Our hurricane dialog never seems to end, and hardly a week goes by without another article appearing in a major journal on the subject of global warming and hurricane activity. In recent weeks, two more major articles have been published adding to the overwhelming evidence that the hurricane – global warming link cannot be supported on theoretical or empirical grounds. These two articles represent more nails for the coffin containing the popularized link between global warming and hurricanes!
Despite the evidence that we have presented over and over, thousands of websites continue to lead us to believe that global warming will increase hurricane frequency and intensity. As we all know, any intense hurricane appearing anywhere on the planet is viewed as further evidence of our impact on the climate system. All self-respecting documentaries on global warming present the destruction caused by Katrina, link the intensity of the storm to warm sea surface temperatures, and then blame greenhouse gas emissions (particularly from the United States) on the entire mess. Al Gore’s blockbuster film was all over this one – the poster for the movie even shows a hurricane pattern in the emission from a smokestack!
The first article this week appears in Geophysical Research Letters and was produced by three scientists in the Department of Atmospheric Science at Creighton University; the work was funded by the National Oceanic and Atmospheric Administration. Englehart et al. focused on tropical cyclones (TCs) in the eastern Pacific (EastPac) off the coast of Mexico. They note “The lack of attention to EastPac storms may partly reflect the fact that the basin’s TC activity tends to move quickly away from land and population.” However, when these storms strike land, they tend to produce huge amounts of precipitation and catastrophic flooding due to the mountainous coastline of Mexico. Some of these storms occasionally move through the Gulf of California and enter Arizona causing flooding in places like Tucson and Phoenix. These storms can even cross the Pacific and make landfall over the Hawaiian Islands. While not nearly as well studied as events in the North Atlantic, these hurricanes in the Pacific Ocean are still important in the context of climate change and tropical cyclone activity.
Englehart et al. gathered data on hurricanes in this region from the National Center for Atmospheric Research, Mexico’s Servicio Meteorológico Nacional, the Mazatlan Observatory, and the National Hurricane Center. In describing the graph below (Figure 1) of their compilation, they note “From this graph it is apparent that long-term TC frequency exhibits a significant (p = 0.05) negative trend (r = -0.30).” Folks, they are becoming less frequent, and the trend is highly statistically significant. They then present the data for early (May to July) and late season (August to November) hurricane activity and find that the downward trend is prominent in the August to November period. They comment that “The fact that a significant trend (p = 0.01) is evident only in the late season provides one (albeit indirect) indication that the apparent decrease in near-shore activity is not substantially conditioned by data and measurement issues.”
Figure 1. Hurricane frequency in the near-shore domain (from Englehart et al., 2008)
Of course, the many in the global warming crowd would argue that storms are becoming more intense, and maybe not more numerous. Well, Englehart et al. present a plot (Figure 2) with maximum wind speeds and minimum sea level pressures. Sorry to disappoint, but there are no upward trends in the data, and if we really want to pour salt in the wounds, the maximum wind speed plot shows a bit of a downward trend!
Figure 2. Time series plot of (a) maximum wind speed (meters per second) and (b) minimum sea level pressure (millibars) associated with hurricane activity in the near-shore domain (from Englehart et al., 2008).
A second study was published in the Journal of Climate by Francis Parisi of Standard and Poor’s and Robert Lund of Clemson University’s Department of Mathematical Sciences. They gathered data on landfalling hurricanes in the United States from 1900 to 2006 (see Figure 3 below). They state “We select 1900 as the study starting date to ensure storm count accuracy. While some work aims to improve accuracy for U.S. hurricane data and adjust counts for undetected storms, the 1900–2006 data are generally considered reliable. To elaborate, towns on the coast were likely dense enough to avoid missing (undercounting) landfalling hurricanes after 1900.” They calculated the return intervals for hurricanes of different strengths using extreme value methods and Poisson regression techniques. They successfully determined the return intervals, and in their discussion of the results, they note “Despite the recent active 2004 and 2005 hurricane seasons, the authors do not find evidence of an increasing trend in hurricane strike frequencies” and “The hypothesis that hurricane strike frequencies are increasing in time is also statistically rejected.”
Figure 3. Annual U.S. landfalling hurricane counts, 1900–2006 counting multiple landfalls as
separate storms (from Parisi and Lund, 2008).
There may not be any pillar of the greenhouse scare that is so consistently unsupported by so much published research. We have kidded about renaming World Climate Report to the World Hurricane Report. Stay tuned – we are confident that we have not seen the end of articles breaking the global warming advocates’ claim that hurricanes are increasing in frequency and/or intensity.
Englehart, P. J., M. D. Lewis, and A. V. Douglas. 2008. Defining the frequency of near-shore tropical cyclone activity in the eastern North Pacific from historical surface observations (1921–2005). Geophysical Research Letters, 35, L03706, doi:10.1029/2007GL032546.
Parisi, F. and R. Lund. 2008. Return periods of continental U.S. hurricanes. Journal of Climate, 18, 403-410.