Jeff co-founded the Weather Underground in 1995 while working on his Ph.D. He flew with the NOAA Hurricane Hunters from 1986-1990.
By: Dr. Jeff Masters , 6:33 PM GMT on January 09, 2006
Good Tuesday to everyone! This week we need to watch a large extratropical low-pressure system sliding down the coast of Africa towards the Cape Verde Islands. This low is similar to the storms that spawned Delta, Epsilon, and Zeta, and has the potential to slowly gain tropical characteristics and become Tropical Storm Alberto later this week. However, it appears that wind shear levels will probably be too high and the water too cool for a new tropical storm to form. The chances of a tropical storm forming this week are probably around 20%.
Has there been an increase in late-season tropical storm activity?
Hurricane experts agree that global warming has not led to an increase in the number of tropical cyclones occurring world-wide, and are currently debating whether or not global warming has affected tropical cyclone intensity (more on this later in January, I've been pulling together a lot of material). Is global warming possibly affecting the length of hurricane season, as well? It seems that an inordinate number of late-season and off-season tropical storms have been forming in the Atlantic the past few years. For example, two December storms formed in 2003, which also had the first-ever April storm, Tropical Storm Ana. Cuba's worst hurricane ever, Hurricane Michelle, hit in November 2001, and the Atlantic's second deadliest hurricane of all time, Hurricane Mitch, lasted into November 1998. Add to this 2005's Greek cousins, Gamma, Delta, Epsilon, and Zeta, which all occurred in November and December. To see if hurricane season is indeed lasting longer, I plotted up the number of days a named storm existed each year from November through April (Figure 1). The data cut-off is 1944--the beginning of reliable hurricane records in the Atlantic, thanks to regular long-range aircraft reconnaissance missions. According to Dr. Chris Landsea's paper, A Climatology of Intense Atlantic Hurricanes, only a very few short-lived tropical storms that formed far out over the open Atlantic were missed by these aircraft missions or ships plying the shipping lanes between Europe and North America. For example, all of 2005's Greek storms were long-lived enough and sufficiently intense that they would have been detected back in the 1944-1960 time frame. Beginning 1960, weather satellites gave us full coverage of all the ocean areas, and it is unlikely we missed any tropical storms after then. Thus, Figure 1 is likely to be an accurate measure of the late-season tropical storm activity for the Atlantic.
Figure 1.Number of days a named tropical storm was present in the Atlantic for each year during November through April, 1944-2005. The 2.5 named tropical storm days from the March 2004 hurricane in the South Atlantic that hit Brazil--Hurricane Catarina--are not included.
Looking at Figure 1, we see a noticeable increase in the number of late-season named-storm days in the Atlantic in the past decade, roughly coinciding with the upswing in Atlantic intense hurricane activity that began in 1995. This increase in late-season tropical cyclone activity was not observed during the previous warm phase of theAtlantic Multidecadal Oscillation (AMO), the natural cycle that greatly influences hurricane activity in the Atlantic. This previous warm phase of the AMO lasted from 1926-1969. Thus, it seems unlikely that the recent upswing in late-season Atlantic tropical storm activity is due to the AMO. Is global warming to blame, then? Global sea surface temperatures in the tropics have increased by .3 degrees C (.5 degrees F) the past century, so it is reasonable to ask if this increase has lengthened hurricane season.
To answer this question, we look at the November though April number of tropical storm days for the Northern Hemisphere's other ocean basins that have tropical cyclones--the Western Pacific (Figure 2) and the Eastern Pacific (Figure 3). Neither ocean basin shows any increase in the length of their hurricane seasons, so global warming has not caused a Northern Hemisphere-wide increase in the length of hurricane seasons. If global warming is to blame for the recent increase in Atlantic late season and off-season tropical storm activity, it is probably through some as yet not understood mechanism, and not directly due to increased the sea surface temperatures over the Atlantic.
Figure 2. Number of days a named tropical storm was present in the Northern Hemisphere's Western Pacific Ocean for each year during November through April, 1945-2005.
Figure 3. Number of days a named tropical storm was present in the Northern Hemisphere's Eastern Pacific Ocean (off the coast of Mexico) for each year during November through April, 1949-2005
Comments will take a few seconds to appear.