Atlantic hurricane season begins Sunday: Will record streak without major hurricane landfall end?

After a six-month break, hurricane season is here again. It is expected to be a relatively quiet season, but that does not mean we should let down our guard; just a single landfall at a vulnerable location can be a major disaster.

Some very lucky years

The U.S. has been extraordinarily fortunate lately: we have not been witness to the fury of a major hurricane (category 3 or higher) landfall since October 2005 when Wilma hit southwest Florida as a Category 3 storm. (Other countries have not had such good fortune these past few years. )

Since the hyper-active 2005 season, the U.S. has had just six Category 1 and 2 hurricane landfalls: Humberto (TX), Ike (TX), Gustav (LA), Dolly (TX), Irene (NC), and Isaac (LA).  Sandy was not technically a hurricane at its NJ landfall, and if it were, it would have been a Category 1 storm.

Days between major hurricane landfalls in the U.S., 1900-2014.  (Roger Pielke, Jr.)

Days between major hurricane landfalls in the U.S., 1900-2013. (Roger Pielke, Jr.)

Such a streak, or “drought”, is unprecendented going back to 1900.  As of the start of this hurricane season, the span will be 3,142 days since the last U.S. major hurricane landfall.  The previous longest span is about 2½ years shorter!  While this is a relief for coastal residents and businesses, it inevitably increases complacency.  The longer you go between events, the less likely you are to fully respect the next one and heed warnings.

Will this year end the streak?  There’s no way of knowing.  Even during an inactive season, a single storm can leave its mark on history (think Andrew in 1992).  And conversely, very active seasons like 2010 can yield no landfalls for the U.S.

Prime time is August through October

Although tropical cyclones can and do form outside of the official hurricane season, 97 percent of a season’s activity is historically confined to the June 1 through November 30 period. Within the season, the core months of Aug.-Sep.-Oct. comprise roughly 78% of the season’s total hurricane activity.

Average seasonal cycle of Atlantic tropical cyclone activity. The yellow-orange-red shading highlights the official hurricane season, while the thick red lines bound the three-month bulk of the season: August through October. (NOAA)

Average seasonal cycle of Atlantic tropical cyclone activity. The yellow-orange-red shading highlights the official hurricane season, while the thick red lines bound the three-month bulk of the season: August through October. (NOAA)

What’s in a name?

Since 1950, tropical cyclones in the Atlantic Ocean have been given names to help quickly and uniquely identify storms. For 29 years, different naming conventions were used. Then in 1979, the current system of alternating male and female names was introduced, and the same six lists are rotated through in six years. So unless a name is retired, it will be used again six years later.

Occasionally, the first time a name is used is also the last. The most recent example is Igor; Igor was introduced to the list when Ivan was retired in 2004. But, Igor was then retired after its one and only use in 2010. According to the National Hurricane Center, names get retired “if a storm is so deadly or costly that the future use of its name on a different storm would be inappropriate for reasons of sensitivity”.

The list of names for the 2014 season includes three new entries to replace names that were retired after the 2008 season: Gonzalo (replaces Gustav), Isaias (replaces Ike), and Paulette (replaces Paloma). Of the 21 names in the list, twelve of them are still in circulation since their first use in 1984, including the first two (Arthur and Bertha).

Seasonal outlook

During an average season (using 1981-2010 as a baseline), there are 12 named storms in the Atlantic basin. Of those, 6-7 become hurricanes, and 2 become major hurricanes (Category 3-5).  I acquired eight different seasonal forecasts by various agencies, and rather than listing all of them here, I will just share the consensus, which often has more skill than any particular forecast anyway. The consensus forecast I calculated is 10 named storms, 4 hurricanes, and 1 major hurricane. In terms of ACE (Accumulated Cyclone Energy), the consensus is about 80 percent of the median value (~73×10⁴ kt²).

 

Related: NOAA predicts below normal Atlantic hurricane season | Forecasters predict lackluster hurricane season thanks to El Niño forecast

One of the main reasons that seasonal forecasters are anticipating reduced activity is a high probability of a strong El Niño forming. Although the phase of ENSO (El Niño Southern Oscillation) is determined by the ocean and atmosphere in and over the tropical Pacific Ocean, its effects are felt around the world, including the tropical Atlantic Ocean. Thirty years ago, Dr. William Gray at Colorado State University found and documented an important relationship: when ENSO is in an El Niño (warm) phase, Atlantic hurricane activity is noticeably suppressed… when it’s in a La Niña (cool) phase, Atlantic hurricane activity is enhanced.

Published in the same paper as that now-well-known ENSO relationship, Gray also linked the QBO to Atlantic hurricane activity. The QBO, or Quasi-Biennial Oscillation, is a shift in the direction of the tropical stratospheric winds with a period of about 27 months. He found that when the winds at 30 millibars (~80,000 feet) were westerly or becoming westerly, Atlantic hurricane activity was enhanced, and when the winds were easterly or becoming easterly, activity was suppressed. As of this month, the QBO is just about to enter an easterly phase, which by itself may not be anything critical, but it’s just another signal that could make a contribution to suppressing the overall activity.

