Rainfall in Sahel region of W Africa & position of ITCZ near Africa Coast - Start of the rainy season in Africa appears to be about normal, though there are some variations from area to area in the Sahel region.

El Nino / La Nina conditions - This basically refers to the impact of sea surface temperatures in the Eastern Pacific.  Let's simplify the impact of this event for illustrative purposes...Think of it this way...given a certain amount of energy, if more of it is in the Pacific, less of it will be available to fuel Tropical development in the Atlantic.  While a slight ElNino inhibited early-season Atlantic Hurricane development, the weak ElNino seems to be moderating.  Models are not in agreement on the latter part of the season, but lean towards a slight cool Pacific LaNina event, which would have a slight enhancing effect on Tropical development in the Atlantic later in the season.

Atmospheric dynamics, including surface pressure, upper-air pressure, surface winds, upper-air winds - Once we have warm water and the right position of the ITCZ, these are the factors which directly affect Hurricane development.  While these factors change day-to-day and season-to-season, it is interesting and vital to note the longer-term variations or cycles for these factors.  Each one is in a long-term cycle which is favorable to support Tropical development in the Atlantic basin.  Taken together, these may explain some of the additional Tropical activity we've had in the past 10 years.  Once again this year, surface pressures in the Caribbean are low (so storms start), upper-atmospheric pressures are high (necessary for outflow and storm "feeding"), surface Trade Winds are light and upper-air winds are light (both lead to less shear and more-intact systems).  These factors seem to support Tropical activity at this tim, and this is forecast to continue for the rest of the Tropical season.

General Discussion of expert predictions:

Media Release July 15, 2003.

Contact: David Jones (284) 494-7559

The Latest Predictions of Tropical Cyclone Activity in the 2003 Hurricane Season have recently been increased!

As we begin to approach the more active period of the 2003 Hurricane Season, it is time to take a look at the latest forecasts for 2003. Six years ago, only Prof. Gray of the Dept. of Atmospheric Science of Colorado State University(CSU) was issuing longer term predictions of the likely level of tropical cyclone activity in the Atlantic, Caribbean and the Gulf of Mexico. Since that time, he has been joined by the US National Oceanic and Atmospheric Administration(NOAA) and recently by the Hazard Research Center of University College, London(UCL). It is therefore interesting to compare their latest forecasts for the 2003 Hurricane Season, as shown by the following table:

Key factors behind tropical cyclone formation in the Tropical Atlantic and the Caribbean are:

· The amount of rain falling in the Sahel Region of Sub-Saharan Africa.

· Sea Surface Temperatures (SST's) in the Tropical Atlantic, Caribbean and the Gulf of Mexico.

· Atmospheric pressures in the Tropical Atlantic

· The strength and latitude of the Inter-tropical Convergence Zone(ITCZ) in the Tropical Atlantic. The further N the ITCZ is in the tropical Atlantic, the more it can induce rotation in tropical waves through the "Coriolis" effect.

· The strength and direction of the mid- and upper level jet stream which is influenced by the "El Nino/La Nina" oscillation in the E. Pacific. During "El Nino" years the jet stream exercises an inhibiting effect; the reverse is true in "La Nina" years.

Taking each of these features in turn:

· Independent research conducted by The Caribbean Weather Center indicates that the amount of rain falling in the Sahel Region of Africa during the critical peak of the season may well be above the long term average. If so, this is likely to result in above-average tropical cyclone activity. However, CSU, UCL and NOAA suggest this feature no longer has the importance previously given to it and have discounted it. The Weather Center finds this odd as the amount of rainfall determines the level of convection activity in tropical waves. Current indications are that rainfall in this region over the last month or so has been more or less at a normal level.

· Both NOAA and CSU are predicting that SST's in the Tropical Atlantic are likely to be above average whereas UCL is predicting average temperatures. SST's currently remain cooler than normal in the western part of the tropical Atlantic.

· As regards atmospheric pressures, all three institutions are predicting lower than normal atmospheric pressures in the Tropical Atlantic which will tend to enhance tropical cyclone formation. At the moment atmospheric pressures are higher than normal with the result that current tropical wave/cyclone activity is either being inhibited or taking place further S in the tropical Atlantic.

