16 June 2010

Climate Change Catnip

At the Washington Post's Capitol Weather Gang, Andrew Freedman grapples with how to discuss climate change in the context of flash floods over the past few weeks:
. . . the question of whether to raise climate change in discussions of flash floods (and other extreme events) constitutes more than a quibble over semantics. The media has a responsibility to report what the science says, even in the context of a breaking news story, such as a flood event or heat wave. The science has become clearer, although by no means certain, that local precipitation extremes may be connected to climate change. Yet, to date, the mainstream media has shied away from raising climate change in extreme event coverage. This is unfortunate, because it constitutes a missed opportunity to make climate change relevant to people in the here and now, rather than an abstract concept in the distant future.
In contrast, Andy Revkin who blogs at the New York Times suggests caution in making connections between a few events and larger climatic patterns.

As the graph above shows (from data of the NWS), there is no evidence for an increase in flood disasters. In fact, there has been a marked decrease. I have also shown on numerous occasions(e.g., PDF) that there is no evidence of an increase in flood disasters in terms of economic damage either, once adjusted for growing wealth at risk.

So what is the thing for journalists to say about climate change and recent flood disasters? Easy. There is presently no evidence for a signal of climate change (human-caused or otherwise) leading to an increase in flood disasters. If there is any signal, it is far too small to see and it will take many decades for such a signal to emerge.

It seems like it would be easy and straightforward to simply say what the science shows, but making climate change connections with disasters seems to be like catnip for journalists and advocates alike.


  1. Roger, you're not making a fair comparison here. My post was not discussing trends in "flood disasters" or fatalities. In fact, I never used the word "disaster" in my post at all. The decrease in fatalities may be related to better societal preparedness and warning capabilities.

    Rather, my post (as well as Andy Revkin's) discusses trends in extreme precip events, as identified by the CCSP report and the Climate Impacts Report. Both reports, along with the IPCC, have identified a trend towards more frequent heavy rainfall events. Whether or not that leads to increased fatalities or "disasters" is a different matter.

  2. Actually, the link embedded in the quote refers to two different meteorological phenomena that might be connected to AGW; increased daily rainfall in the UK, leading to local flooding; and 'extreme' events such as typhoons. It doesn't explicitly say so, but from context I think the link means increased average daily rainfall. It is not clear to me the link says what Freedman says it says; the extreme local precipitation (other than that connected to typhoons, etc.) is connected to climate change.

  3. Gerard: There are other links embedded in the actual post that address that: http://www.climatescience.gov/Library/sap/sap3-3/final-report/#chapters and http://www.globalchange.gov/publications/reports/scientific-assessments/us-impacts/full-report .

  4. Increased flooding would be a problem only if it leads to losses (lives, property or whatever). The 'potential impact' of some consequence of climate change (ignoring adaptation, economic progress and technical change) is just a theoretical indication. The actual impact is the 'residual impact' in IPCC jargon, i.e. the remaining damage (or benefit) after all those factors, especially adaptation in the broad sense, are taken into account.
    However, the overall conclusion of IPCC about precipitation, hurricanes and typhoons is NOT that they would increase in frequency or strength. Hurricanes, for instance, are supposed to DIMINISH in frequency, though a small increase in average strength is also envisaged. As Roger has shown elsewhere, this does not translate into more damage per unit of people or property.

  5. Then perhaps there is an upsurge in reporting of flood disasters?
    Or an upsurge in drawing fatuous conclusions about newsworthy weather events?

  6. -1-Andrew

    Thanks for the comment.

    However, I'll point out that the second sentence of your post refers to "deadly" flash floods, and indeed these are those that ar newsworthy (tell me the last time the WP did a news story on an "extreme" 50 mm rainfall with no deaths or damage;-)

    The CCSP report you cite explains that there have yet to be documented trends in streamflow at the highest levels (i.e., floods).

    Equating projections from climate models to 2100 with deadly events in 2010 is simply misleading in my view.

  7. I took a look at the raw data given in Andrew's first link about high rainfall events, and it's pretty thin. For example, in the central US (see graph on p. 47), where the latest batch of floods has been occurring, a linear fit indeed seems to show an increase from ~0.23 days/year to ~0.29 days per year in days in the top 0.3% of rain in the US over a 113 year period, but the scatter is huge. Given changes in methods of measurement, etc., over that period, I wouldn't hang a cat based on what it shows.

    Writing as someone who spends his life searching for signals in noisy data, I'd have to say I couldn't publish stuff like this in my field. I think a considerable degree of skepticism is warranted.

  8. In regards to the quote, "Yet, to date, the mainstream media has shied away from raising climate change in extreme event coverage": I have never heard the mainstream news say that extreme weather was NOT related to climate change, or even downplay that angle. They always manage to get the words "climate change" or "global warming" in the story somehow, or remind us that it's "consistent with" climate change science without actually spelling out what the connection is. And unless we search out specialists like Roger and others, we're left with the feeling that global warming is connected in a causal way with current weather events. At this point, with 1/2 a degree or so of total warming over 100 years, I don't see how anyone can postulate a link at all.


  9. Give the warmists a break, fer gosh sakes. They've seen almost all of their icons toppled the last few years. Hurricanes. Polar bears. Arctic ice. Phil Jones. Is there any harm in letting them believe that a few flash floods were produced by man's sinful ways?

