the forecaster may label the disturbance an "Invest" and give it a tracking identification number. There is no formal definition of what qualifies as an "Invest". Declaring an "Invest" is merely done so that a set of forecasting aids like computer model track forecasts can be generated for the disturbance. The "Invest" is given a number 90-99, followed by a single letter corresponding to the ocean basin--"L" for the Atlantic, or "E" for the Eastern Pacific. Other warning agencies assign "Invests" for the other ocean basins--"W" for the Western Pacific, "A" for the Arabian Sea, etc.


Figure 2. Preliminary model tracks for Invest 91.

Hurricane Flossie
Hurricane Flossie in the Eastern Pacific has proved resistant to the twin effects of cooler SSTs and increased wind shear, and stubbornly remains a Category 4 hurricane. SSTs are about 27 C under the storm, which normally can only support a Category 2 hurricane. Wind shear is about 15 knots over Flossie, and should increase to 20 knots by Tuesday. Given that Flossie is so intense and well-formed, it will take time for the shear and cooler waters it is now traversing to significantly weaken the storm. Flossie should still be a dangerous Category 2 hurricane Tuesday afternoon, when she make her closest approach to the Big Island. Waves along the southeast coast of the Big Island of Hawaii will build to 12 feet by Tuesday morning, and rains of 10 inches of more could inundate the high mountain flanks of Mauna Loa volcano on the south part of the island. Fortunately, this part of the island is sparsely populated, and flash flooding will probably cause only limited damage. Storm surge will not be a big deal, since islands surrounded by deep water like Hawaii tend to have the storm surge flow around them, instead of up onto shore. The computer model runs continue to show good agreement, giving support to the official forecast calling for Flossie's passage 50-100 miles south of the Big Island. Sustained tropical storm force winds of 39 mph or greater currently extend out about 110 miles from the center of Flossie, so the Big Island could experience some damaging winds in addition to heavy rains and flash flooding. The Hurricane Hunters will be flying several missions into Flossie over the next 24 hours.


Figure 3. Latest satellite image of Hurricane Flossie.

I'll have an update Tuesday morning.

Jeff Masters

Updated: 06:51 PM GMT am 13. August 2007

Permalink

African wave nears tropical depression status; Flossie powers towards Hawaii

By: JeffMasters, 01:31 PM GMT am 13. August 2007

Thunderstorm activity in association with a tropical wave interacting with a upper-level low pressure system over the Western Caribbean remains disorganized. This unsettled weather is expected to move west-northwest into the Gulf of Mexico Tuesday and Wednesday. An upper-level low currently spinning over the Gulf of Mexico is bringing 20-30 knots of shear over the Gulf. This upper-level low is moving steadily westward, and should exit the Gulf by Wednesday, allowing wind shear over the Gulf to drop below 10 knots. I expect a tropical disturbance with heavy rains worthy of being called an "Invest" by NHC will develop in the Gulf on Wednesday. This disturbance may bring heavy rains to Gulf Coast of Mexico and Texas Thursday, as it moves west-northwest. There may be time for the disturbance to grow into a tropical depression before moving inland Thursday. None of the reliable forecast models are predicting a tropical depression will form in the Gulf, though.

What is an "Invest?
When a National Hurricane Center forecaster sees a tropical disturbance that may be a threat to develop into a tropical depression, the forecaster may label the disturbance an "Invest" and give it a tracking identification number. There is no formal definition of what qualifies as an "Invest". Declaring an "Invest" is merely done so that a set of forecasting aids like computer model track forecasts can be generated for the disturbance. The "Invest" is given a number 90-99, followed by a single letter corresponding to the ocean basin--"L" for the Atlantic, or "E" for the Eastern Pacific. Other warning agencies assign "Invests" for the other ocean basins--"W" for the Western Pacific, "A" for the Arabian Sea, etc.


Figure 1. Preliminary model tracks for Invest 90L.

Invest 90L
A strong tropical wave exited the coast of Africa Friday, and is now a 1007 mb low pressure system with heavy thunderstorm activity about 400 miles southwest of the Cape Verde Islands. NHC is referring to this system as 90L. This morning's QuikSCAT satellite pass at 3:21am EDT found a closed circulation centered near 12.5N 29W, with top winds of 40 mph. Satellite imagery thus far has not shown enough organization for a long enough period of time to justify calling 90L a tropical depression, but that should occur later today. The first visible images this morning show some fair spiral banding and a large clump of heavy thunderstorms on the west side of 90L's circulation. If 90L can maintain or improve this appearance for another 12 hours, NHC will likely call it a tropical depression by their 11pm EDT advisory. In fact, some of the products coming out of NHC are already referring to 90L as "Cyclone Four". Wind shear of 20 knots, thanks to strong upper-level winds from the east, are keeping any heavy thunderstorm activity from forming on the east side of 90L's circulation. This shear is expected to decrease to 15 knots by Tuesday morning, which should allow the storm to grow and consolidate. SSTs will gradually warm as the storm heads westward, further aiding the potential for intensification.

What the models say
Several major models--the ECMWF and GFDL--are no longer developing 90L into a tropical storm. The NOGAPS model never has, and continues not to. The GFS, UKMET, and new HWRF model all do develop the system, and bring it to a strong tropical storm or Category 1 hurricane by the time it reaches the Lesser Antilles Islands Thursday night or Friday. Some scenarios to consider:

1) A trough of low pressure is expected to move off the U.S. east Coast Friday, which may be able to deflect 90L northwards enough to pass north of the Lesser Antilles Islands. A high pressure ridge is then expected to build in, forcing 90L more westwards towards the U.S. East Coast. This is the scenario preferred by the Canadian model.

2) 90L will be far enough south and next weekend's trough will be weak enough that 90L will plow through the Caribbean, and not be deflected north of the Lesser Antilles Islands. This is the solution preferred by the GFS, UKMET, and HWRF models.

3) 90L will never develop, or will never become more than a weak tropical storm, due to unfavorable wind shear, dry air, or other factors.

Of the three scenarios, I believe #2 is most likely to occur--90L will develop into a tropical storm or hurricane that will hit the Lesser Antilles, and pass into the Caribbean. It currently appears unlikely that 90L will recurve harmlessly out to sea, since the storm is too far south and the steering pattern is not generating a sufficiently strong enough trough to recurve it. Residents throughout the Caribbean and U.S. should anticipate the possibility that 90L may become a hurricane--and possibly a major hurricane--that will not recurve. If you plan on being in the Lesser Antilles Islands Thursday August 16 - Sunday August 19, keep in mind there is a heightened risk of a tropical storm or Category 1 hurricane during that period. Be prepared to adjust your travel plans.


Figure 1. Latest satellite image of Hurricane Flossie.

