The Heart of the Hurricane Season

28 08 2012

By Dale C. S. Destin

Every year, persons across the Caribbean, Central America and the US Gulf and Atlantic Coasts pray for this period to pass quickly with out a hit – the heart of the hurricane season, which covers the period August to October (ASO). For the period 1851 to 2011, Antigua has had 93 named storms and 44 hurricane giving and average per year of 0.6 named storm and 0.2 hurricane for ASO. For Antigua, the heart of the hurricane season is August to September, which accounts for 79% (Aug 34%, Sep 45%) of all storms to affect the island. ( ). This period has had 83 named storms and 41 hurricane for an average per year of 0.5 hurricane and 0.2 hurricane i.e. one storm every other year and one hurricane every five years.

August 21 and September 3 are very peculiar days for Antigua. These are the two peak days for Antigua for the hurricane season with a record of 7 named storms each to have affected the island on those days from 1851 to 2011. September 3 has produced three hurricanes and August 21 has produced four. ( ) Two notorious tropical cyclones that have struck on August 21 were Category 3 Unnamed Hurricane of 1871 and Category 3 Hurricane Baker of 1950.

On average, ASO has 8 named storms and 5 hurricanes per year. Historically the period has had 1237 named storms of which 806 became hurricanes – 1851 to 2011 – with August contributing 362 named storms and 230 hurricanes; 80% of all storms (Aug  23%, Sep 37%, Oct 20%) form during the period ASO. This year has been no ordinary year contrary to initial forecasts; so far for August, there have been six named storms and three hurricanes, when the month averages two named storms and one hurricane. The record for named storms for August is eight, observed in 2004. The forecast for the overall season has moved from near – below normal to above normal/active season with a consensus of 14 named storms, 6 hurricanes and 3 becoming major hurricanes; already there has been 10 named storms  and 4 hurricanes. ( looks like we will actually see more than 14 when the season ends November 30. Currently, ENSO Positive conditions (El Nino) is trying to squelch of shut down the hurricane season as is the norm; however, the tropical North Atlantic is warming and other factors such as negative North Atlantic Oscillation (NOA) and Arctic Oscillation are in favour of further warming. Of course, there are other factors to consider but these are some of the main drivers of the hurricane season. What do you think will happen? How many more named storms will we see before the season ends?

Record Breaking Start to the Hurricane Season

6 07 2012

By Dale C. S. Destin

Records continue to be broken or equalled this hurricane season, which is supposed to be near normal. Notwithstanding the near normal forecast, the season has had a very frantic start.  Even before the official start to the season on June 1, there were two preseason storms in May. Historically May averages one named storm every 10 years, so to have two storms in May is extremely unusual. As a matter of fact, the only other time this happened was in 1887, well over a hundred years ago. Further, this was only the third time on record two storms formed before the official start of the hurricane season – 1887, 1908 and now 2012.

With the formation of Debby, this makes it the first time on record that four storms formed before the end of June. Debby became the earliest fourth storm in history, beating Dennis of 2005 by 11 days. On average, the fourth named storm forms by August 23. June has never had more than three storms in a given season and this is the 12th time two or more storms have formed in a year in June.

The forecast for the hurricane season is still for near normal activity; however, with a quarter of the number of storms already in the history books, it is quite likely many of the groups who produce forecasts for the season must be scratching their heads and wondering what the hell may have gone wrong with the Atlantic. I now believe that this season will be marginally above normal and the next round of forecasts will reflect this. With four storms already gone, the average season producing eight named storms over the period August to October and the possibility of El Nino, an inhibiting factor, but forecast to develop very late in the hurricane season, I would say an above normal season is fast becoming the most likely outcome.

The x-factor continues to be whether there will be an El Nino episode and when. I look forward to the next round of seasonal forecasts, which are due early August. Will more records be equalled or tumble this year? Follow us as we keep you up to date with all the latest happenings of the Atlantic Hurricane Season.

Tropical Cyclones up to the end of June:

Name                         Dates                        Max Wind (MPH)

TS Alberto             19-22 May                         60

TS Beryl                 26-30 May                        70

H  Chris                19-22 Jun                          75

TS Debby               23-27 Jun                         60


1 06 2012

By Dale C. S. Destin


The general consensus among tropical cyclone experts is for a near normal Atlantic Hurricane Season for 2012. The consensus forecast calls for 12 named storms, 6 hurricanes and 2 major hurricanes. A normal season averages of 11 named storms, 6 hurricanes and 3 major hurricanes (See table 1).  The science behind the outlook is rooted in the analysis and prediction of current and future global climate patterns as compared to previous seasons with similar conditions. For this season, the experts are citing a neutral to possibly warm Pacific Ocean (possibly an El Nino) and a neutral to cool Atlantic Ocean as the reason for a near normal season prediction. A warm Pacific Ocean causes hostile upper level atmospheric conditions for tropical cyclones and a cool Atlantic Ocean reduces the energy available for cyclone formation and also causes hostile atmospheric conditions for tropical cyclones (above normal trade winds, surface pressure and vertical wind shear).

