Wealth and climate change are increasing the cost of disaster cleanup

Political scientist Roger Pielke Jr. is right when he says that one reason we now spend more on disasters than we did in the past is that we are wealthier than we were in the past. But he's wrong in claiming that climate change has nothing to do with the increased cost. Physicist and climate scientist John P. Abraham explains a couple key flaws in Pielke's argument at The Huffington Post.

Notable Replies

  1. ~"Fivethirtyeight asked me to write a response to the article and subsequently decided not to run it."

    Why? Is Fivethirtyeight running out of internets to post stuff on? It's not like an on-line magazine is in any way limited in the number of articles it can publish, the way a print medium would be. One would think every article they run (presumably without any reward for the author in this instance!) will generate some traffic/ad revenue, so the decision most likely wasn't motivated by economic considerations. Of this sort.

  2. I don't see where Abraham is "attacking the messenger".

    You seem to be a little confused. Pielke does not directly cite any data on extreme weather events in his piece. He does cite the IPCC report saying: "medium evidence and high agreement that long-term trends in normalized losses have not been attributed to natural or anthropogenic climate change." This is not an example of what you suggest: this is not counting the numbers of tornadoes or hurricanes. Pielke does none of that in his piece.

    Instead, Pielke misinterprets that line from the IPCC to indicate that there has been no increase in extreme weather events correlating with increased average global temperature. Abraham points out that when you look directly at the incidence of extreme weather events -- as you yourself suggested but which Pielke did not bother to do -- there is, in fact, an increase of extreme weather events correlating with the increase in average global temperature. And Abraham cites a great many examples. Each example is a study -- a specific incidence of someone performing a test.

    Who is it that's trying to "silence" Pielke? No one that I can see. I do see Abraham rebutting his argument by pointing out that Pielke's interpretation of data leaves something to be desired. But fivethirtyeight didn't print that rebuttal so who is it getting silenced, exactly?

  3. Let's keep this simple. Pielke's argument consists of two parts:
    1) The cost of weather-related disasters is increasing due to increasing value of what is destroyed.
    2) There is no increase in frequency or intensity of weather-related disasters to contribute to the increase in cost of such disasters.

    Pielke does a fine job supporting point (1). But everyone is already acknowledging that (1) is true. No one thinks this is a particularly controversial point. Costs from weather-related disasters are largely driven by the value of what is destroyed.

    However, it's also completely possible that in addition to the increased cost due to higher property value etc. that there is also an increase in frequency or intensity of weather-related disasters.

    Pielke offers no evidence whatsoever that this is not the case. He asserts that this is not the case, but to support his point he cites the IPCC report saying something that sounds similar but does not mean the same thing. In contrast, Abraham points to numerous peer-reviewed studied finding that the frequency and/or intensity of weather-related disasters has increased.

    So you don't have to prove (1). That's taken care of. What I'm waiting for is some data on (2). Neither you nor Pielke have provided any.

    By the way:

    You're being pretty dishonest here. Abraham doesn't end on that hypothetical question. Here's the full statement:

    "Heat waves and heavy rains are not drivers of disaster costs? Just don't tell that to sufferers of floods from Irene, in Colorado, Duluth, Europe, or in the U.K., to name a few. Also, don't tell residents of France in 2003, Russia in 2010, Oklahoma and Texas in 2011, California this year, Australia, or just about any U.S. citizen in 2012."

    Many of the words in the part you left out are links directly detailing the costs of flood and drought-related disasters. Abraham is supporting his argument with data, not leaving a dangling rhetorical question. Cutting out Abraham's full response to this argument is a move right out of the Pielke playbook.

    Someone who likes fivethirtyeight but thinks Pielke is a poor choice for a science writer is well within his or her rights to petition fivethirtyeight to find a new science writer. Doing so isn't "silencing" Pielke any more than complaining to a manager about a particularly rude waiter. Even if he gets fired from fivethirtyeight, Pielke will still be free to write what he wants and publish anywhere that wants to print his stuff.

    Since fivethirtyeight is a private employer and the people petitioning fivethirtyeight are customers it seems like this is just capitalism in action -- customers giving feedback to a company about how it could better serve their needs. You're not against capitalism, are you?

  4. This gets sillier and sillier. Pielke's major source for stats on this is a Munich RE report. Pielke at one point argues that drought and floods have no significant impact on insurance costs for weather-related disasters. But look at page 6 from the Munich RE webinar that Pielke links to:


    Estimated insured losses due to thunderstorms, drought, and wildfire top losses due to tropical cyclones by about 20%. I guess for Pielke 60% is "not significant."