Hurricane Flossie
Hurricane Flossie in the Eastern Pacific has proved resistant to the twin effects of cooler SSTs and increased wind shear, and stubbornly remains a Category 4 hurricane. Wind shear is about 15 knots over Flossie, and should increase to 20 knots by Tuesday. Given that Flossie is so large and well-formed, it will take time for the shear and cooler waters it is now traversing to significantly weaken the storm. Flossie should still be a dangerous Category 2 hurricane Tuesday afternoon, when she make her closest approach to the Big Island. Waves along the southeast coast of the Big Island of Hawaii will build to 12 feet by Tuesday morning, and rains of 10 inches of more could inundate the high mountain flanks of Mauna Loa volcano on the south part of the island. Fortunately, this part of the island is sparsely populated, and flash flooding will probably cause only limited damage. Storm surge will not be a big deal, since islands surrounded by deep water like Hawaii tend to have the storm surge flow around them, instead of up onto shore. The computer model runs continue to show good agreement, giving support to the official forecast calling for Flossie's passage 50-100 miles south of the Big Island. Sustained winds of 40 mph currently extend out about 90 miles from the center of Flossie, so the Big Island could experience some damaging winds in addition to heavy rains and flash flooding. The Hurricane Hunters will be flying several missions into Flossie over the next 24 hours.

On July 21, Tropical Depression Cosme passed 185 miles south of the Big Island, bringing rains of 3-5 inches over a six hour period. Cosme helped with the summer-long drought that has affected the entire island chain, but Hawaii could use another near miss by Flossie to alleviate the moderate to severe drought conditions that persist.

I'll have an update on things later this afternoon.
Jeff Masters

Permalink

African wave more organized; Hurricane Flossie still a Cat 4

By: JeffMasters, 02:02 PM GMT am 12. August 2007

Thunderstorm activity in association with a surface trough of low pressure over the Western Caribbean remains disorganized. None of the computer models than reliably forecast tropical storm formation are calling for development in this region. Wind shear is about 20 knots, which will keep any development slow.

A strong tropical wave exited the coast of Africa Friday, and is now a 1006 mb low pressure system with heavy thunderstorm activity about 200 miles south of the Cape Verde Islands. NHC is referring to this system as 90L. This morning's QuikSCAT satellite pass at 3:48am EDT found an elongated circulation centered near 12.5N 23W, with top winds of 30 mph. Satellite imagery at the NRL Monterey web site that most people use to track eastern Atlantic storms is not available this weekend. A substitute source of imagery is available at the main Navy weather web site. The thunderstorm activity surrounding 90L is currently not very expansive, thanks in part to relatively cool SSTs of 26-27C and high wind shear of 20 knots. However, the cloud pattern has gotten more organized this morning, with some spiral banding starting to occur. SSTs should increase above 27C by Tuesday, as 90L moves away from the African coast at 15-20 mph. Wind shear should drop to 10-20 knots, and I expect 90L will become a tropical depression on Monday or Tuesday.



What the computer models say
Watching the computer model runs for 90L is not for the faint of heart. All the major models except the NOGAPS develop the system into a tropical storm or hurricane that tracks westward over the Atlantic, reaching the lesser Antilles Islands as early as Thursday night, August 16. There are four possible scenarios to consider:

1) A strong trough of low pressure is forecast to move off the East Coast of the U.S. at that time, and this trough may deflect 90L northwards so that it misses the Lesser Antilles Islands, and then recurves harmlessly out to sea.

2) In keeping with the steering pattern we've observed since late July, the trough is expected to rapidly move onward, allowing a ridge of high pressure to build in. If the trough is not strong enough to recurve 90L out to sea, the storm will be forced to the west once more and eventually hit the East Coast of the U.S. This is the solution of last night's ECMWF model.

3) 90L will be far enough south and next weekend's trough will be weak enough that 90L will plow through the Caribbean, and not be deflected north of the Lesser Antilles Islands. The storm would eventually track into the Gulf of Mexico. This is the solution preferred by this morning's GFS model.

4) 90L will never develop, or will never become more than a weak tropical storm, due to unfavorable wind shear, dry air, or other factors. This is the solution of the NOGAPS model.

Of the four scenarios, I believe #2 or #3 are most likely to occur--90L will develop into a tropical storm or hurricane that will affect the Caribbean and/or U.S. East Coast. Residents throughout the Caribbean and U.S. should anticipate the possibility that 90L may become a hurricane--and possibly a major hurricane--that will not recurve. If you plan on being in the Lesser Antilles Islands Thursday August 16 - Sunday August 19, keep in mind there is a heightened risk of a tropical storm or hurricane during that period. Be prepared to adjust your travel plans.


Figure 1. Latest satellite image of Hurricane Flossie.

Hurricane Flossie
Hurricane Flossie in the Eastern Pacific put on an impressive burst of intensification yesterday that put her at Category 4 strength for a full day. This intensification happened despite the presence of waters that were only 27.5 C, with a very limited total heat content (a relatively shallow layer of warm water, instead of the deep layer usually needs for a hurricane to rapidly intensify). Flossie has likey peaked in intensity, and should decay to Category 1 hurricane status by Tuesday, thanks to the twin influences of cooler SSTs and increased wind shear. Waves along the southeast coast of the Big Island of Hawaii will build to 12 feet by Tuesday morning, and Flossie is expected to make her closest approach to the island Wednesday morning. This morning's model runs have come into better agreement, giving support to the official forecast calling for Flossie's passage 50-100 miles south of the Big Island. Sustained winds of 40 mph currently extend out about 90 miles from the center of Flossie, so the Big Island could experience some damaging winds. Heavy rains and flash flooding are also likely on the Big Island. Right now, it doesn't appear that the Hilo airport on the northeast end of the island will need to close, but the Kona airport on the west side may be forced to close on Wednesday when Flossie passes the island and the winds become more onshore.

On July 21, Tropical Depression Cosme passed 185 miles south of the Big Island, bringing rains of 3-5 inches over a six hour period. Cosme helped with the summer-long drought that has affected the entire island chain, but Hawaii could use another near miss by Flossie to alleviate the moderate to severe drought conditions that persist.

Jeff Masters

Updated: 02:04 PM GMT am 12. August 2007

Permalink

African wave a threat to develop; Hurricane Flossie intensifies to Category 4

By: JeffMasters, 03:01 PM GMT am 11. August 2007

Thunderstorm activity in association with a surface trough of low pressure near Jamaica remains disorganized. None of the reliable computer models are calling for a tropical depression to form in this region any more. The four reliable models for forecasting tropical storm formation are the GFS, ECMWF, NOGAPS, and UKMET. The NAM and Canadian GEM models are not reliable for forecasting formation of tropical storms.

