Forecast Histories

Have you ever wondered if hurricane forecasters are any good at their jobs? Are meteorologists at the National Hurricane Center and Colorado State University accurate at predicting activity for each Atlantic Basin hurricane season? Don’t be fooled: the U.S. (especially Florida) might have gotten walloped in 2004 and 2005, but all in all, the staffs at both the NHC and CSU know what they’re doing (mind you, they’re not perfect, but they ain’t so bad either).

Seasonal predictions aren’t necessarily so much “forecasts” as they are “hindcasts”- much of what the NHC, CSU and others (including the Coastal Fluid Dynamics Laboratory (CFDL) in Raleigh, NC and the Tropical Storm Risk Consortium (TSR) in London) theorize is based on historical activity, modeling and what meteorologists “think” will happen. While historical data is important (hindcasting), severe weather meteorologists also look at a number of atmospheric factors as seasons approach to fine tune long range forecasts (see Reference #6). Among them:

El Niño: As outlined in the terminology section, El Niño is the warming of Pacific Ocean currents along the coasts of Peru and Ecuador that influence dramatic changes in the weather across the globe. Warm El Niño weather weakens trade winds which in turn raises the temperature of surface water that flows from west to east. The warmer waters produce strong upper level winds; the winds in turn “sheer off” the tops of weather patterns diminishing the chances of severe storm development, or keeping developed hurricanes at a low category of intensity.

El Niño runs on a sort of schedule, about every three to five or seven years. The term is Spanish for “the boy;” the weather phenomenon got its name during the 19th century when fishermen noticed the pattern of weather changes. Those changes showed up around Christmastime- “the boy” was named after Jesus.

In terms of hurricanes, El Niño is a good thing in that the systems diminish Atlantic hurricane activity, although that warmer water produces more rain in coastal areas as well as warmer temperatures in the winter. The bad news- all that rain increases the chances of area flooding during peak times from December through April.

La Niña is considered the polar opposite to El Niño: La Niña is unusually cooler water in the Pacific Ocean, leading to drier summers and warmer winters in the southeastern U.S. That drier air cools the trade winds and reduces or eliminates upper level winds. The result: storms have more of an opportunity to develop and can be more intense. La Niña is not a good thing if you live in a hurricane prone area (take note those on the eastern seaboard and the Gulf of Mexico).

In a published study in the July 3, 2009 issue of Science, climatologists at Georgia Tech’s School of Earth and Atmospheric Sciences reported that El Niño might not be as simple a phenomenon as once thought and there might be different classifications of the weather pattern. The Georgia Tech team announced they had discovered El Niño Modoki, or a hybrid/modified version of the El Niño weather trend that instead of retarding the development of strong storms, instead behaves more like La Niña and aids in that development (Modoki by the way is Japanese and means similar, yet different). In instances of El Niño Modoki, development of Eastern Pacific warming (EPW) occurs further west leading to Central Pacific warming (CPW). Although the signature is that of a typical El Niño weather pattern (warm weather heating up the waters), the results instead are more like that of a La Niña system. The initial study was inconclusive as to why warmer weather in different parts of the Pacific produces such drastically different results, instead concentrating on the possible discovery of a trend leading to a new classification and a better distinction of the influence of weather patterns.

There is encouraging news in this theory if it pans out: the researchers believe El Niño Modoki is more predictable and can be identified earlier than the typical El Niño pattern (that pattern really only materializes about a month before the start of the hurricane season). If scientists recognize the signs of El Niño Modoki versus El Niño, they can now factor in that information earlier in forming predictions of an upcoming hurricane season. Furthermore, recognizing the phenomenon of El Niño Modoki decreases the likelihood of complacency and false security in the years of warmer temperatures in the Pacific Ocean. NOAA data compiled between 1899 and 1996 had shown a trend of 36% fewer named Atlantic basin storms and a 6% decrease in intensity during periods of El Niño (see Reference #7). The co-authors of the study were Judith A. Curry and Peter J. Webster with research by Hye-Mi Kim.

Water Temperatures: The simple part to remember is warm water gives hurricanes “fuel” to strengthen; cool water starves a storm system. Without warm ocean water (with heat from condensing water vapor), a storm will lose its strength. Now the complicated part: warm water moves around and isn’t always so easy to locate. The Loop Current (also previously mentioned in the terminology section), is the name used to describe tropically warm to hot water that travels around the eastern portion of the Gulf of Mexico. It is created when water from the Caribbean Sea passes through the Yucatán Strait (between Mexico and Cuba). Warm water in the Gulf can be as deep as 100-150 feet, but in the Loop Current, that depth can be as much as 250 feet. As mentioned, hurricanes strengthen over warm water and lose strength when cooler water rises to the surface. High temperature water (above 80 degrees) is an almost never ending fuel supply for a hurricane; if that supply is hundreds of feet deep, colder water has almost no chance of getting to the surface and cutting short a hurricane’s energy.