    And check page 7 for the frequency of extreme weather events. I guess I can see why Pielke wanted to keep his attention on the money and not on the data about frequency or strength.

  5. Neither you nor Pielke have provided any evidence concerning the actual frequency or intensity of storm events. Without doing so, you cannot conclude that there is no increase in the frequency or intensity of storm events. You seem to be contradicting your initial post here when you said:

    This is exactly the data I'm asking you (or Pielke) to provide to support Pielke's conclusion. It is, in fact, the evidence required to do so. Now you seem to be arguing that this is impossible to provide and even if it was provided it wouldn't prove anything. That's an interesting change of opinion in the space of a few comments.

    The Munich RE data Pielke uses to prove his (1) actually contradicts this. You can see clearly on page 6 of the Munich RE webinar presentation that there has been a steady increase in frequency of extreme weather events. Perhaps this is why Pielke used that strange word "spike" in his analysis -- his own dataset shows an increase but not a "spike". Then again, a "spike" is often statistically insignificant whereas the steady increase in the Munich RE dataset does look quite statistically significant. However, if you would like -- and I've asked for this three times now -- to provide some data to the contrary I would be happy to consider it.

    As it is, merely insisting Pielke demonstrated that there has been no increase in frequency or intensity of storms when he hasn't provided any evidence to bear on the question isn't going to impress anyone. It actually makes your argument look pretty bad, especially when you initially acknowledged that such evidence is required to make this determination and only now are backing away from that position, saying: "At this point in time, the data is scientifically insufficient but you may believe whatever you want."

    On the contrary. There is plenty of evidence out there about the frequency and intensity of weather events and how they change over time. Try here, here, and here. I'm not going to spend all day doing your research for you, though, so if you have some studies that don't show any trend then please link them instead of just insisting over and over that Pielke demonstrated his claim without providing any evidence for it.

    But you don't have to take my word on it. Here's Nate Silver talking about the shortcomings of Pielke's article:

    As I mentioned, the central thesis of Roger’s article concerns the economic costs associated with natural disasters. But we also allowed a number of peripheral claims into the piece. For instance, Roger made a number of references to the overall incidence of natural disasters, as opposed to their economic cost.

    We think many of these claims have support in the scientific literature, specifically including the 2013 IPCC report. But there is a range of debate among experts about others. Either way, these claims shouldn’t have been included in the story as offhand remarks. We should either have addressed them in more detail or scrubbed them from the article.

    ...Furthermore, there was some loose language in the article. We pride ourselves on precise, matter-of-fact language. These things reflect a poor job of editing on our part.

    As you can see, Silver's own reservations about the article match my complaints exactly.

    That's not really true. The warmest 10 years on record have all occurred since 1998. The last 15 years is the hottest 15 year period on record. The "flattened" global temps you mention are the result of starting a trendline on 1998 which is the third-hottest year on record. The hottest and second hottest are 2010 and 2005 respectively. If you cherry pick your trendlines you can make them go down but you can't make the last 15 years anything but the hottest 15 years in the history of human beings measuring temperature. Please take a look at that last link. It will make it pretty clear how bullshit this "flattening" argument is. Someone told me the link isn't working for them so here's the URL: http://americablog.com/wp-content/uploads/2013/05/skeptscience-realistclimateXX.gif

    No, the fact that he claims weather events have not increased in frequency or intensity when his own dataset contradicts that claim is what makes him dishonest. And some other stuff you can find easily enough by googling "pielke dishonesty".

    You seem to be misunderstanding the argument here. Pielke quotes IPCC saying "attributable signal" and he uses this as evidence that there is no signal at all. That's not valid reasonsing. It's changing the goalposts.

    You know, I don't like to get into credentials in these sorts of discussions because it amounts to an argument from authority. However, Abraham has a PhD in mechanical engineering and specializes in thermodynamics and fluid flow whereas Pielke has a PhD in political science. Which is more relevant to climate science (the science itself, not subsequent policy discussions) is left as an exercise for the reader.

    Well I'm not a "fan" of such efforts either, but I also don't have any particular problem if the readers of a publication don't like one of the writers and try to draw management's attention to this fact. Pielke has no constitutional right to a job at fivethirtyeight.com but he does have a constitutional right to free speech. If he loses his job he is not "silenced" because he still has that right to free speech and never had a right to the job in the first place. I think your loose use of the term "silenced" is actually pretty insulting to people living under political regimes where they are legitimately silenced by government force.

Continue the discussion bbs.boingboing.net

12 more replies