24 September 2012

Drought and Climate Change

The amount of nonsense in public debate on extreme events and climate change remains at a high level. This is great news for me because it provides plenty of opportunities to discuss what the actual science says and how we think we know what we know. As I have long argued, accurate representations of the state of science of extremes is far more important as a matter of scientific integrity than to the hyper-politiczed debate over climate change. Put another way, it is highly unlikely that misrepresentations of the state of science will do much to move action on energy policies, but they could damage the integrity of leading institutions of science.

This is a short post about drought, which simply summarizes the bottom-line conclusions of two of the most recent major scientific assessments of extreme events and climate change, one by the US government, released in 2008 under the Bush administration (PDF, and then reaffirmed in the CCSP Unified Synthesis under the Obama Administration, here) and the second from the IPCC.

First, from the US government's assessment of extreme events in the US, here is what it concluded about drought (here in PDF, at p. 42):
The most widespread and severe drought conditions occurred in the 1930s and 1950s (Andreadis et al., 2005). The early 2000s were also characterized by severe droughts in some areas, notably in the western United States. When averaged across the entire United States (Figure 2.6), there is no clear tendency for a trend based on the PDSI. Similarly, long-term trends (1925-2003) of hydrologic droughts based on model derived soil moisture and runoff show that droughts have, for the most part, become shorter, less frequent, and cover a smaller portion of the U. S. over the last century (Andreadis and Lettenmaier, 2006). The main exception is the Southwest and parts of the interior of the West, where increased temperature has led to rising drought trends (Groisman et al., 2004; Andreadis and Lettenmaier, 2006). The trends averaged over all of North America since 1950 (Figure 2.6) are similar to U.S. trends for the same period, indicating no overall trend.
Got that? Over the climate time scales "droughts have, for the most part, become shorter, less frequent, and cover a smaller portion of the U. S. over the last century." At the top of this post is Figure 2.6 from that report, showing drought incidence in the US (red) and North America (blue) from 1900.
The IPCC in 2012 conducted a survey of drought globally, and concluded with the following (here in PDF, at p. 171):
There is not enough evidence at present to suggest high confidence in observed trends in dryness due to lack of direct observations, some geographical inconsistencies in the trends, and some dependencies of inferred trends on the index choice. There is medium confidence that since the 1950s some regions of the world have experienced more intense and longer droughts (e.g., southern Europe, west Africa) but also opposite trends exist in other regions (e.g., central North America, northwestern Australia).
Got that? Some places have become dryer, others wetter, and not much confidence in asserting the presence of any trends at the global scale.

Now it is of course true that the recent assessments of the US government and IPCC are not the last words on these subjects. They represents attempts by governments to have scientists systematically arbitrate questions that can resolved empirically, such as "has drought increased or decreased?"

Anyone who shows up in public debate (or the scientific literature) with an alternative view (e.g., droughts have become worse in the US or there is more certainty at the global level) had better come with some strong evidence. The need for strong evidence comes not simply from the authority of these assessments, but because they represent a condensation of a large amount of literature. To overturn the conventional wisdom expressed in the assessments requires overturning the arguments found in the literature on which the assessment is based. The same logic goes for people who would challenge other findings from such assessments, such as in making a claim that the globe is not warming.

The findings of assessments are not however received truth. As we have seen, those leading and participating in such assessments can err egregiously -- where an error means a failure to accurately reflect in the assessment the underlying scientific literature. Here overturning the assessment is relatively easy, as it does not require over turning the literature, but simply representing it accurately. This is why the IPCC SREX came to different conclusions on extremes than did AR4.

Unfortunately, there are no shortcuts here to reliable knowledge - just because the IPCC erred on one subject does not mean that it necessarily erred on others. Each knowledge claim must be evaluated on its merits. Shortfalls in an assessment process may cause you (or me) to lose confidence in the integrity of that process, but it does not eliminate the need to evaluate knowledge claims on their merits using the methods of science.

The good news is that for extreme events and climate change there is a well-developed scientific literature on this topic, and it does not always jibe with what you might read in the news or hear from experts. Given the politicization of the climate debate, and the penetration of that politicization into academia, I don't expect this to change anytime soon, if ever.

As always -- Caveat Emptor!


  1. "- just because the IPCC erred on one subject does not mean that it necessarily erred on others. Each knowledge claim must be evaluated on its merits."

    That's a nice, idealistic position to take. Now, if you were being tried for a very serious crime, and the prosecution was caught falsifying one piece of evidence, would you want your defense lawyer to say the same thing?

    Unlike you, I find it perfectly reasonable to question the entire case, based on individual failures. We know that people will bend evidence when inspired by belief - this is a trivial fact. The assumption that scientists are naturally objective seekers of truth is naive in the extreme. Stephen J. Gould made a minor career out of pointing out the ideological tilting of scientists of the past.