Of note: both Hurricane Katrina and Hurricane Rita (2005) passed over the Loop Current strengthening each from a Category 1 to a Category 5 in a matter of days. Furthermore, Katrina remained a much more powerful storm because it crossed into a Loop Current Eddy, a separate ring of warm water that can break away from the Loop Current.
Bermuda High: The “semi-permanent” phenomenon centers over Bermuda (off of the coast of the Carolinas) each summer and acts as a clockwise rotation scooping up Atlantic Ocean storms and swinging them to the north away from eastern seaboard. Strong Bermuda Highs are somewhat regular occurrences in the first two thirds of a storm season and help keep storms away from the U.S. That’s the good news: the bad news is occasionally a system becomes so large or drifts so far south it will drive a storm system into the warm waters of the Gulf of Mexico or directly towards the Yucatán Peninsula. Also, the Bermuda High dissipates rapidly around early October making the U.S. more susceptible to a storm making landfall. When the phenomenon moves towards the eastern part of the Atlantic, it is known as an Azores High.

African/Saharan Dust: A relatively new tool added to the arsenal of hurricane forecasting, meteorologists have discovered dust from the African continent blowing into the Atlantic Ocean suppresses severe storm activity. Here’s how it works: dust cloud density cuts down on solar energy that heats water; the result is lower ocean temperatures and a shortage of energy (dryer air) to fuel a hurricane. The Saharan air layer emerges in the spring and summer and settles into the atmosphere between 5,000 and 25,000 feet high. Go figure. As with anything else, one man’s joy is another man’s misery: high dust mass reducing hurricane activity only occurs when African rain is at its lowest during periods of drought (see Reference #7).

Remember that seasonal forecasting is harder and less accurate then storm forecasting. During storm forecasting, meteorologists have the benefit of measuring current atmospheric conditions, while for seasonal forecasting meteorologists look at what factors should be prevalent over a storm season; those factors sometimes change mid stride.
Here’s a rundown on seasonal forecast histories and how the predictions measured up against storm activity. El Niño and La Niña references through 2009 came from Emily Nipps July 14, 2009 sidebar (see Reference #8):

2002
El Niño
(Moderate)
2003
Neutral
2004
El Niño
(Weak)
2005
Neutral
NHC Predicted
Storms:
9-13 11-15 12-15 12-15
NHC Predicted Hurricanes 6-8 6-9   6-8 7-9
NHC Predicted  Intense: 2-3 2-4 2-4 3-5
CSU Predicted
Storms:
12 12 14 13
CSU Predicted Hurricanes: 7 8 8 7
CSU Predicted
Intense:
3 3 3 3
Actual
Storms:
12 16 15 28*
Actual
Hurricanes:
4 7 9 15
Actual Intense Hurricanes: 2 3 6 7



 
2006
El Niño
(Weak)
2007
La Niña
(
Moderate)
2008
Neutral
2009
El Niño
(Moderate)
NHC Predicted
Storms:
13-16 13-17 12-16 9-14
NHC Predicted Hurricanes: 8-10 7-10 6-9   4-7
NHC Predicted  Intense: 4-6 3-5 2-5 1-3
CSU Predicted
Storms:
17 17 15 12
CSU Predicted Hurricanes: 9 9 8 6
CSU Predicted
Intense:
5 5 4 2
Actual
Storms:
10* 15** 16 9
Actual
Hurricanes:
5 6 8 3
Actual Intense Hurricanes: 2 2 5 2
2010
La Niña
(Moderate)
2011
La Niña
(Weak to Moderate)

2012
El Niño

2013

 

NHC Predicted
Storms:
               14-23                12-18
NHC Predicted Hurricanes:                  8-14                 6-10
NHC Predicted  Intense:                   3-7                  3-6
CSU Predicted
Storms:
15                   16 10
CSU Predicted Hurricanes: 8                    9 4
CSU Predicted
Intense:
4                    5 2
Actual
Storms:
                  19                   19***
Actual
Hurricanes:
                   12                    7
Actual Intense Hurricanes:                     5                    3      

Finally, in addition to CSU and NHC, other entities also issue long range tropical forecasts. Among them: AccuWeather,Florida State University’s Center for Ocean-Atmospheric Prediction Studies (COAPS), North Carolina State University’s Coastal Fluid Dynamics Lab (CFDL), Penn State University’s Earth System Science Center (ESSC), Tropical Storm Risk Inc. (TSR), UKMet’s GloSea and Weather Systems International (WSI).

*You will find some references that 2006 only had nine storms. A post-season review revealed that on July 17th-18th, the same system that spawned Tropical Storm Beryl also spun off a previously unnoticed system off the coast of Nova Scotia. The analysis was done in December of 2006; the storm was given the name of “Unnamed Tropical Storm” and pushed season storm totals from nine to ten. The same thing occurred in 2005 when meteorologists upped the storm total from 27 to 28 after a post-season review…

**For 2007 there were technically 15 named storms although some people will say Sub-Tropical Storm Andrea didn’t really count. I think it does, but if you only see 14 storms for the 2007 season, now you know why…
***In 2011 you’ll notice the NHC only used 18 names but there were 19 storms. Similar to 2005 and 2006, the agency discovered after the fact a short lived tropical storm from early September that formed b/w Bermuda and Nova Scotia. The system chronologically fell in between storms Irene and José and was given the generic moniker Unnamed Tropical Storm. In that same review of satellite recordings, the NHC also upgraded Tropical Storm Nate to Category 1 hurricane status pushing the number of hurricanes for the season from 6 to 7…

In 2000, both Gray and the NHC predicted 11 named storms would form in the Atlantic Basin (there were actually 14 that year). For 2001, Gray predicted 10 storms in the Atlantic Basin, the NHC predicted 8-11 (there were actually 15 that year). 2000 was the last year the NHC used a single number for storm predictions; starting in 2001, the agency went to a range of numbers based on threes (see Reference #9).