A strong tropical wave exited the coast of Africa yesterday, and is now a 1006 mb low pressure system with heavy thunderstorm activity southeast of the Cape Verde Islands. The system is not well-organized at present, and has lost some of its thunderstorm activity since exiting the coast of Africa. This is typical for such systems, which take a day or two to adjust to their new oceanic environment before they build new thunderstorms. Wind shear has dropped from 30 knots early this morning to about 20 knots over the region, and wind shear is forecast to continue to drop as the wave continues westward. There is burst of dry air and African dust emerging from the African coast just north of the Cape Verdes, but this is probably too far away to affect the storm.

This system has the potential to become a tropical depression as early as Monday. Most of the computer models develop it into a tropical depression that moves west to west-northwest over the Atlantic towards the Lesser Antilles Islands, and I think this is a reasonable forecast that has a 60% chance of coming true. We are into mid-August, when these waves traditionally start to develop, and the dry air and dust associated with the Saharan Air Layer (SAL) does not appear to be a major impediment at present. Wind shear should be low enough and water temperatures warm enough. Water temperatures are about 26-27C south of the Cape Verdes Islands, and gradually warm to 27-28C southwest of the islands, so the chances for development increase as the system gets further west.

If you live in the Lesser Antilles Islands, or plan on visiting next week, keep in mind that this forecast puts the Lesser Antilles at increased risk of a seeing a tropical cyclone beginning Friday, August 17. For the period August 17 - August 20, there is at least a 10% chance that a tropical storm or hurricane will affect these islands. Given the high uncertainties that a storm will form, or even make it across the Atlantic without recurving, should it form, one should not be canceling any travel plans at this point. However, you should be giving serious thought to your hurricane plan and what you would do if a hurricane did blow through the islands. Also, keep in minds that several of the computer models develop the next wave moving off the coast of Africa, a week later than the current wave. We are into the active part of hurricane season, and we can expect that one of these waves will develop and threaten the Lesser Antilles Islands--and points beyond--in August.


Figure 1. Latest satellite image of Hurricane Flossie.

Hurricane Flossie
Hurricane Flossie in the Eastern Pacific put on an impressive burst of intensification this morning, and is now a major Category 4 hurricane. Flossie will remain over waters warm enough ( > 80 F) to support a hurricane through the weekend, then hit cooler waters and high wind shear on Monday that should be able to weaken her to a Category 1 hurricane or tropical storm by the time the storm approaches the Hawaiian Islands on Tuesday. Just how strong Flossie is able to stay will influence her track as she approaches Hawaii. A severely weakened Flossie will not be large enough to "feel" the influence of a trough of low pressure expected to path north of the islands on Tuesday. This trough could potentially recurve the storm and bring it over the islands. If Flossie does manage to stay large and strong, the storm is more likely to feel the trough pass perilously close to the islands as a strong tropical storm or Category 1 hurricane. The Hurricane Hunters are scheduled to fly into Flossie beginning Monday at 9am Hawaiian time.

Below are a few wunderphotos of some impressive severe thunderstorms that moved through Ohio and Pennsylvania yesterday. Wow!

Jeff Masters

Updated: 06:08 PM GMT am 11. August 2007

Permalink

Tropical update; record Arctic sea ice loss in July

By: JeffMasters, 02:29 PM GMT am 10. August 2007

Thunderstorm activity in association with a surface trough of low pressure near Jamaica has diminished this morning. However, we will have to watch this area this weekend, as most of the computer models forecast that low levels of wind shear will develop over the western Caribbean and southern Gulf of Mexico. Two computer models, the NOGAPS and ECMWF, predict a tropical depression could form in the region as early as Tuesday. The GFS and UKMET models do not. As there will be plenty of moisture and low wind shear, the limiting factor would seem to be lack of a trigger to get things spinning. We'll have to wait and see on this.

An area of disturbed weather off the North Carolina/Virginia coast is associated with the tail end of a cold front that pushed off the East Coast. Sea surface temperatures are warm enough (80-84 F) and wind shear is low enough (10-20 knots) to allow some slow development over the next few days. This disturbance is expected to slide off to the northeast away from land, and I don't expect a tropical depression will form.

Thunderstorm activity has picked up on the west coast of Africa, and it appears that one or two strong African waves will push off the coast over the coming week. Most of the computer models forecast that one of these waves will develop into a tropical depression. We are into mid-August, when these waves traditionally start to develop, and the dry air and dust associated with the Saharan Air Layer (SAL) does not appear to be a major impediment at present. Thus, I think it likely a tropical depression will form off the coast of Africa 3-7 days from now.

Hurricane Flossie in the Eastern Pacific remains a threat to the Big Island of Hawaii, and could bring rains there as early as Wednesday.

Record July Arctic sea ice loss
Sea ice extent in the Arctic in July 2007 set a record low, posting a large 7% decline compared to July 2006. July marked the third month this year that a record monthly low was set. Arctic sea ice coverage in July has declined by about 26% since measurements began in 1979 (Figure 1). The trailing end of Figure 1 shows a very striking drop, so it's worth investigating this decline in more detail.



Figure 1. Arctic sea ice extent for July, for the years 1979-2007. July 2007 had the lowest Arctic sea ice extent since satellite measurements began in 1979. July sea ice coverage has declined about 26% since 1979. Image credit: National Snow and Ice Data Center.

What caused the July sea ice loss?

Sea ice in July over the past 28 years has shown a steady decline, punctuated by ups and downs characteristic of year-to-year natural variability in the weather patterns over the Arctic. The steady decline is largely due to increasing temperatures in the Arctic from global warming, but a significant portion is due to changing wind patterns. As I discussed in detail in a blog earlier this year, much of the 1990s saw lower than average pressure over the Arctic, which drove stronger than average westerly winds along the north coast of Canada. These west-to-east winds acted to push ice out of the Arctic through Fram Strait, the region between Greenland and Europe. Was a similar wind pattern responsible for the July 2007 decline in ice?



Figure 2. Surface wind for the Arctic averaged for July 2007.

A plot of the surface wind speed for July 2007 (Figure 2) shows that the meteorology of July 2007 led to a wind pattern the opposite of the one in the early 1990s that pushed so much ice out of the Arctic. In July 2007, surface winds blew from east to west along the north shore of Canada, rotating clockwise around a high pressure system over the North Pole. It is more difficult to flush ice out of the Arctic with this kind of wind pattern. There were very strong north-to-south winds over Fram Strait--in excess of 7 m/s (about 14 mph). The wind is normally nearly calm in this region in July, so these strong winds did account for a small portion of the July 2007 record sea ice loss.

By comparing the sea ice coverage in July 2006 versus 2007 (Figure 3), we can identify areas along the northern coasts of Russia and Canada where most of the melting in 2007 occurred. A plot of the temperature anomalies (how far temperature differed from average) for July 2007 (Figure 4) show that the greatest ice loss in July 2007 occurred where temperatures much above average occurred. Thus, it appears that warming, and not wind patterns, was primarily responsible for the record July 2007 sea ice loss.

I asked Dr. Mark Serreze, a Arctic sea ice expert at the National Snow and Ice Data Center, about the record July sea ice loss. His comments:

We see a strongly anticyclonic atmospheric circulation at sea level in July over the Arctic Ocean. This seems to be fostering rather clear skies, promoting strong melt. Also, if you look at the temperature anomalies for July, there is an area of very warm conditions along eastern Siberia, on the west side of the anticyclone where winds have a southerly component. Southerly winds will also "push" the ice away from shore, helping to reduce ice extent along the Siberian coast. Having said this, we are also strongly seeing "memory" of past conditions. We started out on a bad footing with ice extent in May 2007 well below norms. There also seems to be very little thick ice in the Arctic Ocean--as the ice is thinner, large areas can melt out in summer. At the current rate of loss, it's a good bet that we will exceed the record 2005 September ice minimum. However, last year, we were on track to set a new record, until it got colder and stormier in August. In essence, we were "saved by the bell". Hence, we'll just have to wait and see.



Figure 3. Comparison of Sea ice extent for July 2006 and July 2007. Major sea ice loss in July 2007 compared to July 20006 occurred along the north coast of Russia and Canada.



Figure 4. Temperature anomalies (how far temperature differed from average) for July 2007. Much warmer than average temperatures were observed over the land areas adjacent to where the major sea ice losses occurred.

The implications
This July's major loss of sea ice will amplify sea ice loss the remainder of the summer, due to a positive feedback loop. As sea ice melts in response to rising temperatures, more of the dark ocean is exposed, allowing it to absorb more of the sun's energy. This further increases air temperatures, ocean temperatures, and ice melt in a process know as the "ice-albedo feedback" (albedo means how much sunlight a surface reflects). The July 2007 ice loss may mean that a runaway "ice-albedo feedback" has taken hold, which will amplify until the Arctic Ocean is entirely ice-free later this century. Other scientists will disagree, but I believe that such a runaway ice-albedo feedback has taken hold.

The melting of the Arctic sea ice will not raise ocean levels appreciably, since the ice is already floating in the ocean. However, it will bring warmer temperatures to the Arctic, which will accelerate the melting of the Greenland Ice Cap. This ice sheet holds enough water to raise sea level 20 feet--though much less melting is expected this century, with only a 0.6-1.9 foot sea level rise predicted. Loss of Arctic sea ice will also dramatically change the global weather and precipitation patterns. For example, the jet stream should move further north, bringing more precipitation to the Arctic, and more frequent droughts over the U.S. In any case, the reduced Artic sea ice should give us another delayed start to winter in the Northern Hemisphere this year.

Beginning today, the National Snow and Ice Data Center has begun a blog providing expert analysis of this summer's record Arctic sea ice loss. Expect weekly updates from now until Arctic sea ice reaches its minimum extent in September.

Jeff Masters

Climate Change

Updated: 08:02 PM GMT am 16. August 2011

Permalink

NOAA hurricane forecast still calling for very active season

By: JeffMasters, 03:10 PM GMT am 09. August 2007

NOAA released their August Atlantic Hurricane Season Outlook Update today. The forecast calls for a likely range of 13-16 named storms, 7-9 hurricanes, and 3-5 major hurricanes. The forecast is almost unchanged from their May 22 forecast of 13-17 named storms, 7-10 hurricanes, and 3-5 major hurricanes (a normal season has 10-11 named storms, 6 hurricanes, and 2 intense hurricanes). The forecast team cites the lack of El Nino, sea surface temperatures in the western tropical Atlantic and Caribbean Sea well above average (+0.56 C), and the elements shown in Figure 1 below, as justification for their continued forecast of much above normal hurricane activity this year. They also note:

Historically, similar conditions have typically produced 2-4 hurricane strikes in the continental United States and 2-3 hurricanes in the region around the Caribbean Sea. However, it is currently not possible to confidently predict at these extended ranges the number or intensity of landfalling hurricanes, or whether a given locality will be impacted by a hurricane this season.


Figure 1. Graphic of the meteorological justifications for NOAA's forecast of a much-above normal Atlantic hurricane season.

NOAA (13-16 storms), the Colorado State Dr. Gray/Phil Klotzbach team (15 storms), and TSR (15 storms) are all calling for an awful lot of tropical storm activity in a relatively short period of time. I think it is likely that the total number of named storms this year will be at the lower end of NOAA's range--13. This is close to what the new UKMET office forecast was calling for in June--12 named storms. Given the current SST patterns and behavior of the steering currents, at least one major hurricane affecting the Caribbean and one major hurricane hitting the U.S. is a good bet this season.

Tropical update
The Tropical Atlantic is relatively quiet. The Intertropical Convergence Zone (ITCZ), the region of strong thunderstorms usually present between Africa and the Lesser Antilles Islands, has been almost absent the past three days. Thunderstorm activity in association with a surface trough of low pressure is over the Dominican Republic and extending northwards over the southeastern Bahama Islands and southwards over the central Caribbean. There is 20-30 knots of wind shear over this region, and wind shear is expected to remain too high to allow any development over the next few days. An area of thunderstorms off the North Carolina/Virginia coast is associated with the tail end of a cold front that pushed off the East Coast yesterday. Sea surface temperatures are warm enough (80-84 F) and wind shear is low enough (15-20 knots) to allow some slow development over the next few days. However, conditions are marginal enough that I'd be surprised to see a tropical depression form here.



What the computer models forecast
Most of the computer models are forecasting very low levels of wind shear for the western Caribbean and southern Gulf of Mexico developing early next week. Two computer models, the NOGAPS and ECMWF, predict a tropical depression could form in this region as early as Tuesday. The GFS and UKMET models do not, but do show something developing off the coast of Africa next week. We are starting to approach the peak part of hurricane season, and I expect that our next tropical depression will form in one of these two areas by the end of next week.

Pacific storms
Tropical Storm Flossie formed yesterday in the Eastern Pacific, and could be a threat to Hawaii by Wednesday. In the Western Pacific, Tropical Depression Pabuk hit China this morning, bringing up to 10 inches of rain. Pabuk killed 11 in the Philippines earlier this week in rain-triggered landslides. However, Pabuk ended a 3-month long drought in the Philippines that had priests throughout the country urging their parishioners to pray for rain. Rains from Pabuk also added to flooding problems in Vietnam, where at least 43 people have died in severe flooding this week.

Jeff Masters

Permalink

Major air pollution episode underway

By: JeffMasters, 02:49 PM GMT am 08. August 2007

The Tropical Atlantic is exceptionally quiet, with very little thunderstorm activity anywhere. There are no threat areas to discuss, and none of the reliable models are forecasting development of a tropical storm over the next seven days.

What may be the most deadly weather event to affect the U.S. this year is underway across large portions of the central and eastern U.S. A large area of high pressure with light winds has settled over the region, bringing unhealthful levels of ozone and fine particulate matter to many major cities. High pressure systems are regions where the air gradually sinks, warming as it approaches the surface. This warming, sinking air creates a layer of air aloft (typically near 3000 feet in altitude) that is warmer than the air beneath it. This "upper air inversion" acts as a lid on the atmosphere, keeping pollutants trapped near the surface. Updrafts carrying surface air into the inversion suddenly encounter air that is warmer and less dense, so the updraft dies and the pollutants that they were trying to carry aloft settle back down towards the surface. If the high pressure region is large, an extensive area of light winds at the surface will exist, keeping the pollutants trapped under the inversion from being blown away horizontally. If the high pressure system stays in place for several days, pollutants will accumulate day by day, reaching levels harmful to human health and triggering a sharp rise in the death rate. "Particulate matter," also known as particle pollution or PM, is the pollutant that causes the largest rise in the death rate. Particulate matter less than 2.5 microns in diameter (PM 2.5) is particularly harmful to the lungs. A double-whammy dose of ozone pollution is also occurring this week, since temperatures are warm enough to drive the chemical reactions that form ozone.


Figure 1. Current wind speeds over the Southeast U.S.

Figure 2. Current visibility over the Southeast U.S.

The air pollution episode began August 1 over Indiana and Kentucky, then spread eastward to Ohio and Pennsylvania. By late last week, the stagnant conditions migrated southwards to the Southeast U.S. Wind speeds near the surface (Figure 1) have been less than 5 mph over much of the Southeastern U.S. the past few days, leading to visibilities less than seven miles due to fog and pollution (Figure 2). A small portion of the pollution is due to smoke from forest fires burning in the Northwestern U.S. (Figure 3).


Figure 3. Satellite image from NASA's Aqua satellite on Saturday, August 4, shows a large area of pollution over the Central and Southeast U.S. Forest fires burning in Montana and Idaho (bottom image) are visible, but smoke from these fires is being wafted northeastward into Canada.

On Tuesday, levels of fine particulate matter pollution exceeded the federal air quality standard of 35 ug/M3 over PA, NJ, MD, CT, DE, OH, SC, TN, GA and AL. The pollution episode is expected to be at its worst Wednesday in Georgia and North Carolinas; air pollution action days have been declared in major cites in those states, as well as St. Louis, Cincinnati, Philadelphia, Knoxville, and Wilmington. The air pollution episode should continue into Thursday, then ease on Friday when rains, higher winds, and more mixing of pollutants is expected.

Health Tip: Cut back on strenuous outdoor exercise when air quality is expected to be unhealthy. Exercise during the early morning or late evening hours when ozone levels are at the lowest levels of the day. This is especially important for children and other sensitive groups. This air pollution episode has likely contributed to the premature deaths of several hundred people already. As I discussed in detail in a blog in May, air pollution is thought to be responsible for tens of thousands of premature deaths across the U.S. each year. A major air pollution episode like this week's will contribute significantly to that toll.

Check out the EPA Airnow web site for detailed information on this ongoing pollution episode. Also, thanks go to Alex J. Sagady for providing information for this blog.

Jeff Masters

Air and Water Pollution

Updated: 09:36 PM GMT am 13. Juli 2011

Permalink

Former NHC chief Bob Burpee dies

By: JeffMasters, 01:51 PM GMT am 07. August 2007

The Tropical Atlantic is quiet. There are no threat areas to discuss, and none of the reliable models are forecasting development of a tropical storm over the next seven days.

Dr. Robert Burpee, former head of the National Hurricane Center, died on Tuesday July 31 after a long illness. He was 65. He served as director of NHC from July 1995 to August 1997, and was forced to step down because of serious medical problems. I had the honor of working with Bob from 1986 through 1990, when I flew with the NOAA Hurricane Hunters as a Flight Meteorologist. Bob served as director of NOAA's Hurricane Research Division (HRD) at that time, and flew on many hurricane hunter missions, accumulating 260 penetrations into the eyes of hurricanes. My most memorable flight with him came in 1989, when we endured a flight into Category 5 Hurricane Hugo which nearly killed us. When we finally escaped the eye of Hugo after being trapped in it for an hour, I remember his masterful understatement, "It feels a little better now, outside the eye!" Bob was always enthusiastic, knowledgeable, and made a great leader. It a shame Bob's career was cut short by his illness--he had unmatched integrity and scientific knowledge. Max Mayfield's blog has a nice tribute to Dr. Burpee.


Figure 1. From left to right: Paul Willis, Frank Marks, and Bob Burpee of NOAA's Hurricane Research Division plan a hurricane research flight. We flew that day out of San Juan, Puerto Rico and studied Category 3 Hurricane Emily as it made landfall in Hispaniola (September, 1987).

Jeff Masters

Updated: 02:00 PM GMT am 07. August 2007

Permalink

A quiet week ahead in the tropics?

By: JeffMasters, 02:10 PM GMT am 06. August 2007

Well, I'm back to blogging after an enjoyable week off camping and paddling along Lake Huron. It was a good time to be gone--significant landfalling storms in late July and early August are uncommon, and this year was no exception. The quiet conditions look likely to continue this week. There are no threat areas to discuss in the Atlantic, and none of the reliable computer models are forecasting any development over the next seven days. The Intertropical Convergence Zone (ITCZ), the band of active thunderstorms between Africa and the Lesser Antilles Islands that frequently serves as a genesis region for August tropical storms, is relatively quiet right now. However, it would not surprise me to see some activity in the ITCZ late this week. Ocean temperatures continue to warm, headed towards their early September peak. Wind shear and Saharan dry air/dust activity are near normal or below normal between Africa and the Lesser Antilles. The long-range steering current forecast from the GFS model continues to show a near-normal pattern, with the jet stream not forecast to "lock in" to a configuration that will enhance the risk of a hurricane strike to any particular region. But it isn't until around August 18 that hurricane activity historically begins its major activity period, so we may still have another quiet week or two ahead.


Figure 1. This morning's visible satellite image from the Navy web site, showing a relatively quiet Intertropical Convergence Zone (ITCZ) and a modest cloud of African dust just north of the Cape Verde Islands.

I'll be back with a new blog on Tuesday.

