Our paper entitled "Global warming and United States landfalling hurricanes" was published in Geophysical Research Letters on January 23, 2008. After publishing, the paper was covered by more than 300 newspapers worldwide plus many TV and radio shows. The report by The Associated Press and some media includes the following statement:"Critics said Wang's study is based on poor data that was rejected by scientists on the Nobel Prize-winning Intergovernmental Panel on Climate Change. They said that at times, only one in 10 North Atlantic hurricanes hit the U.S. coast, and the data reflect only a small percentage of storms around the globe. Hurricanes hitting land "are not a reliable record" for how hurricanes have changed, said Kevin Trenberth, climate analysis chief for the National Center for Atmospheric Research in Boulder, Colo."
We think that the above statement is misleading. First, we cannot find that the report of the Intergovernmental Panel on Climate Change (IPCC) rejected the data of U.S. landfalling hurricanes. Second, it should be pointed out that U.S. landfalling hurricanes, which were used in the peer-reviewed literature (e.g., Landsea 2005; Parisi and Lund 2008), account for about one third of the total North Atlantic hurricanes instead of "one in 10". U.S. landfalling hurricanes are about 5% of all global ocean basin hurricanes (i.e., hurricanes in the North Atlantic, Eastern Pacific, Northwest Pacific, South Pacific, and the Northern and Southern Indian Oceans). Because of the relatively small number, one may argue that U.S. landfalling hurricanes are not reliable indicators of hurricane activity or say that the data of U.S. landfalling hurricanes is "poor data". However, we have to keep in mind that other hurricane data have serious problems. Hurricanes were undercounted before the era of aircraft reconnaissance (around the mid-1940s) and satellite technology (the mid-1960s) since hurricanes over the open ocean during that time can be rarely detected. These hurricane data cannot be used for indicating how hurricanes have changed over the long term, without taking into account the potential missed hurricanes over the open ocean. In contrast, by far the most reliable Atlantic hurricane measurement over the long term (for example, from the early 20th century to the present) is the number of U.S. landfalling hurricanes since it excludes hurricanes over the open ocean (which never make a landfall) that can be hardly detected before aircraft reconnaissance and satellite eras. Therefore, we conclude that U.S. landfalling hurricanes are most reliable data (in contrast to "poor data" or "are not a reliable record") for studying Atlantic hurricanes over the long term. In addition, the most social impact and importance are hurricanes making landfalls. We thus believe that using U.S. landfalling hurricanes is a proper choice for our study.
Landsea, C. W., 2005: Hurricanes and global warming. Nature, 438, E11-E13.
Parisi, F., and R. Lund, 2008: Return periods of continental U.S. hurricanes. J. Climate, 21, 403-410.