Time-height plot of monthly mean zonal winds at the equator since 2001.  Westerly winds are shaded gray, easterly winds are white, and the contour interval is 10 m/s.  (Adapted from Freie Universitat Berlin)

The heartbeat of the QBO: time-height plot of monthly mean zonal winds at the equator since 2001. Westerly winds are shaded gray, easterly winds are white, and the contour interval is 10 m/s. (Adapted from Freie Universitat Berlin)

Understanding uncertainty

One aspect of hurricanes that is both commonly used and commonly misunderstood is the track forecast and the “cone of uncertainty”. The National Hurricane Center (NHC) uses a cone on its forecast track graphics to convey typical forecast errors. While a center line is drawn for clarity, forecasters stress not to focus too much on it because there are likely to be errors in the forecast, and the storm will affect areas far away from the center.

The cone is designed to enclose two-thirds of the recent forecast errors, meaning that on average, there is still a 1-in-3 chance that the center of the storm will track outside of the cone. Of course, the center of the storm is just a small part of the story. Since every storm extends well beyond its center, the impacts of a storm will be felt several hundreds of miles away. Thus, the forecast cone is NOT a storm impacts cone.  The impacts of a storm, whether damaging winds, flooding rain, or storm surge, can extend hundreds of miles from the center.

Presently, the same cone size is used for an entire season and for all storms. The 2014 cone size is determined by NHC’s track errors from the 2009-2013 seasons. There is greater uncertainty at longer forecast times, and that is why the cone is smallest near the storm’s current position and largest at the longest-range forecast. But, as track forecasts improve, the cone gets smaller! The figure below compares the cone used in 2008 to the cone that will be used in 2014 for an identical [hypothetical] storm and forecast. This is great progress, as reducing the uncertainty in a forecast can reduce the time and resources spent on preparations and evacuations.

Comparison of this year’s forecast cone (red) to the cone used six years ago (green) for a hypothetical storm and track forecast.

Comparison of this year’s forecast cone (red) to the cone used six years ago (green) for a hypothetical storm and track forecast.

In addition to track uncertainty, there is also uncertainty in the intensity forecast. For example, the average intensity error for a two-day forecast is 15 mph, but there will be some individual examples with much greater error due to rapid intensification or rapid weakening, which are still poorly forecast.

A very important point to make is that the intensity, as defined by a category on the Saffir-Simpson Hurricane Wind Scale, only accounts for the peak wind speed found somewhere in the storm. It is critical that we all understand the destructive power of the storm extends far beyond that single strongest wind speed value: storm surge, inland flooding, tornadoes, and the outer wind field are all destructive and deadly forces that reach hundreds of miles away from the center. You could live 200 miles away from the landfall location and only get tropical storm force winds, but get inundated with an 8-foot storm surge. In other words, there’s more to the story than the category!

New storm surge products

Cause of death in the U.S. directly attributable to Atlantic tropical cyclones, 1963-2012. (Rappaport, 2014).

Cause of death in the U.S. directly attributable to Atlantic tropical cyclones, 1963-2012. (Rappaport, 2014).

When you think of a hurricane, chances are, you think of the destructive wind associated with it. However, according to the latest statistics, wind only accounts for 8% of the deaths associated with tropical cyclones in the U.S.  Storm surge is the number 1 killer by a significant margin at 49%. Rain and inland flooding is in second place at 27%.

One problem is that a lot of coastal residents don’t know their elevation, or whether they’re at risk for storm surge inundation.  You do not need to be immediately on the coast either… rivers, inlets, and canals that are hydrologically connected to the ocean/bay can also rise quickly.  A storm surge occurs when wind-driven water piles up on land, and water levels rise by 1, 5, 10, or even 25+ feet depending on the exact details of the storm and the terrain.  It is much more complicated than simply assigning a surge depth to a category of storm.  Also be aware that the phase of the normal local daily tide makes a difference — a peak storm surge arriving during high tide will be far more destructive than if it had arrived during low tide.

Some recommended resources:

   * Storm surge information and product examples: http://www.nhc.noaa.gov/surge/

   * New experimental maps that will be available this year: http://www.nhc.noaa.gov/experimental/inundation/

   * Follow the NHC Storm Surge Unit on Twitter: https://twitter.com/NHC_Surge

Storm surge in New Jersey during Sandy in October of 2012... not even a Category 1 hurricane.  (photo by Scott Anema)

Storm surge in New Jersey during post-tropical storm Sandy in October of 2012. (photo by Scott Anema)

Getting ready…

Without knowing if and when a hurricane might come to visit your area, what can you do now?  If you live near the coast, there are action items that you can address anytime to maximize your preparedness and minimize your stress. 1) know your evacuation zone and storm surge vulnerability, 2) exterior: have trees trimmed away from your house and power lines, remove debris and trash, locate and inspect your window protection, 3) interior: buy non-perishable staples like water and canned goods, check your insurance policies for coverages and accuracy, and 4) think about evacuation plans (where would you go, what are your criteria).

Also on Capital Weather Gang

D.C. area forecast: Murky 60s and 70s today clear into comfortably warm weekend