30 November 2010

Why There are No Trends in Normalized Hurricane Losses

The graph above shows data on normalized US hurricane losses 1900 to 2009 and was presented in a talk I gave today.  Why is there no trend in the data?  The two graphs below explain why.  You can do the math.

There are no trends in normalized damage since 1900 because there are no trends in either hurricane landfall frequency (data from NOAA) or intensity (data from Chris Landsea through 2006) over that same period (but rather, a very slight decline in both cases).  If our normalization were to show a trend then it would actually have some sort of bias in it.  It does not, thus we can have confidence that the societal factors are well accounted for in the normalization methodology.

4 comments:

  1. Roger,

    Given the spike in PDI since 2003, when do you expect that we might see an upward trend in damages? Or is the 2003-2006 spike an anomaly?

    ReplyDelete
  2. -1-Marlowe Johnson

    Thanks for the question. We have a paper on exactly this question under review at present, stay tuned.

    ReplyDelete
  3. But the IPCC Bible says, in its SPM,
    "There is observational evidence for an increase in intense tropical cyclone activity in the North Atlantic since about 1970, correlated with increases of tropical sea surface temperatures".
    And behold, in chapter 3, the Gospel according to St Jones, verse 3.8.3, it is written, that the North Atlantic PDI index "showed substantial upward trends", and that the trends found by wonderful counselor Emanuel and Webster et al "appear to be robust".

    ReplyDelete
  4. Sorry, I should have read your previous post before commenting, and described chapter 3 as the Gospel according to St Kevin and St Philip.

    Perhaps a reminder of why Chris Landsea resigned from the IPCC might be timely.

    ReplyDelete