Forecast (Groups) Source Forecast Date Named Storm Hurricane Major Hurricane
NOAA May 24, 2012 9 – 15 4 – 8 1 – 3
CSU Jun 1, 2012 13* 5 2
Apr 4, 2012 10 4 2
TSR May 23, 2012 13 6 3
Apr 12, 2012 13 6 3
UKMO May 24, 2012 7 – 13
NCSU 2012 7 – 10 4 – 7 1 – 3
Consensus^ Jun 1, 2012 12 6 2
61-yr Antigua Climatology1 1950 – 2010 0.7 0.4 0.2
61-yr Atlantic Climatology 1950 – 2010 10.9 6.2 2.7
30-yr Atlantic Climatology 1981 – 2010 12.1 6.4 2.7
Table 1: The Atlantic Hurricane Season Forecast – 2012. ^The Consensus forecast is themean of the June and May forecasts issued by the various groups. The Season averages (1950 – 2010): 10.9 named storms, including 6.2 hurricanes of which 2.7 intense hurricanes. 1Storms passing within 105 nautical miles of Antigua. NOAA – National Oceanic Atmospheric Administration. CSU – Colorado State University.  TSR – Tropical Storm Risk. UKMO – United Kingdom Met Office. NCSU – North Carolina State University. Consensus – The average of all the forecasts by the Met Service. *Includes the two pre-season storms – Alberto and Beryl; for the remainder of season 11 storms are predicted.

What does this mean for Antigua and Barbuda?

 Although there have been great advancements in the science of tropical cyclone (depression, tropical storm and hurricane), the science has not yet reached the stage where accurate predictions can be made of how many cyclones will form in a given year. Also, the science cannot accurately predict when and where these systems will move or make landfall months in advance. The details of the large-scale weather patterns that direct the path of these cyclones cannot be predicted more than a few days into the future.

There is no clear correlation between the number of tropical cyclone in the Atlantic yearly and the number that affects Antigua and Barbuda. As well, there is no detected trend in the percentages of Atlantic Tropical Cyclones that reach our islands yearly. However, for a near normal season, the climatology (1950 – 2010) suggests the following for Antigua and Barbuda (the 95% confidence interval in bracket):

  • 58% probability or (41 to 74%) probability of at least one named storm
  • 32% probability or (19 to 50%) probability of at least one hurricane
  • 19% probability or (9 to 36%) probability of more than one named storm
  • 10% probability or (3 to 25%) probability of more than one hurricane
  • 0 – 3 named storms, including:
  • 0 – 2 hurricanes

Over all, there is a 46% (33 to 59%) probability of at least one named storm per season, 28% (17 to 41%) probability of at least one hurricane, 16% (8 to 28%) probability of more than one named storm and an 8% (3 to 18%) probability of more than one hurricane. The most likely months for tropical cyclones to affect Antigua and Barbuda are August and September. Of all cyclones to impact Antigua, 79% (63 to 90%) occurs over the period August – September. For a near normal season, this number decreases slightly to 73% (54 to 86%) with the average impact date of August 30 (± 20 days). By climatology, at least one named storm impacts Antigua and Barbuda every two to three years on average and at least one hurricane every two to six years. See graph 1 below for a distribution of Antiguan Tropical Cyclones for the period 1851 to 2011.

The 2011 Hurricane Season

The 2011 Atlantic Hurricane Season produced nineteen (19) named storms. Of the nineteen (19) storms, seven (7) became hurricanes and four (4) strengthened to achieve major hurricane status (category three (3) or higher on the Saffir-Simpson Hurricane Wind Scale). The strongest tropical cyclone was Major Hurricane Ophelia with peak winds of 140 mph, category 4. By the national definition, Irene and Maria hit Antigua and Barbuda, Ophelia hit Barbuda and brushed Antigua; all were tropical storms at the time of affecting the area. Damages to the islands were minor. The season was well above normal (extremely active) with respect to named storms and near normal in terms of hurricanes. Further, the season was above normal in terms of major hurricanes and Accumulated Cyclone Energy (ACE) index. This season tied with the 2010 and 1995 for the second highest number of named storms for the period 1944 to 2011 and tied with 2010, 1995 and 1887 for the third highest since record began in 1851. The above normal season was attributed primarily to above normal sea surface temperature in the Atlantic and the lingering effects of a La Nina Episode.

It Only Takes One

Regardless of the numbers, we should always approach the hurricane season in the same manner each year: be aware and be prepared. The prevention of the loss of life and property from tropical cyclones is a responsibility that should be shared by all. As a reminder, recall our lesson from Hurricane George of 1998: it only takes one hurricane to make it a bad season. Further, Hurricanes Jose and Lenny impacted us during a near normal season – 1999; thus, we cannot be complacent because the forecast calls for a near normal season.

