The skies are turning brick red, with high clouds masking the bright sun. The sea is churning as mighty waves crash loudly on the windward coast. Not so long ago, these were the only signs of an approaching storm. Its strength, position and track could only be guessed at, and planning for its arrival was almost impossible.
These days, we watch the reports on television, and when the cable company turns off the satellite dish, we tune in to the local radio station or log on to our favourite hurricane tracking website. We know it’s getting close when the island powers down; if we have a generator, we can track the hurricane until it’s about to hit. Then there’s nothing to do but find somewhere safe to ride out the storm.
The modern science of hurricane forecasting has become a lifeline for the Caribbean during the peak months of the Atlantic hurricane season, and that is when the National Hurricane Center (NHC) in Miami, Florida, becomes the focus of attention. Tucked away in a bunker-like building at the back of Florida International University’s campus just west of Miami, it is home to some of the world’s top hurricane experts. These are the people who take the pulse of all tropical storm activity that passes through the Caribbean. It is in this building that computer models, aircraft observations, phone calls and teletypes all come into the hands of the six hurricane specialists who churn out the anxiously awaited forecasts. For the Caribbean, the information is crucial.
The NHC is the only official warning agency for cyclone activity in the Atlantic, Caribbean and Gulf of Mexico as designated by the World Meteorological Organization (a UN body). The meteorologists here release the official forecast track and information, while each Caribbean country that lies in the path of an incoming storm is responsible for issuing its own watches and warnings.
At first glance, the operations centre looks like a television newsroom. Computer monitors line the desks in the open “fish bowl” area where a floor-to-ceiling map of the world provides the backdrop. A dozen or so names are posted on the side of the map, all waiting to be assigned to that season’s storms, to name the hurricanes that may dissipate or destroy in a matter of hours. Across the cavernous room is a media “pen” where camera crews from all over the globe jostle for space during the height of hurricane season. When hurricanes threaten to make landfall in the US, this place turns into a media zoo.
With billions of dollars and hundreds of thousands of lives at stake, hurricanes have become the big news events of the summer season. Just how accurate the forecasts can be depends heavily on the hurricane specialists who man this place 24 hours a day, seven days a week. They divide their time between what is called the “Atlantic Desk” and the “Pacific Desk”. From June through November, it is the Atlantic Desk that keeps them hopping.
Stacy Stewart has 16 years’ experience in tropical cyclone forecasting, the last three of which have been spent monitoring the Caribbean region. When asked what it’s like at the height of hurricane season, Stewart takes a deep breath: “To put it bluntly — intense. It can get very intense. Very trying. You really have to watch yourself . . . it’s very easy to make a mistake if you are not watching yourself.”
Just how busy does the Caribbean keep Stewart and his colleagues? “The islands lie in Hurricane Alley,” he says. “And if you live in an alley eventually a car is going to run you over.” But where you are in the Caribbean will make a big difference to the chances of being slammed by Mother Nature. After a storm has gathered strength moving across the Atlantic, it generally travels through the Caribbean in a westerly to north-westerly direction — if it hits the Caribbean at all. “The north-east Caribbean islands are the ones that get pummelled the most,” says Stewart. “So anywhere north of about Martinique over to the island of Hispaniola, those tend to be the most vulnerable areas for significant hurricane strikes.”
The further south you go, the less chance there is of a hit. This is because thunderstorms in the eastern Atlantic need the earth’s rotation to induce the spin factor required to form a cyclone. On the equator, the spin is effectively zero: you have to be 10º N in the Atlantic for there to be sufficient rotation to start a storm. That is why Trinidad is unlikely to get hit hard by a hurricane, and in Guyana there’s virtually no chance at all.
Hurricanes usually have a forward motion of 10 to 20 mph, and the front-right quadrant is always the most intense. In the eastern Caribbean, this means that any island just to the north of the eye will get the strongest winds and heaviest rain, and its windward coast will get the worst battering.
The forecasters this year predict that the Atlantic and Caribbean Basin will see 11 named storms, six of which will become hurricanes. They also say that the Atlantic-Caribbean basin has moved into an era of heightened hurricane activity. Six out of the last seven years have seen an above average number of storms. The five-year period between 1995 and 1999 was the record breaker: 65 named storms and 41 hurricanes.
