Is the Atlantic hurricane season getting longer?

It seems like there have been an unusual number of early and late season tropical storms and hurricanes in the Atlantic in recent years. In 2008, we had four named storms in July, and the second most powerful November hurricane on record. Both 2007 and 2005 had rare December storms, and 2003 featured Tropical Storm Anna, the first April tropical storm ever recorded. This year, Hurricane Tomas made 2010 the fourth straight year with a November hurricane, something that has never occurred in the Atlantic since accurate records began in 1851. Is hurricane season getting longer? Dr. Jim Kossin of the University of Wisconsin published a 2008 paper in Geophysical Research Letters, "Is the North Atlantic hurricane season getting longer?" He concluded that yes, there is an "apparent tendency toward more common early- and late-season storms that correlates with warming Sea Surface Temperature but the uncertainty in these relationships is high".


Figure 1. Observed sea surface temperature (SST) trends during the official North Atlantic hurricane season (June-November) for the period 1950-2007. Units are °C per century. The dashed rectangle denotes the tropical storm formation region south of 30° North latitude and east of 75° West longitude. Data are from the NOAA Extended Reconstructed Sea Surface Temperature V3 product [Smith et al., 2008]. Image credit: Kossin, J., 2008, "Is the North Atlantic hurricane season getting longer?", Geophysical Research Letters, Vol. 35, L23705, doi:10.1029/2008GL036012, 2008.

Methods
Dr. Kossin utilized the "best track" database of Atlantic tropical cyclone activity going back to 1851. However, since lack of satellite and aircraft reconnaissance data before 1950 makes the early part of this record suspect, he limited his analysis to the period from 1950 onward. The era of best data--the satellite era beginning in 1980--was also looked at separately, to ensure the highest possible data quality. The area studied was only a portion of the Atlantic--the tropical storm formation region south of 30° North latitude and east of 75° West longitude. This region has shown considerable warming of the Sea Surface Temperatures (SSTs) since 1950, in excess of 1°F (0.6°C) (Figure 1). A statistical method called "quantile regression" was employed. The method looked at how certain thresholds that mark the beginning and end of hurricane season have changed over the years. For example, the date where 5% of all tropical storms form earlier than that date, was called the 0.05 quantile, and the date where 5% of all tropical storms form later than that date, was called the 0.95 quantile. Kossin was able to show that the date of the 0.05 quantile got steadily earlier and the date of the 0.95 quantile steadily got later since 1950. Hurricane season for both the period 1950-present and 1980-present got longer by 5 to 10 days per decade.


Figure 2. Trends in tropical storm formation dates, in the region south of 30° North latitude and east of 75° West longitude, at the 0.05.0.95 quantiles. Trends are based on the periods (left) 1950-2007, and (right) 1980-2007. The dates (month/year) associated with the 0.05, 0.25, 0.50, 0.75, and 0.95 quantiles for each period are shown on the top axis (these threshold dates are based on the full sample for each period). Shading denotes the 90% confidence interval. Image credit: Kossin, J., 2008, "Is the North Atlantic hurricane season getting longer?", Geophysical Research Letters, Vol. 35, L23705, doi:10.1029/2008GL036012, 2008.

Relationship with Sea Surface Temperature
The broadening of the Atlantic hurricane season found was strongly dependent upon Sea Surface Temperatures. Both the onset and end of hurricane season shifted by 20 days per degree C of warming of the SST. With global warming projected to increase tropical Atlantic SSTs 1-2°C by the end of the century, can we then expect a 40-80 day increase in the length of hurricane season? Dr. Kossin doesn't explore this possibility, and doesn't blame the observed increase in the length of the season on human-caused global warming of the oceans. There is reason to believe that future warming of the Atlantic SSTs won't necessarily broaden the area over which tropical storms will form, though. Papers by Henderson-Sellers et al. (1998) and Knutson et al. (2008) theorize that as SSTs warm, the lowest temperature at which tropical storms can form will also increase. The current minimum temperature of 26.5°C (80°F) may increase to 28.5°C for a 2°C warming of Atlantic SSTs. Johnson and Xie (2010) have found observational evidence that the lowest temperature at which tropical storms can form has indeed been increasing at about 0.1°C per decade in the Atlantic, in line with climate model predictions.

References
Henderson-Sellers, A., et al., 1998, "Tropical Cyclones and Global Climate Change: A Post-IPCC Assessment", Bull. Am. Meteorol. Soc. 79, 19–38.

Johnson, N.C., and S.P. Xie, 2010, "Changes in the sea surface temperature threshold for tropical convection", Nature Geoscience doi:10.1038/ngeo1008

Knutson, T.R., J.J. Sirutis, S.T. Garner, G.A. Vecchi, and I.M. Held, 2008, Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions", Nature Geoscience 1, 359 - 364 (2008), doi:10.1038/ngeo202

Kossin, J., 2008, "Is the North Atlantic hurricane season getting longer?", Geophysical Research Letters, Vol. 35, L23705, doi:10.1029/2008GL036012, 2008.

Jeff Masters