Accordingly, the Meteorological Service will play its usual role in alerting the public of any tropical cyclone that may form and threaten Antigua and Barbuda, the Leeward Islands and the British Virgin Islands. We endeavour to provide weather and climate information for the protect life, property, livelihood and the enhancement of the economy. Although the hurricane season officially runs from June 1 through November 30 each year, tropical cyclones can and have occurred outside the season – be prepared!

For more information see the links below or email me at You are also welcome to follow us via twitter facebook youtube and blog . Click here for pdf format


Colorado State University, Fort Collins, 2012 Atlantic Season Hurricane Forecast [online]<>%5BAccessed 1 June, 2012]

National Oceanic and Atmospheric Administration, Silver Spring, NOAA 2012 Atlantic Hurricane Season Outlook [online].              Available from: <> %5BAccessed 24 May, 2012]

North Carolina State University, Raleigh, 2012 Atlantic Tropical Cyclone Outlook [online]. Available from: <>%5BAccessed 24 May, 2012]

Tropical Storm Risk, London, Pre-Season Forecast for Atlantic Hurricane Activity in 2012 [online]. Available from: <>[Accessed 24 May, 2012]

United Kingdom Met Office, Exeter, Seasonal Forecasting of Storms [online]. Available from: <>%5BAccessed 24 May, 2012]

The Forecast for 2012 Atlantic Hurricane Season

4 05 2012

Thus far for the year, there have been two forecasts for the North Atlantic Hurricane Season, and both have predicted below normal activity. However, when one looks at the actual forecasts, it could easily be argued that these forecasts are really predicting near normal conditions. Here are the numbers for both forecasts:

                                    Klotzbach and Gray       Saunders and Lea

                             (Colorado State University)     (Tropical Storm Risk)

Named Storms          10 (12)                                            13 (11)

Hurricanes                  4 (6)                                                  6 (6)

Major Hurricanes      2 (2)                                                  3 (3)

The numbers without brackets are the forecast values, and the numbers in brackets are the normal or average values for the season, depending on the base period. The bracketed numbers differ because of the different in base period. From their numbers, I would say near normal conditions are being predicted since this is a climatological Statistical forecast, and in climatology near normal is any value that falls within the middle tercile or 33.3% of the historical data. In this case, near normal would be the range 9 to 12 named storms, 5 to 7 hurricanes, 2 to 3 major hurricanes. Saunders and Lea’s forecast is actually forecasting above normal named storms.

Persons should not put too much confidence in these forecasts thought, as historically, the ability to produce an accurate forecast for the Atlantic Hurricane Season from April lead time is very low. In other words, the Meteorological Community has very little skill in producing an accurate forecast for the hurricane season from this early in the year; these April forecasts are often wrong. See the skills graph below:


December and April’s forecasts show very little to negative skill

The figure above displays the recent 10-year (2002-2011) skill for the forecast number of North Atlantic Hurricanes issued by different organisations.  The forecast precision is assessed using the Mean Square Skill Score (MSSS), which is the percentage improvement in mean square error over a climatological forecast. Positive skill indicates that the model performs better than a climatology forecast (what obtains on average), whilst a negative skill indicates that it performs worse than climatology. Two different climatologies are used: a fixed 50-year (1950-1999) climatology and a running prior 10-year climate norm. The figure compares the forecast skill of the TSR (Tropical Storm Risk), NOAA (National Oceanic and Atmospheric Administration) and CSU (Colorado State University) seasonal hurricane outlooks 2002-2011 as a function of lead time. NOAA does not release seasonal outlooks before late May. It is clear there is little skill in forecasting the upcoming number of Atlantic hurricanes from the prior December and April. Skill climbs slowly as the hurricane season approaches. Moderate skill levels are reached by early June and good skill levels are achieved from early August.

Regardless to the forecast number of named storms and hurricanes, we should always prepare the same for every season as it only takes one hurricane to make it an active season for you and ruin your year. We have recent examples of near normal seasons being quite deadly and costly. Recall in 1992, there were 7 named storms, 4 hurricanes and 1 major hurricane – Andrew. Andrew hit the United States and at the time was the costliest hurricane causing US$26 billion dollars and killing 65 persons. Closer to home, in 1989, there were 11 named storms, 7 hurricanes and 2 major hurricanes, one of which was Hurricane Hugo. Many persons remember Hurricane Hugo; this system absolutely flattened Montserrat; it left in its wake 11 deaths, 3000 homeless  and damage of about EC$1 billion. Hugo also caused significant damage to Antigua costing about EC$200 million.


Colorado State University, Fort Collins, Extended Range Forecast of Atlantic Seasonal Hurricane Activity and Landfall Strike Probability for 2012  [online]. Available from: <> [Accessed 26 Apr 2012].

Texas A&M University, Corpus Christi, Montserrat: Emergency Planning, Response and Recovery Related to Hurricane Hugo [online]. Available from: <> [Accessed 24 April 2012].

Tropical Storms Risk, London, Silver Spring, April Forecast Update for Atlantic Hurricane Activity in 2012 [online]. <>[Accessed 26 Apr 2012].

%d bloggers like this: