Environmental Defense Critique of the Landsea Paper

Posted: 31-Jul-2006; Updated: 06-Nov-2006

Christopher Landsea (of the National Hurricane Center) and his colleagues argue in the July 2006 issue of Science that existing hurricane data are not reliable because the techniques for estimating hurricane intensity have changed over the years (see our summary or Science abstract). In particular, they maintain that pre-1990  methods for determining hurricane intensity probably tended to underestimate the actual intensity. As a result, they allege, these methods artificially produced the upward trends in hurricane strength found in the works of Emanuel and Webster et al.

Landsea et al. overlook several important points:

• Both Emanuel and Hoyos et al. found that the year-to-year variations of hurricane intensity were strongly correlated with the variations in sea surface temperature. Measurements of sea surface temperature and hurricane intensity are made independently. It is statistically very improbable that such a strong correlation would have been found if the variations in hurricane intensity were an artifact caused by measurement errors.
• In a 1993 paper Landsea argued that older methods tended to overestimate (not underestimate) hurricane intensity.
• In Emanuel’s first paper, which showed that the destructive potential of North Atlantic hurricanes has increased dramatically since the mid-1970s, Emanuel corrected for measurement biases using a method developed—ironically—by Dr. Landsea, and was criticized by Landsea for doing so.
• In the current Landsea paper, the authors provide only anecdotal evidence that pre-1990 measurement methods underestimated hurricane intensity.

All of the scientists studying the relationship between hurricanes and global warming have spent a considerable amount of time discussing data quality and the assumptions underlying their analyses. In ocean basins outside the Atlantic, there has not been a rigorous statistical analysis of exactly how intensity measurements may be influenced by the  various techniques that have been used over the years.

Fortunately, this sort of analysis is underway and the results are certain to shed light on this interesting issue. But even as this discussion continues, two essential points should be not be forgotten: 1. There is a very strong and undisputed relationship between hurricane intensity and sea surface temperatures, which are certain to continue rising because of global warming; and 2. Global warming has been shown to be a significant factor in producing the extremely warm ocean temperatures during the summer of 2005 that led to its record-breaking hurricane season.

Goldenberg SB, Landsea CW, Mestas-Nunez AM, et al.
The recent increase in Atlantic hurricane activity: Causes and implications, Science 293 (5529): 474-479 July 20, 2001.

Owens BF, Landsea CW
Assessing the skill of operational Atlantic seasonal tropical cyclone forecasts, Weather And Forecasting 18 (1): 45-54 February 2003.

Landsea CW
A Climatology of Intense (or Major) Atlantic Hurricanes, Monthly Weather Review 121 (6): 1703-1713 June 1993.