    The IPCC is a human effort, and we should expect to find human failings. It is not 'science' that is writing the IPCC reports; it is small groups of individual scientists. We now know that when you shine a light on individual claims, then not only fail in accuracy, but those responsible refuse to admit error when it is pointed out to them in plain English.

    Worse still, when others involved in the greater project see this behavior, they stay silent. Why? Certainly not out of love of the truth. Every individual involved in writing any version of the IPCC reports who remains silent while shenanigans go on invites an assumption of the worst where their own work is involved. An honest person, doing honest work, does not countenance dishonesty or skullduggery. When one does so, one loses the assumption of good will.

  2. Boy, is all that fence -sitting getting uncomfortable for you. All the outright untruths from fellow scientists must be very annoying. This is as close as you have been in calling liars -the warmist group of so-called scientists-those who continue to prove that money corrupts science.

  3. -1- Mark B,

    I agree with Roger that each claim requires its own proof. But given the number of "problems" I won't be very interested in hearing from the IPCC about any of those claims.

    I guess I'm just a bit skeptical about their reports.

  4. Good piece, but 2008 was still the Bush administration, not Obama's as you said.

  5. -4-Jeff

    Thanks, you are right of course, but the report was subsequently re-reviewed by the Obama Administration as part of the CCSP synthesis. I will update to make clear.


  6. just because the IPCC erred on one subject does not mean that it necessarily erred on others.

    True, but logic suggests that genuine errors would be randomly spread. Are there any examples in which the IPCC erred on the less alarming side? (Or do those errors get found and fixed up almost immediately?)

    And the reaction to the pointing out of errors seems asymmetric as well. Does the head of the IPCC accuse people suggesting the reports are insufficiently alarmist of conducting "voodoo science"? Or do only sceptics get such treatment.

  7. Now... All the IPCC has to do is copy your post, and these reports, into their newest assessment, and then say...

    "... but that was before the drought this year, which was notable".

    They then can just leave things like that, iarguing the two statements are completely independent of one another, with no implications therein. And then all would still be true, and worthy of publication within IPCC manuscripts.

  8. Politics is (sometimes at least) about "sides". Science is about hypotheses and evidence. That's why, as Roger says, each knowledge claim must be evaluated on its merits. The climate change issue has become so politicized that even scientifically literate people forget this.

    If climate science were really about a single overwhelmingly relevant hypothesis, then the debate between the "sides" might be a somewhat reasonable proxy for the debate about the science. But it's not.

  9. Roger is so interested in sports that he has a second blog to cover it. I generally stay away from sports analogies, but here you go. For some years in baseball, there were whispers about particular players and steroids. A very common comeback to them was the indignant "How dare you accuse a man without proof? - Harrumph harrumph!

    Then, more and more players got caught, and it was generally the exact players who had been suspected.

    Given that the player's union fought tooth and nail against testing, and that every caught player had indignantly denied the accusation in the past, a new assumption has set in. it is the 'reasonable to assume guilt until proven innocent' assumption. I see no problem with taking this line. Given the player's union's persistent actions to block testing, and the refusal of most players to speak out against steroid cheats, the burden has shifted on to the players. They HAD an assumption of innocence, and they, as a group, abused it.

    In the case of the IPCC, I think reasonable people can agree that there are going to be errors, and that reasonable people may disagree on the analysis of evidence. But when someone looks at 'A' and finds flim-flammery, and someone else looks at 'B' and finds the same - and no one seems to care to correct the record - then I think a reasonable person can decide that 'C,' 'D' and 'E' can no longer be trusted. At some point, the 'fool me once' principle comes into play.

    This sort of thing doesn't prove that the rest of the IPCC document is baloney, but it does invite REASONABLE skepticism. If I signed my name on the damn thing, and someone else in another section was shown to be fiddling the numbers, I'd be quick to separate my own reputation from their escapades. Aren't there supposed to be 2500 scientists involved in assembling this report? If there isn't a single one who will speak out - even for purely selfish reasons - how can I pretend that we're in an ideal world and each statement should not only stand on its own, but be trusted?

  10. Awesome and succinct.

    Did find a typo, though.

    "loose confidence" should be "lose confidence"

    Thanks for your efforts!

  11. Leaving aside hammering the IPCC for a moment - not that it is not playing politics to a fare thee well - let us consider the blatant misrepresentation of what the IPCC actually does say and reflect on the context that reveals : partisan lying and blatant misrepresentation.
    But if I suggest that the recipient of global taxation is not the agent to be entrusted with blind confidence - that is a 'conspiracy theory' rather than common sense.

  12. -11-Cartoonasaurus

    Fixed! Eagle eye appreciated, thx;-)