Press Sources:

Riccardi, Nicholas. “Eminence Grise of Hurricane Forecasting.”
Los Angeles Times
May 30, 2006

“Max Mayfield, National Hurricane Center Director.”
USA Today
May 15, 2000

Spotts, Peter N. “Active Hurricane Season, but Who Listens?”
Christian Science Monitor
May 23, 2006

NOAA/NHC Sources:

Vaccaro, Chris, Susan Bachman. “NOAA Expects Busy Atlantic Hurricane Season.”
NOAA Public Affairs
May 27, 2010

Gillis, Carmeyia, David Miller. “NOAA Issues Atlantic Hurricane Season Outlook, Encourages Preparedness.”
NOAA Public Affairs
May 21, 2009

Gillis, Carmeyia, David Miller. “NOAA Predicts Near Normal or Above Normal Atlantic Hurricane Season.”

NOAA Public Affairs
May 22, 2008

Gillis, Carmeyia, Greg Romano. “NOAA Predicts Above Normal 2007 Atlantic Hurricane Season. 13 to 17 Named Storms Predicted.”

NOAA Public Affairs
May 22, 2007

Gillis, Carmeyia, Chris Vaccaro. “NOAA Predicts Very Active 2006 North Atlantic Hurricane Season. Residents in Hurricane Prone Areas Urged to Make Preparations.”
NOAA Public Affairs
May 22, 2006

Lepore, Frank, Greg Romano. “NOAA Issues 2005 Atlantic Hurricane Season Outlook. Another Above Normal Season Expected.”
NOAA Public Affairs
May 16, 2005

Lepore, Frank. “Above Normal 2004 Atlantic Hurricane Season Predicted.”
NOAA Public Affairs
May 17, 2004

Lepore, Frank. “NOAA Forecasters Say Six to Nine Hurricanes Could Threaten in 2003. NOAA and FEMA Stress Preparedness for Residents in Hurricane Prone Areas.”
NOAA Public Affairs
May 19, 2003

Lepore, Frank. “Hurricane Forecasters Say 6 to 8 Hurricanes Could Threaten in 2002. NOAA Expects Normal to Slightly Above Normal Atlantic Storm Activity.”
NOAA Public Affairs
May 20, 2002

Lepore, Frank. “Hurricane Forecasters Expect Normal Atlantic Storm Activity in 2001. NOAA Says 5 to 7 Hurricanes Could Threaten.”
NOAA Public Affairs
May 21, 2001

Colorado State Sources:

Gray, William M. and Philip J. Klotzbach. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and Landfall Strike Probability for 2010.”

Colorado State University Department of Atmospheric Science

April 7, 2010

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2009.”
Colorado State University Department of Atmospheric Science
April 7, 2009

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2008.”
Colorado State University Department of Atmospheric Science
April 9, 2008

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2007.”
Colorado State University Department of Atmospheric Science
April 3, 2007

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2006.”

Colorado State University Department of Atmospheric Science
April 4, 2006

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2005.”
Colorado State University Department of Atmospheric Science
April 1, 2005

Gray, William M., Philip J. Klotzbach and William Thorson. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2004.”
Colorado State University Department of Atmospheric Science
April 2, 2004

Gray, William M., Philip J. Klotzbach, Christopher W. Landsea, Eric Blake, William Thorson and Jason Conner. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2003.”
Colorado State University Department of Atmospheric Science
April 4, 2003

Gray, William M., Christopher W. Landsea, Eric Blake, Philip J. Klotzbach, William Thorson and Jason Conner. “Extended Range Forecast of Atlantic Seasonal Hurricane Activity and U.S. Landfall Strike Probability for 2002.”
Colorado State University Department of Atmospheric Science
April 7, 2002

Gray, William M., Christopher W. Landsea, Paul W. Mielke Jr., Kenneth J. Berry, Eric Blake, John Sheaffer and William Thorson. “Updated Forecast of Atlantic Seasonal Hurricane Activity and U. S. Landfall Strike Probability for 2001.”
Colorado State University Department of Atmospheric Science
April 6, 2001

Reference #6: Nipps, Emily. “Why El Niño is a Welcome Visitor.” St. Petersburg Times. July 14, 2009: Page 1A.

Reference #7: Nipps, Emily. “Storm Predictors.” St. Petersburg Times. April 13, 2009: Page 1A.

Reference #8: Nipps (Fast Facts) July 14, 2009.

Reference #9: Brink, Graham (with research by Angie Drobnic Holan). “An Unpredictable Season.” St. Petersburg Times. November 30, 2006: Page 1B.