Jeff Masters

Updated: 02:12 PM GMT am 06. August 2007

Permalink

A Quiet Weekend In The Tropics

By: JeffMasters, 02:03 PM GMT am 04. August 2007

Guest Blogger Margie Kieper

................................................Time Magazine Aug 13 2007


Time Magazine's headline article this week is on New Orlean's [lack of] recovery on Katrina. Two years have passed and the city is not any more ready for a major hurricane (or a minor one) than it was two years ago. Is anyone surprised? It is a harshly-worded article and a scathing indictment of USACE and Louisiana politicians. The lede to the article does get something correct:

The most important thing to remember about the drowning of New Orleans is that it wasn't a natural disaster. It was a man-made disaster, created by lousy engineering, misplaced priorities and pork-barrel politics. Katrina was not the Category 5 killer the Big Easy had always feared; it was a Category 3 storm that missed New Orleans, where it was at worst a weak 2. The city's defenses should have withstood its surges, and if they had we never would have seen the squalor in the Superdome, the desperation on the rooftops, the shocking tableau of the Mardi Gras city underwater for weeks.
However, the scientific explanation they provide for the sinking of NOLA (and southeast Louisiana and surrounding areas) is not correct:

The straitjacketed river now carries less than half its original sediment load down to Louisiana. So there's little new land-building material to offset the natural erosion of the coast, much less the unnatural rising of the sea fueled by global warming.

The result is that New Orleans is sinking, and about 30% of the coast's wetlands have slipped into the Gulf...
This does not address subsidence, and that is a serious omission. The subsidence in SE Louisiana is caused by the sediment that has flowed down to the mouth of the river for centuries. The extra weight of the land is continuing to cause that area to sink, regardless of whether it is in a swamp that has been drained or not, and the area affected by this subsidence includes the Mississippi coastline as well, and likely all the way to the Florida coastline. As an expert in this area noted to me after reading the article, "If all the possible problems are not included then all of the possible solutions are faulty."

The article focuses on Louisiana and New Orleans politics, and USACE and its past and proposed management of the Mississippi River. It blasts "MR GO," and calls the Industrial Canal lock a "white elephant," and notes that NOLA's main hurricane project "was 37 years behind schedule when Katrina hit." The article notes:

Louisiana Senators Vitter and Mary Landrieu promptly proposed a bloated quarter-trillion-dollar Louisiana reconstruction bill, drafted by lobbyists for oil, shipping and other corporate interests. The request included $40 billion for the Corps10 times the agency's budget for the rest of the nationincluding nonreconstruction projects like the Industrial Canal lock and a New Iberia port deepening that had flunked the Corps' cost-benefit tests. It also included pre-Katrina coastal levee schemes, with names like Morganza-to-the-Gulf and Donaldsonville-to-the-Gulf to suggest their grandiose sweep. The bill stalled after it was widely mocked as legislative looting, but it sent the message that pre-Katrina priorities were still in effect. Vitter kept pushing a measure to help timber companies harvest cypress swamps. Landrieu tucked language into emergency bills ordering the Corps to redo its New Iberia analysis and fast-tracking the Industrial Canal lock. "Katrina was just a perfect excuse to pull the old pork off the shelf in the name of otherwise-we-drown," says Tulane law professor Oliver Houck, the sage of Louisiana environmentalism. "And away we go: another Louisiana hayride."
The article is packed with information and well worth reading. Will it result in any action to address the issues it raised? Only time will tell.


* * * * * * *

In the last 24 hours, Invest 98W, northwest of the Marianas and Guam, has steadily developed and is now Tropical Storm Pabuk. It is scheduled to strengthen and hit Taipei, Taiwain, in three days. Environmental conditions appear good for strengthing from 00Z 6 Aug through 00Z 7 Aug.

Pabuk IR


The area of disturbed weather associated with Tropical Depression 06W is generating a lot of rain over southern Vietnam, as it slides up the coast, with the convection sheared to thw southwest.

* * * * * * *

And in all probability a quiet week in the North Atlantic, although there is continued activity percolating in the other basins.

Here's an image of the extratropical ex-Chantal, courtesy of Dundee, as it slides bewteen Iceland and Scotland. Evidence of the Saharan Air Layer (SAL) can be seen in the lower portion of the image.

Ex-Chantal, now ET


This image of the ITCZ in the NATL was formed by combining GOES and METEOSAT images. Waves of SAL continue to accompany each tropical wave off Africa, and the SAL continues to impact convection in the ITCZ between about 25W and 60W, so it is unlikely to see any development in this area (and by development I mean at minimum a tropical depression). Dry air continues to move into the Eastern Caribbean as a result of the continued SAL moving into the eastern NATL.

RGB image, central and east tropical NATL

Updated: 04:08 PM GMT am 08. August 2007

Permalink

Friday Funnies

By: JeffMasters, 02:05 PM GMT am 03. August 2007

Guest Blogger Margie Kieper


What's wrong with this picture?

a) how many times do I have to tell you, don't clean the machine guns and grenade launchers on my clean white towels.
b) it's correct to camouflage by wearing brown in front of a brown building, as long as you have a tan -- not a sunburn.
c) never grow biceps larger than your head.
d) this man is not forecasting hurricanes.
e) absolutely nothing.

Stacy Stewart in Iraq

This is a photo of senior NHC forecaster Stacy Stewart on combat duty in Iraq, as detailed by the Miami Herald, "Today's outlook for hurricane forecaster Stacy Stewart: mostly sunny, high of 114 degrees, constant threat of death." Stacy recently deployed to Iraq, where he'll be serving a tour of duty through at least early summer 2008.


And what's wrong with this picture?

a) there's no tropical cyclones in the Caribbean or GOM.
b) there's no tropical cyclones in the Caribbean or GOM; absolutely nothing.
c) oh dear -- there are blobs in the Caribbean and GOM -- should I be worried?
d) those Canadians don't feel the need to hide part of the satellite image under a big round logo

NATL vis+ir


The early morning Tropical Weather Outlook notes:

A TROPICAL WAVE OVER THE CENTRAL CARIBBEAN SEA IS MOVING RAPIDLY WESTWARD AND IS EXPECTED TO BRING CLOUDINESS AND SHOWERS TO JAMAICA...THE CAYMAN ISLANDS AND CENTRAL AMERICA TODAY AND SATURDAY. THERE IS STILL SOME POTENTIAL FOR TROPICAL DEPRESSION FORMATION BEFORE THE WAVE REACHES CENTRAL AMERICA IN A DAY OR SO.

CLOUDINESS AND SHOWERS OVER THE NORTHEASTERN GULF OF MEXICO ASSOCIATED WITH A BROAD TROUGH OF LOW PRESSURE REMAIN DISORGANIZED. DEVELOPMENT...IF ANY...SHOULD BE SLOW TO OCCUR.

Now that the hysteria over a mere vigorous tropical wave has subsided (somewhat), remember what I mentioned at the beginning of the week -- in a couple of days there could be a chance for tropical development as the wave moves into the Western Caribbean or the East Pacific. Now you can become hysterical.


What's wrong with this story?

Once upon a time there were three little pigs.

The first little pig built his house out of straw. The second little pig built his house out of sticks. The third little pig built his house out of bricks.

One day the big bad wolf came along and said "Let me in, Let me in, little pig or I'll huff and I'll puff and I'll blow your house in!"

"Not by the hair of my chinny chin chin", said the first little pig.

And he huffed, and he puffed, and he blew the house down.

Next the wolf came to the house of sticks. "Let me in, Let me in, little pig or I'll huff and I'll puff and I'll blow your house in!"

"Not by the hair of my chinny chin chin", said the second little pig.

And he huffed, and he puffed, and he blew the house down.

Finally, the wolf came to the house of bricks. The three little pigs yelled out the window to the wolf, "This brick house was built to withstand 155 mph winds, tested in the Wall of Wind, and you'll never blow it down! We feel quite safe here."

The wolf purchased an ACME Do-It-Yourself Hurricane Kit from his friend Wiley E, and generated a storm surge that washed away the brick house and the three pigs with it.
a) the three little pigs didn't have a hurricane plan
b) the three little pigs didn't evacuate
c) don't build for wind in a surge zone

Every year now, we see news articles about coastal surveys, that seem to indicate a lot of people living on the coast aren't ready for hurricane season. Whether that is the case or not, one of the things from this year's survey noted was, "One out of three (34%) do not know if their home is located in an evacuation zone." Well there are a lot of things that remain to be fixed or improved regards hurricane preparation on the coast, but this is one that can be taken care of!

If you live in Florida, you have access online to the best emergency management website in the country: FloridaDisaster.org. Included on this website are storm surge maps for coastal counties. For instance, if you live in Pensacola, you can pull up the Escambia county map and see, to some extent, what category hurricane would flood your area:

Escambia surge zone


If you live in Mississippi and go to MS EMA, you won't find anything nearly as interesting. However you are fortunate, because a very detailed surge map exists on the USACE website, generated by a Hurricane Evacuation Study that used the results of thousands of SLOSH runs. The flooding that occurred with Katrina correlates very well to these maps. With the Jackson County maps, it can be seen that the surge can travel up the river plain all the way to the county line, and looking at a particular plate shows the surge detail for Pascagoula. That's right, almost the entire city floods in a direct hit from a category 2 hurricane.

Jackson county surge zone


Jackson county surge zone


So if your county has one of these maps available, you can look at it now, and have an idea of how vulnerable your home is to storm surge from hurricanes. But remember that regardless, your county emergency management will be well-informed via an NHC liason and will be executing evacuation plans based on the information provided by NHC and the local NWSFO. All of the Mississippi coastal residents who died from surge in Katrina were in locations where evacuation had been requested. If you are told to evacuate -- there's a good reason.

* * * * * * *

The Colorado State Univeristy forecast will be out shortly and numbers have been reduced from 17/8/4 to 15/8/4.

A tropical depression that formed in the South China Sea is forecast to become a minimal tropical storm, tracking towards North Vietnam. Just an interesting note: this TD developed under a considerable amount of shear, but it had a solid low level circulation, warm SSTs, considerable ocean heat content, and was in a moist environment.

Usagi, now a tropical storm, has moved away from Japan and is undergoing extratropical transition. Usagi brought high rainfall to the area. Apparently the Japanese take delivering the mail as seriously as we do, even delivering the mail during a typhoon: this BBC news story notes that "a postman was injured after wind overturned his van."

* * * * * * *

Regards the bridge collapse in Minneapolis -- the StarTrib had these quotes, which make it appear that the hot weather may indeed have been a contributing factor -- although the focus now seems to be mostly on a failure with the steel truss, which was designed as non-redundant (any one piece failing can cause the entire truss to fail):

"Things can happen with temperature, and with construction, or a lot of other confounding factors." --Catherine E. Wolfgram French, a civil engineering professor at the University of Minnesota.

"I would start at the foundations." --Michael Ramerth, a principal at MBJ Consulting Structural Engineers in Minneapolis.
* * * * * * *

Have a nice weekend, and don't forget to watch that tropical wave in the Caribbean and the East Pacific. :)

In other words, there is no tropical activity of note in the North Atlantic, so relax and enjoy the weekend.

Updated: 03:31 PM GMT am 03. August 2007

Permalink

Usagi Landfalls in Southern Japan

By: JeffMasters, 02:07 PM GMT am 02. August 2007

Guest Blogger Margie Kieper


Thursday afternoon update: Recon once again found no closed circulation in the tropical wave and associated disturbed weather in the eastern Caribbean. And Generalissimo Francisco Franco is still dead.

* * * * * * *

Usagi has made landfall in southern Japan in Kyushu, apparently at tropical storm strength or possibly Category 1 hurricane, for those familiar with the US scale -- or as a Severe Tropical Storm or weak Typhoon as classified by Japan Meteorological Agency (JMA). JMA's hourly observations from the Kyushu area and just north, show a low pressure of 973.5 mbar, highest sustained (10-min) winds of around 50 kt, and considerable rainfall: Nobeoka recorded 46.5 mm in one hour, and 183 mm total. The coast of the Kyushu Prefecture, including Miyazaki, Aburatsu, and Nobeoka, experienced the eyewall winds on the northwest side. Cooler water contributed to the rapid weakening.

Usagi is a large storm and its effects are still being felt as it travels over the islands of Japan.

The last vis image from last night showed the eye covered in cirrus; however microwave imagery continued to depict a large eye, with very little remaining convective banding, at the time the northwestern eyewall was over the Pacific coast of Kyushu:

Usagi vis


Usagi vis


Here's a photo of the waves at the shore in Japan prior to Usagi's arrival, taken by Typhoon Hunter:

Usagi arriving in Japan


* * * * * * *

There is still no sign of tropical development in satellite imagery of the much-publicized tropical wave in the eastern Caribbean. We will continue to report faithfully on this non-development, and hope to be reporting on the lack of development of this tropical wave throughout the 2007 season ("Breaking news just in...Generalissimo Francisco Franco is still dead"). However, to quote from the Atlantic TWO, "THE SYSTEM STILL HAS SOME POTENTIAL TO BECOME A TROPICAL DEPRESSION DURING THE NEXT DAY OR SO."

Recon is scheduled for both this area of disturbed weather and the one in the Gulf of Mexico. The Caribbean invest is being monitored by three AF Hurricane Hunters who deployed to St. Croix the day prior to the scheduled mission, but for the disturbance in the northeast GOM the Hurricane Hunters will not have to travel very far from their home base in Biloxi, MS.

For those wondering about the potential of the tropical wave following 99L, all tropical waves in the hurricane season have potential. The experts at NHC keep track of things like that, and they document this information in the TWO (Tropical Weather Outlook).

* * * * * * *

I see there are many questions regarding recon. The readings taken at flight level, if about about 300 meters, can be taken as surface windspeeds. Yesterday winds were light and variable (mostly 1 kt to 20 kt) and there was a broad center area of low pressure (a trough). The key readings required for a closed surface circulation will be westerly winds on the SW leg (winds coming from 240 to 300 degrees). My recommendation: don't bother reading the HDOBS because when the recon flight is completed, the results will be in today's afternoon Atlantic Tropical Weather Outlook.