· None of the institutions give any predictions as to the likely strength and latitude of the ITCZ. Weaker high pressure in the Tropical Atlantic will allow the ITCZ to move further N and thus increasing the effect that the spinning of the Earth ("Coriolis Effect) has on tropical cyclone formation. At the moment, the ITCZ is hovering around 9N-10N compared with 5N-8N a month ago.

· A predominantly easterly-flowing upper level jet stream in "El Nino" years tends to inhibit the development of intense tropical systems by wind-shearing ("slicing off") the tops of convection clouds. The latest indications are that "La Nina" in the E. Pacific is fairly well established and that the protective shield provided by the upper level stream has all but disappeared.

In summary, the latest key factors tends to confirm that 2003 will experience an above-average level of tropical cyclone activity. Already we have seen 3 named storms. OF GREATER CONCERN IS THAT TWO OF THE INSTITUTIONS INDICATE A DISTINCT PROBABILITY OF TWO TROPICAL STORMS AND ONE HURRICANE MAKING LANDFALL IN THE EASTERN CARIBBEAN.

The Caribbean Weather Center does not attempt to make longer-term forecasts of tropical cyclone activity, but monitors on a daily or weekly basis during the Hurricane Season any changes in key formation factors. The one feature that we feel fairly sure about is that there will be a tendency to more late season activity, and the last few years have provided ample evidence of this. Every Monday as part of our regular Weather Updates on ZBVI Radio we issue a tropical weather outlook for the following 7days.

JUST REMEMBER, WHATEVER THE NUMBERS….IT ONLY TAKES ONE!

The 2002 Hurricane Season saw more Named Storms than predicted but suppressed Hurricane and Intense Hurricane activity!

The 2002 Hurricane Season, which ended officially on November 30, 2002, saw more named storms than predicted but less than average hurricane and intense hurricane activity. For the second year in a row, no hurricanes were experienced in the E Caribbean and only 3 named storms have passed through the region in the past 2 years. Despite this low level of activity, the initial predictions for the 2003 Hurricane Season call for a heightened level of activity, as shown by the following table:

The principal reasons for the lower level of activity in 2002 were as follows:

· Drought or semi-drought conditions in the Sahel region of sub-Saharan Africa where tropical waves are spawned.

· Sea surface temperatures (SST’s) in the tropical Atlantic cooler than predicted. All tropical cyclones need access to warm, moist sea water.

· Atmospheric pressures over Africa and the tropical Atlantic were above normal inhibiting tropical wave development and forcing those that did form much further south in the tropical Atlantic where they are less likely to develop.

· The ITCZ or the Inter-tropical Convergence Zone (where the NE trade winds of the N hemisphere meet the SE trade winds from the S hemisphere) stayed mainly S of 10N for much of the season. As a result, there was very little convergence of the convection activity associated with the ITCZ and that associated with tropical waves.

· "El Nino", the cyclical warming in the E Pacific which tends to have a inhibiting effect on tropical cyclone formation, came into play at the statistical height of the hurricane season in mid-September.

Taking a look now at the initial predictions for the 2003 Hurricane Season, the main reasons for an expected increase in activity, according to Prof. Gray of Colorado State University, is for warmer than usual SST’s in the tropical Atlantic next year as well as a reversal of the ‘El Nino" effect. According to Gray, the period from 1995-2002 comprised the most active hurricane seasons on record and he believes that we are in a multi-decade period of increased tropical cyclone activity, such as that occurred in the 1940’s and 1950’s.

Just remember, whatever the numbers, it only takes one!

August 9, 200

The 2001 Hurricane Season was kind to the E Caribbean but the W Caribbean did not fare so well!

The 2001 Hurricane Season saw well above the average level of activity with 15 named storms of which 9 became hurricanes of which 4 were intense hurricanes. The final forecast was for 12 named storms of which 7 would become hurricanes of which 3 would be intense. The long-term annual average is 9.3 named storms, 5.8 hurricanes and 2.2 intense hurricanes.

According to Prof. William Gray of Colorado State University, this was an unusual year in that most of the tropical cyclone activity occurred September through November. The first hurricane did not form until Sept. 8 - the latest-forming first hurricane since 1984. Other unusual occurrences included one of the most active periods during the last 50 years during October and November, with four hurricanes and two major hurricanes. In addition, two hurricanes happened simultaneously in November, the first time that has happened since 1932. Finally, 1995-2001 are the seven most active consecutive years on record, with 93 named storms, 57 hurricanes and 27 major hurricanes.

The E Caribbean came off fairly lightly and did not experience any hurricanes this year. The only tropical conditions experienced in the Virgin Islands were when TS Dean was forming just to the NW of the islands in the afternoon of August 22nd. The Leewards and Windwards also fared well with only 3 tropical storms (Chantal, Iris and Jerry) passing over the islands but causing no significant damage.

On the other hand, the W Caribbean did not fare so well with two category IV destructive hurricanes, Iris and Michelle.

As Iris passed S of Jamaica, it strengthened into a compact category IV destructive hurricane with 125kt(140mph) winds as it made landfall on the S coast of Belize in the evening of October 8th. Iris destroyed the entire banana crop and caused heavy flooding and major infrastructure damage. Belize had only just finished rebuilding last year’s damage from Keith, also a category IV hurricane. Iris was deadliest hurricane of the 2001 Hurricane taking more than 30 lives.

Michelle formed off the E coast of Nicaragua. As it headed NW towards the Yucatan Channel and strengthened, it brought very heavy flooding to that country and Honduras. On November 3, Michelle, now a destructive category IV hurricane, changed course and headed towards W Cuba, Making landfall on November 4 with winds of 115kts(130mph). At least 17 lives lost were attributed to Michelle.

Even though the 2001 Hurricane Season ended on November 30, at the time of writing the 15th named storm of the season, Olga, is one of the few storms that have survived into December. It is currently meandering midway between the Bahamas and Bermuda as a tropical storm.

Altogether, a most unusual hurricane season!

The 2000 Hurricane Season in the Caribbean was much quieter than predicted!

As the 2000 Hurricane season draws to a close, it is time to take a look at how the Caribbean was effected. The final predictions for this year’s Atlantic Basin hurricane season were 11 named storms of which 7 would become hurricanes of which 3 would become intense. In other words, an above average season. To date (late November) we have seen 14 named storms of which 8 became hurricanes of which 3 were intense hurricanes at some stage in their cycle, shown by the following table:

So it would seem as though this year’s predictions were very much on the ball. However, that is not definitely not the case as regards the Caribbean where, with one exception, we have had a very light hurricane season. Only 6 named storms threatened or were close to threatening the Caribbean this year. They were Chris, Debby, Ernesto, Helene, Joyce and Keith.

Let us take a brief look at each of these systems and their impact. Well Chris fizzled out as a minimal tropical depression just east of Antigua. Debby on the other hand tracked through the Leewards and the NE Caribbean as a minimal category I hurricane or more like an English Channel gale. Debby never developed an eye and brought more rain than winds. Ernesto tracked well to the NE of the Leewards and never really threatened the Island Chain. Helene also fizzled out east of Antigua and only regenerated in the NW Caribbean. Joyce on the other hand tracked unusually far south and passed between Grenada and Trinidad but once again as a tropical depression. On the other Keith, a Category IV Hurricane caused extensive flooding in Belize where some 35 inches of rain fell over a 3-day period.

What were some of the reasons for the lighter activity in the Caribbean? Well, for the first part of the season atmospheric pressures in the tropical Atlantic were unusually high. As a result, tropical waves were further south than usual and therefore less likely to develop. Another factor was Sea Surface Temperatures (SST’s) in the Tropical Atlantic. These were below normal right through September with the result that tropical waves were starved of sufficiently warm moist sea water. On the other hand, very warm SST’s in the NW Caribbean contributed to the intensification of Keith. Finally and despite "La Nina", at times there was a fairly strong upper level easterly-flowing jet stream which inhibits convection activity and therefore tropical storm development.

So, despite dire predictions, once again the Caribbean has come through once again relatively unscathed!

In the last 110 years that detailed records have been kept, the Caribbean has never experienced a tropical cyclone like Lenny. There have been two late-season tropical storms, one in 1909 and tropical storm Klaus in 1984, that followed a northeastwards path across the Caribbean. What made Lenny unique was the fact that not only was it a late-season hurricane that tracked from west to east, but also that it almost reached Catastrophic Category 5 status on Wednesday afternoon, November 17th, 1999when it was approximately 50 miles due S of Road Town, Tortola and battering St. Croix. While the Caribbean Weather Center in downtown Road recorded a maximum gust of only 55mph, at Cable & Wireless’ facility approximately 1500ft up at Chalwell two maximum gusts of 185mph and sustained winds well in excess of 100mph were reported for more than 3 hours from midday onwards.

Let us take a closer look at what caused Lenny to form when it did, what caused it to track from west to east and what allowed it to develop into a destructive category IV hurricane. Well the blame can, to some extent, be placed on "La Nina". The world climate is currently in a strong "La Nina" or cool period. During "La Nina" years, atmospheric pressure patterns change the World and generally we have seen unusually low pressure in the tropical Atlantic and the Caribbean during the latter part of the hurricane season. In the western Caribbean, prior to Lenny there had been unusually low atmospheric pressure some three weeks before Lenny was spawned. This allowed a large area of convection activity (thunderstorms) to develop. All that was needed was a tropical wave to come along to kick-start it and thus, Lenny was born on November 13th, 1999 southwest of Jamaica. However, already at that time, the winter west to east weather patterns were already bringing cold fronts to the NW Caribbean. Lenny, a late season hurricane, was influenced by these weather patterns and began moving east. At the same time, a persistent low pressure trough, first to the NE of the Virgin Islands then east of the Leewards helped to draw Lenny east-northeastwards across the Caribbean. There was no upper wind shear and sufficient warm, moist sea water allowed Lenny to rapidly develop into a dangerous hurricane.

With a normal east to west moving hurricanes, the strongest winds are usually experienced in the NW segment of the system. With a "reverse" hurricane like Lenny, the strongest winds were in the SE quadrant and since Lenny tracked south of the Virgin Islands, the BVI was spared the full wrath of Lenny. When Lenny was some 80 miles southeast of the BVI, the trough that had been to the east of the Leewards had closed and Lenny met instead weak high pressure which caused it to stall and meander back and forth between St. Maarten and St. Bart’s. It was this meandering over a 24 hour period that brought extensive destruction to St. Maarten and Anguilla.

However, destruction was not limited to the NE Caribbean. As Lenny moved from west to east, the "reverse" wind circulation generated an unusually large NW swell which caused widespread damage to the western shores of the Leewards and Windwards Islands.

The Caribbean Weather Center’s forecasting model predicted the possibility of Lenny developing into a major hurricane, tracking from west to east and developing a large NW swell for more than a week ahead. I began putting out warnings on ZBVI Radio, the other radio stations and nets that I provide weather forecasts for as well as on the Weather Center’s Website and via the Internet. Perhaps because of the uniqueness of this system, sufficient heed was not given to these warnings by some islands. Thankfully, the BVI was fully prepared.

A 1998 analysis of Eastern Caribbean hurricane tracks for the years 1950-1997 provides some interesting landfall statistics !

In a recent articles, I wrote about the latest forecast from Colorado State University for the 1998 hurricane season and pointed out that the forecast envisaged an increased chance of the classic "Cape Verde" hurricanes forming this year in the tropical Atlantic. I also mentioned that article that a long term analysis of tracks of hurricanes forming in this area shows that their tracks tend to lie between 15ºN and 25ºN at 60ºW and thus closely approach or pass over the northern part of the Island Chain and the NE Caribbean. "Cape Verde" hurricanes tend to be the more intense type of hurricane and thus there is an increased risk that we shall see a major hurricane in the Eastern Caribbean this year.

I was recently asked if I could be more specific as to likely landfall of classic hurricanes in the E Caribbean. So I decided to analyze the tracks of all the hurricanes that passed between 10ºN and 19ºN at 60ºW during the period from 1950 to 1997. All together there were some 20 hurricanes during that period that met the criteria. I then analyzed each system as to area of formation and the time of formation with "Early" being prior to August 15th and "Late" after September 30th. If it formed east of 35ºW then I considered it to be the classic "Cape Verde" type. I further analyzed the tracks to see which of these systems approached within 75 miles of the more popular anchorages from Puerto Rico to Trinidad. Here’s the result:

Perhaps the most surprising result is that neither the southern Windwards nor Trinidad is immune from hurricanes despite an apparent belief to the contrary. What is clear, however, is that the classic "Cape Verde" hurricane is more likely to affect the more northern islands in the Island Chain. Having said that, an analysis of the tracks of all of the classic type of hurricanes for the past 47 years indicates that roughly two-thirds crossed the 60ºW meridian between 18ºN and 25ºN rather than south of it.