As Stewart walks through the Ops Center, he picks up the phone to broadcast the forecast of the first tropical cyclone in the Pacific. Hurricane Alma is about 700 miles off the coast of Cabo San Lucas, Mexico, and they are keeping a close watch on her track. When it gets busy in the summer months, Stewart and his fellow forecasters spend a lot of time talking to meteorologists in the Caribbean. “They pass information along to us and we include it in our forecast.” But it is the computer models that give the NHC their primary guidance. “I would not want to try to put out a forecast without the models,” says Stewart.
There are several different models that predict where and when the storms will hit. “Our responsibility, as forecasters, is to look at all the available models and compare them.” If all the models give the same answer then the forecast will be pretty reliable. It is when the models give different answers that the forecasters earn their money. “We have to know when to jump off one and start going with another model, and it’s not always easy to make that transition.”
The forecasters prefer to rely on science more than gut instinct. “We don’t forecast based on hunches, but on average there’s about one out of 10 storms where the models just don’t perform very well at all. Then you have to rely on your experience — you stand back and look more at the big picture.”
One storm that didn’t co-operate with the scientific models was Hurricane Lenny which swept through the eastern Caribbean late in the 1999 hurricane season. The forecasters nicknamed it “wrong-way Lenny,” as it came from the west and hammered the western sides of islands that usually escape the brunt of the storms. It was a one-in-a-hundred-year event, but that made little difference to fishermen in the eastern Caribbean, whose boats were still in the water when the storm hit.
The problem is that there is a vast amount of ocean and atmosphere between Africa and the eastern Caribbean where the models have to make up data, rather than using real information. The occasional passing ship may send in a report, and satellites beam back some information, but for the models and forecasters, it’s usually a big black hole.
When a storm closes to within a few hundred miles of landfall in the Caribbean, the forecasters produce the final tool in their armoury — reconnaissance aircraft. A group of pilots, commonly referred to as “hurricane hunters”, are sent by the forecasters at NHC into the eye of the storm. They report back what is happening in the atmosphere around and within the hurricane, and that information feeds directly into the forecasts. The National Oceanographic and Atmospheric Association (NOAA) has two aircraft that are used for this data gathering. The US Air Force Reserve flies a third plane, a modified C-130, into the storms. The real-time data that is fed from the planes via satellite back to the NHC is, of course, vital, but for some of the Caribbean islands, the information can come too late. “Reconnaissance aircraft have to stay around the hurricane for many hours, so their range is fairly limited. Often storms have to be within 300 miles or so of the Leeward Islands before aircraft are dispatched for recon work,” says Stewart.
For storms crossing the Atlantic, constant improvements in the models and the increasing amount of data available to forecasters means that their accuracy for landfall three days in advance has increased four-fold in the past 30 years. However, for the small islands of the eastern Caribbean, a 130-mile-error can make the difference between a direct hit and a windy day. And predicting the strengthening or weakening trends of storm2™s remains a challenge.
Forecasters at NHC have the most trouble with storms early and late in the season which seed rapidly and develop sporadically within the Caribbean Sea. “Every once in a while we get a runaway storm where the bottom just falls out and we go from a tropical depression to a category 4 hurricane in a day and a half, and that’s kind of scary, especially if it’s happening right off your coast. That’s what we don’t like to see. When something develops and develops rapidly right in your back yard, that creates big problems,” says Stewart.
A much less sophisticated operation is the contact between the NHC and ham radio operators throughout the Caribbean. Some of these radio operators have meteorological instruments supplied by NHC. There has been an amateur radio station at the NHC for the last 22 years and Caribbean “hams” have supplied forecasters with hundreds of supplemental surface weather reports. The hams are also a vital link in getting out calls for help through their co-ordination with the Red Cross and local emergency managers.
Although a lot of the hurricane research at NHC is aimed at preventing losses in the US mainland, the hurricane forecasters and researchers are well aware of the vulnerability of the small Caribbean islands. For most of the islands, wind and rain are the main problems. One key fact, more than any other, should be remembered by islanders, says Stewart: “Forecasts and category numbers are given for 30 feet above sea level. The higher you live, the higher the winds — so a forecast category 2 hurricane hitting an island with lots of mountains will be at least a category 3 hurricane for most residents.” As windspeed increases, damage increases much more quickly, so if the wind is a little stronger than forecast, then the damage can be much worse.
Every island can recall in vivid detail what it was like to live though their worst storm. Most of the devasting hurricanes in the Caribbean have occurred during the middle part of September — the real peak of the action for the Caribbean.