* * * * * * *

Regarding the bridge collapse in Minneapolis -- which I can see outside the office window of my 25-flight-plus downtown location -- weather was likely a factor.

In the 1970s the newest way to build bridges was to incorporate steel encased in concrete. Concrete provided the compressive strength and steel the tensile strength. There are many, many bridges built using this technology, from about 40 years ago, that are still in use today. However bridges that are built today are built of concrete with the tensile support provided external to the concrete, via cabling, beams etc.

The problem is that the two materials contract and expand differently due to changes in temperature, with the steel changing more than the concrete. Over time this would create small stress fractures in the concrete, near the steel, and then over a longer time period this leads to greater fatigue fractures. Eventually the concrete has a catastropic failure and falls apart, and the remaining steel, which does not have significant compressive strength, collapses. This bridge was designed as one long open steel truss spanning the Mississippi, and I expect that the analysis will find that there was a failure of one of the concrete supports. There was a video showing part of the bridge at failure, which showed that the south side of the truss failed suddenly -- the side off-camera -- and that the shock through the truss caused a secondary failure on the other side of the bridge, in the truss, dropping it about 10 feet, for a couple of seconds, before the entire truss collapsed straight down.

Minneapolis has been in a continuing heat wave this week, and it seems each summer is hotter than the norm here, when you used to be able to count summer days above 85 degrees on one hand. We also get temperature extremes on the cold side in the wintertime; minus fifteen is not unusual. It has been hot the last week, and yesterday the high was 93 degrees in the late afternoon, and still 89 degrees at 6pm. This high of 93, followed daily highs of 90, 93, 96, 89, 89, 89, 91, and 91. When the temperature is this high, the steel would have expanded. At the time of the bridge failure, road crews were jackhammering on the roadbed above the truss; this was noted by someone who was on the bridge and was lucky enough to survive. This likely transmitted vibrations into the truss. Harmonic vibrations -- the right ones -- can trigger structural failure in already-fatigued structures. It was likely a combination of factors that provided the "last straw" that led to failure of some key part of the structure, most likely a concrete support. However there is also the possibility that the initial failure was in a key section of the truss that was fatigued by the heat and vibrations, but once the first failure occurred, the shock through the structure would have immediately triggered other failures, leading to a cascade that brought the entire truss down, with shock waves also bringing down the bridge structures on either side.

Updated: 07:33 PM GMT am 02. August 2007

Permalink

Japan Prepares for Usagi

By: JeffMasters, 01:29 PM GMT am 01. August 2007

Guest blogger Margie Kieper


Wednesday afternoon update (yet again): From the Atlantic TWO: "AN AIR FORCE RECONNAISSANCE PLANE INVESTIGATING THE AREA OF LOW PRESSURE AND A TROPICAL WAVE NEAR THE WINDWARD ISLANDS THIS AFTERNOON WAS NOT ABLE TO FIND A CLOSED CIRCULATION." And that is that -- for now. Tonight -- we'll Usagi (it sounds like it could be a verb!).


Wednesday afternoon update #2: There appears to be a broad low with no closed circulation on the latest recon obs for 99L -- which is back on the NRL web site (sigh). Winds are light and variable, and no westerly winds at all on the leg from the SW in to the center. Let's pretend we're in the cockpit of that flight, flying back and forth... Plaintive voice on the radio back to NHC: "Nothing here -- can we go home now?" :)

Will update on this in the evening.

Usagi is steadily weakening. This is what I didn't understand from the Monday and Tuesday forecasts -- how strengthening could be forecast for an area of ocean where the heat content drops off rapidly. Usagi will still be a significant typhoon when landfalling in Japan, but not the Cat 4 that it was earlier.

They did it again in the forecast discussion for TS Erick..."A WEAKENING BLOB OF DEEP CONVECTION." Sigh (again).

Here's a link to the UK forum that will be updating "Typhoon Hunter's" Usagi experience (gotta love those names, huh). Here's a link to Japan's radar.


Wednesday afternoon update: Initial recon obs for the disturbance just about to enter the eastern Caribbean appear to show support for a closed circulation at 1000 ft (HD OB 14 and 15 showed WSWly winds), with very light winds and a lowest pressure reading so far of 1002 mb. If a closed circulation is found, this disturbance will probably be labeled a tropical depression. Most recent water vapor imagery shows a moister environment in the eastern Caribbean, but convection is still disorganized.

The latest forecast from JTWC for Usagi shows it decreasing to 95 kt intensity just prior to making landfall in Japan. A microwave image from this morning shows the change in organization, with what appears to be an open eyewall to the southeast.

Usagi microwave


We'll hopefully be hearing some reports from "Typhoon Hunter," who headed for Japan a couple of days ago to intercept Usagi.

* * * * * * *

During the day in the West Pacific, Usagi maintained a classic presentation on visual satellite imagery:

Usagi vis


But at this same time, on microwave, the outer bands of convection took on a ring-like appearance, outside of the eyewall convection:

Usagi microwave


This convective pattern has since shifted, with strong convection remaining in the eyewall and the eye temperature remaining warm. JTWC has backed off intensifying in the forecast, and has maintained the 120 kt intensity, but still forecasts Usagi to hit southern Japan at the eastern tip of Kyushu with a maximum intensity of 115 kt, and JMA forecasting 80 kt (10-min wind).

* * * * * * *

***Horrors!*** I'm shocked -- shocked, I tell you. NHC has used the word "blob" in a forecast discussion on TS Erick: "...WITH THE CENTER OF CIRCULATION LOCATED NEAR THE EASTERN EDGE OF AN AMORPHOUS BLOB OF DEEP CONVECTION."

* * * * * * *

There continues to be no closed surface circulation associated with the area of disturbed weather approaching the eastern Caribbean, and invest 99L, which is now a tropical wave with some associated convection, and was dropped around 12Z this morning. The mid-morning Tropical Weather Outlook seems to want to keep it alive, however, even while noting, "ENVIRONMENTAL CONDITIONS DO NOT APPEAR ESPECIALLY FAVORABLE FOR DEVELOPMENT," so perhaps it will make a reappearance later today. Barbados weather obs, which can be followed here, will tell the story (pressure is currently rising).

More importantly, the TWO notes the disturbed weather in the GOM on the tail of the front that moved off the East Coast yesterday.

More updates in mid-afternoon.

* * * * * * *

For Jeffs early August hurricane outlook, which came out Sunday, link here.

Updated: 09:20 PM GMT am 01. August 2007

Permalink

About

Jeff co-founded the Weather Underground in 1995 while working on his Ph.D. He flew with the NOAA Hurricane Hunters from 1986-1990.

Local Weather

Overcast
69 °F
Bedeckt