Early Riser: Does a Busy 2017 Equate to an Early Start for North Atlantic Hurricanes in 2018?

Although my colleague Peter Holland declared the Atlantic hurricane season officially closed, there are reports in the media that environmental conditions in the North Atlantic basin may be favorable enough to sustain tropical cyclones, potentially adding to what has already been a very active year and perhaps foretelling an early start to hurricane activity in 2018.

The official hurricane season for the North Atlantic runs from June 1 to November 30 and encompasses over 97 percent of annual climatological activity. Out-of-season tropical systems — events happening through December to May — are rare, but not unprecedented. The U.S. National Oceanic and Atmospheric Administration (NOAA) Hurricane Research Division (HRD) states that since 1851, 88 storms have been observed in the Atlantic during the off-season — that’s about one out-of-season storm every two years. In theory, this could be an underestimate, as the peer-reviewed literature suggests an undercount of Atlantic tropical cyclones prior to the satellite reconnaissance era.

Figure 1: Number of storms that have formed for each calendar month during the off-season. Data: National Oceanic and Atmospheric Administration (NOAA) Hurricane Research Division (HRD).

Looking at the National Hurricane Center’s records reveals there has been a named storm in each month of the year, demonstrating that tropical storms and hurricanes can develop at any time of year if the right environmental conditions are present. Tropical storms forming in the off-season tend to develop from non-tropical sources, such as upper-level lows and frontal systems, which can intensify into tropical storms over warm sea surface temperatures. This was the case earlier this year when Tropical Storm Arlene formed from what was a non-tropical low. The system drifted south across the Atlantic for several days and intensified to a named storm over warmer waters. It marked the third consecutive year a storm had been named before June 1, and the fifth year in the last decade.

Notably, most out-of-season storms occur in May, immediately prior to start of the season, or in December, immediately after the end of the season, with 49 and 17 named storms forming in each month respectively. There have been seven storms named in December since 1975 — an average of one storm every six years.

The good news is that these out-of-season tropical systems are generally weaker than the storms that develop during the peak months of the hurricane season. A marked increase in wind shear — a change in wind direction or speed with height — across the Gulf of Mexico and Caribbean Sea, and cooler sea surface temperatures (SSTs), acts to suppress cyclone formation. Moreover, cyclogenesis in the off-season historically shifts to the central and south-central Atlantic, well east of Bermuda, meaning the risk of a potential landfall in the U.S. is also considerably reduced.

However, there have been several out-of-season storms with some notable characteristics:

Hurricanes Can Form in the Off-season

  • Although not officially listed in the historical record, Hurricane Amanda is one of two out-of-season hurricanes to have reached Category 2 status and the only U.S. landfalling hurricane in the month of May on record. The system made landfall near Apalachicola, Florida on May 28, 1863.

They Can Impact the U.S. East Coast…

  • Hurricane Able is one of only four hurricanes to form in May on record. Although the system did not make landfall in the U.S., it attained hurricane status near the coast of Florida in May 1951 and tracked within 70 miles (115 kilometers) of Cape Hatteras as a Category 1 hurricane.

And the Caribbean…

  • Tropical Storm Olga was a short-lived storm that made landfall in the Dominican Republic in December 2007. The system caused significant damage in the Dominican Republic, Puerto Rico, and Haiti.
  • In December 2003, Tropical Storm Odette became the first tropical storm on record to form in the Caribbean Sea. The system made landfall in the Dominican Republic, causing widespread damage.

The Threat Is Not Just From Wind

  • Tropical Storm Bonnie made landfall near Charleston, South Carolina in May 2016 and brought heavy rainfall to coastal areas of the Carolinas, causing extensive flood inundation in Ridgeland, South Carolina.

These events are a gentle reminder that tropical cyclones can be impactful even in the off-season. So what is the likelihood of a significant event between now and the start of the official start of the 2018 North Atlantic hurricane season on June 1?

As the article from Bloomberg points out, sea surface temperatures in the western tropical Atlantic and Caribbean Sea remain anomalously warm at 26.5 to 28.5 degrees Celsius, warm enough to a support tropical system should one develop. However, a look at the medium-term operational model weather forecasts and the Global Tropics Hazards and Benefits Outlook from NOAA shows no indications of cyclone development in the North Atlantic, so it appears likely that activity in the basin in 2017 has ended.

Figure 2: North Atlantic Sea Surface Temperatures (SSTs) as of December 12, 2017. Data: National Weather Service (NWS), National Hurricane Center (NHC), National Centers for Environmental Prediction (NCEP), National Oceanic and Atmospheric Administration (NOAA).

Obviously it is too early for a skilled forecast of activity this spring, but we should continue to monitor the equatorial Pacific closely. La Niña conditions, the cooler phase of the El Niño-Southern Oscillation (ENSO), officially emerged in November — sea surface temperatures across the central and eastern equatorial Pacific Ocean are currently 0.8 degrees Celsius below average — and are forecast to continue through to the spring, as demonstrated by the latest model consensus released by NOAA on December 11, 2017. The La Niña phase of ENSO normally enhances hurricane activity in the North Atlantic, owing to weaker wind shear and increased atmospheric instability, so conditions next spring in the Atlantic basin may be more favorable than average.

Figure 3: Weekly sea surface temperature anomaly (°C) for the Pacific Ocean for November 27 to December 3, 2017. Data: Australian Bureau of Meteorology (BOM)

Figure 4; Nino3.4 SST anomaly forecast from the NCEP coupled forecast system model version 2 (CFSv2). The threshold for El Niño conditions is a Niño 3.5 SST anomaly greater than 0.5°K, while for La Niña conditions the threshold is a Niño 3.4 SST anomaly of less than -0.5°K. Data: National Weather Service (NWS), National Centers for Environmental Prediction (NCEP), Climate Prediction Center (CPC).

Of course, La Niña conditions do not guarantee increased tropical activity. Five recent spring La Niña years (2006, 2008, 2009, 2011, and 2012) yielded only three out-of-season storms, and two of them formed in 2012 alone. Even if La Niña conditions persist, several other factors determine cyclone development, including the presence of a disturbance, tropospheric moisture, and warm sea surface temperatures.

Looking further ahead to the 2018 hurricane season, Tropical Storm Risk (TSR), a renowned publisher of seasonal tropical cyclone activity forecasts, recently published its extended range forecast for North Atlantic hurricane activity. TSR calls for a slightly above-average season, citing reduced trade wind speed in the tropical North Atlantic and Caribbean Sea to be the key meteorological driver. In its December outlook for the 2018 season, Colorado State University (CSU) identified SSTs and the state of ENSO among several factors that could influence activity in 2018. However, it goes without saying that uncertainties at this time of year are significant and the forecast skill at this forecast lead time is historically low for named tropical cyclones at just seven percent (2008-2017, TSR).

RMS, as always, continues to monitor these forecasts as they are released and updated and will publish a comprehensive summary in the weeks ahead of what could be another active hurricane season in 2018.

Based in London, James works within Model Development as a member of the RMS Event Response team, supporting real-time Event Response operations and assisting on various Event Response projects. James holds a bachelor’s degree in Physical Geography and Geology from the University of Southampton and a master’s degree in Applied Meteorology from the University of Reading.

Leave a Reply

Your email address will not be published. Required fields are marked *