  10. Bill: Can you please send links to stories about the Oklahoma City event that mention climate change? Same for Arkansas please. Thanks.


  11. "The science has become clearer, although by no means certain, that local precipitation extremes may be connected to climate change."

    What a ridiculous statement. Of course local weather is linked to climate. That goes for the medians as well as the extremes.

  12. -10-Andrew

    I just did a Google News search of "Arkansas Flood Climate Change" and a bunch of stories came up, yours was at the top;-)

    Same for Oklahoma ...

  13. Mike M.,

    Apparently you stopped paying attention to Arctic ice in April. Currently, Arctic ice area and extent are at or near record lows. The probability that the ice extent minimum will exceed 2009 is getting smaller every day and the prospect of a new record low minimum is not out of the question any more.

  14. Re Andrew #10:

    I checked on both Oklahoma and Tennessee flooding, along with the word climate. Newspaper stories so far (haven't found any in video media yet--ABC, NBC, CNN, etc.), and I think the last article is yours (and you are definitely more nuanced than the others, so props to you for that):

    "Scientists blame the unusual weather patterns from DC snow storms over the winter, to increased tornadoes across the Midwest, and freak flood-causing rainfall--on climate change."

    "The devastating rain and flooding that recently paralyzed Nashville is just the sort of extreme precipitation event that global warming is expected to make more common."

    "With precipitation extremes already increasing in frequency and severity due to global climate change, it is especially important that journalists and the public learn how to discuss the climate context of such extreme weather events, provided such discussions stick to the scientific evidence."

    Unfortunately I am leaving on vacation shortly, but will try to follow up on this tomorrow.


  15. Freedman's post spells out his motives quite well. He bemoans "a missed opportunity to make climate change relevant to people in the here and now, rather than an abstract concept in the distant future."

    Translated into plain English, Freedman wants journalists to use deadly floods to scare people about global warming.

    Does WaPo endorse this kind of agitprop among its journalists?

  16. Oh, I think the arctic will do just fine this year, De Witt...


  17. "Yet, to date, the mainstream media has shied away from raising climate change in extreme event coverage":"

    This isn't being covered by national media at all.

    "Never in recorded weather history has Seattle gone this far into a year without hitting 75 degrees."

  18. Roger: I meant other media than my post on WashPo. Sheesh. Also, I specifically said "mainstream media" which connotes major publications. I should have been clearer in specifying major daily newspapers, broadcast networks, and cable news. If you read the articles that come up in a google news search for such terms you may find that many of them don't talk about climate change at all, which is what I found when researching my post.

    Bill: Thx for the links! I don't see any news source in your list that qualifies, in my mind, as a major mainstream media outlet. But I was not sufficiently clear (as I noted just now to Roger) in what I meant by "mainstream media" in my blog post. And I think it's probably pointless to debate that now anyway.

  19. Using deaths from flooding as an argument for or against climate change cannot be done without discussing improvements in weather forecasting, flood control and more. We can call it a Pielke error, equivalent to discussing hurricane damages without accounting for the costs of hardening houses, the investment in prediction and observation, etc.

  20. Andrew Freedman is a propagandist looking to further his cause, not a journalist.

    I first came across him at the Weather Channel when Heidi Cullen was running her bully pulpit there.

    Andrew would pump out wildly one-sided articles and then I would drop by to present, "the rest of the story" that Andrew somehow missed in his relentless "research".

  21. increased daily rainfall in the UK, leading to local flooding;

    Nothing at all to do with less uncovered land for run-off, more tightly constricted rivers or urban expansion now forcing people onto flood plains??

    I would suggest that all these things are more important to the amount of flooding in UK than the pathetic amount of extra rain a degree or two will bring.

  22. Re: Andrew #18:

    I only had about 10 minutes yesterday to do the first search, so I wasn't very in depth. There were a lot of hits, but I tried t stay away from the obviously smaller outlets and re-posters. I did a Google News search this morning and didn't find anything, so I went to CBS and ABC and searched them directly. In 45 minutes I found perhaps a dozen articles total in the past 2 years, so it doesn't appear that my hyperbole from my first post is accurate. My apologies to Andrew for stirring the pot needlessly.


  23. -19-EliRabbit

    You and I agree.

    As we wrote in the Hohenkammer workshop consensus, f you want to document climate change, look at climate data, not socio-economic data. The lack of a climate change signal in socio-economic data does not imply no climate change, but you know this.

  24. Mike M. #16

    How is the fact that there's 1.7 Mm2 less ice area on 6/11/2010 than on 6/11/1980, a drop of nearly 20% and a thirty year record low, supposed to convince me or anyone else that the ice will do just fine?

  25. Eli
    "...equivalent to discussing hurricane damages without accounting for the costs of hardening houses, the investment in prediction and observation, etc."
    ....Or perhaps even discussing the decreasing trend in the frequency of landfalling hurricanes. Coincidentally (or not) that was the only strong prediction from Knutsons latest modeling effort, everything else being a 50/50 shot, despite Emanuels alarmist interpretation of it.

    Anyway all supposed trends in extreme weather must have been found by chopping short the record because all the long-term trends are flat. The cherry-picking in this area is rife and crude.