Jamaica remembers Hurricane Gilbert which devastated the island on 12 September, 1988. The eye of Gilbert holds the record as the lowest pressure ever recorded in the western hemisphere, and is regarded as the strongest Atlantic storm on record. Without a hit in the previous three decades, Jamaica was relatively unprepared, and damage was severe.
Montserrat was flattened by a direct hit from the category 4 Hugo on 17 September 1989, a storm which continued across the USVI and Puerto Rico and ended up slamming into the US East Coast in South Carolina. This storm was the second most costly ever, causing US$7 billion in damage in US territories alone.
An early August storm, Hurricane Allen, in 1980, holds the windspeed record for Atlantic hurricanes, with maximum sustained winds of 165 knots (190 mph). Allen also holds the record as the longest hurricane at category 5, spending almost three days with sustained winds over 155 mph. Allen clipped Barbados and St Lucia at category 4 strength, but never quite made landfall at category 5. The only hurricane to do that in the Caribbean (excluding the Bahamas) was the famous early September 1928 hurricane, which hit Puerto Rico at category 5, devastating Guadeloupe and Montserrat on the way.
The most deadly hurricane is one for which few records exist. Known as The Great Hurricane of 1780, it devastated many islands of the eastern Caribbean before heading across the Caribbean Sea. In just a few days an estimated 22,000 people were killed.
For Stewart and his colleagues, there is big motivation in reducing the devastation and loss of life. “We don’t do the job because we like to see hurricanes. We know they are going to develop. Its not a question of ‘if’, but ‘when’?’ So every year we know we are going to have at least some hurricanes out there. So it’s a job that we enjoy doing, but not from the standpoint of ‘oh goody, here comes another hurricane’. It’s a very challenging job, and its very rewarding when you have put out a good forecast and you know you’ve helped people.”
To the islanders of the Caribbean, the NHC is a beacon of light. Maybe this year the Caribbean will be spared, but if it isn’t, then the people in “Hurricane Alley” will be relying on Stewart and his colleagues to get it right.
You can find the National Hurricane Center and all the information it generates on the worldwide web at www.nhc.noaa.gov.
THE MAKING OF A CARIBBEAN HURRICANE
Atlantic hurricanes form off the western coast of Africa. Disturbances in the atmosphere created over the coast of West Africa move out over the warm waters of the Atlantic. The Cape Verde Islands, sitting at 15°N, 500 miles off the African coast, are used as a reference point as they are the only land masses before the stretch of thousands of miles of open ocean to the Caribbean. “If it develops south of the Cape Verde Islands, there’s at least a 50/50 chance that it will pass through the Caribbean,” says Stacy Stewart of the National Hurricane Center.
The key, says Stewart, is “organised and persistent thunderstorms, because without the thunderstorms you will never get a tropical cyclone forming.” The Atlantic Ocean is warm year-round, but the thunderstorms don’t usually start to get organised until late July. Thunderstorms act as an agent for heat exchange from warmer ocean to the cooler atmosphere. This upward heat exchange drives air upwards, causing lower pressure at the surface. Typical low pressure systems start as open tropical waves. If thunderstorms persist, then the wave will close, forming a low pressure system, and then a tropical depression.
The warning sign for hurricane-watchers, Stewart says, is a drop in pressure. Once that happens they know they are on their way to a storm. “As the pressure continues to drop, it draws air in from greater distances, and that causes more thunderstorms. The more thunderstorms you have, the more the pressure drops. Now the winds increase as the pressure drops and you get a tropical storm.”
Once a system has organised and it has sustained windspeeds of greater than 39 mph, it is classified as a Tropical Storm, and given a pre-determined name. If the system develops further and winds reach a sustained speed (measured as an average over one minute) of 74 mph, then it becomes a Hurricane. The Saffir-Simpson scale is used to classify hurricanes, with 1 being the weakest and 5 the strongest (winds greater than 155 mph).
The satellite images of hurricanes which have become so familiar in recent years have removed a lot of the mystique aournd the killer storms. The most recognisable feature is the eye of the hurricane, which marks the centre of rotation and the point of lowest pressure. Around the eye is the eye wall, where the most intense thunderstorms form. The remainder of the central core of a tropical cyclone is known as the central dense overcast, where winds gradually diminish outwards. Outside of this, the cloud cover splits into bands of intense showers and gusty winds. Systems vary greatly in size, but most are between 100 and 300 miles across.