Archive for the ‘Raging Logical Positivism’ Category
Via John Fleck, I found this opinion piece by Bjorn Lomborg on climate change, water, and adaptation in Bangladesh. The main thrust of the article is that if the goal is to reduce the harmful consequences of anthropogenic climate change, resources would be better spent on projects to improve access to the basic necessities of the poor and the developing world in general, rather than on reducing carbon emissions.
Not that it should matter what I think, but in this instance I would agree. I think that improving basic access to clean water, sanitation, health care and nutrition, and infrastructure in the developing world addresses a host of interrelated ethical and moral, social, economic, public health, energy, national security, and environmental issues. I think such resources would be well-spent and substantially enhance the adaptive capacity of countries in the face of future climate change. This is not an either/or proposition, however, — the sense one inevitably gets from Lomborg’s writing and those that subscribe to his particular tactics and approach — increasing adaptive capacity, reducing vulnerability, and reducing emission have to proceed together. There is no one single solution nor policy that addresses the myriad current and future challenges that human populations will face from anthropogenic climate change.
So far, so good. But then Lomborg writes the following rather incredible sentence,
Cutting carbon emissions will likely increase water scarcity, because global warming is expected to increase average rainfall levels around the world.
There is a technical term for this type of statement. Dangerously misleading. And it is misleading in important ways that conceal one of the very real and likely imminent challenges we face as a consequence of anthropogenic climate change and is directly relevant to policy. I’ll maintain that if you get the science wrong, you reduce your chances of developing effective adaptation and mitigation policies.
Consider the Intergovernmental Panel on Climate Change‘s Fourth Assessement Report (AR4). Figure 3.3 from the IPCC AR4 Synthesis Report shows the precipitation predictions from the multi-model averages based on the SRES A1B scenario — what this means is that it is the average result of over 20 different climate models running one of the mid-range emissions scenarios.
Warm colors indicate drier conditions by 2099, cooler colors are wetter. The left panel is December through February, while the right panel is Northern Hemisphere summer.
What immediately jumps out at you will be the spatial patterns. Regions in the subtropics, including parts of the southwestern United States and much of Mexico and Central America, southern South America, North Africa and the Mediterranean, parts of southwest Asia, and south Africa and Australia show future dry conditions, while the equatorial tropics and the high latitudes get wetter. There are also important seasonal patterns. South and southeast Asia (from where Lomborg reports for the Wall Street Journal), for instance, show a drier future dry season and a wetter monsoon season. The Amazon is projected to have a dry June-August, but wetter although less change in boreal winter.
Finally, notice the stippling. Whereas areas with colored shading indicate 66% of the models agree on the sign (not magnitude) of the change, stippling indicates where 90% of the models agree on the sign. Two of the most robust projections of the model are the high latitude increase in precipitation, and the extratropical drying. Note that there is less model certainty in the monsoon region, the setting of Lomborg’s opinion piece.
This latter phenomenon is a robust projection of the suite of global climate models now available, and is colloquially known by climatologists as ‘the rich get richer, the poor get poorer‘ — that is, wet tropical and high latitude areas get more rain, already semi arid regions receive less in the future.
Let’s return to Lomborg’s statement then that,
Cutting carbon emissions will likely increase water scarcity, because global warming is expected to increase average rainfall levels around the world.
In fact, some of the most vulnerable regions of the world will experience severe reductions in precipitation as a consequence of carbon emissions. Lomborg’s statement is partially backwards and partially not useful — not cutting carbon emission will likely increase water scarcity (in some important regions!), and moreover is likely to do so in the places least likely to be able to adapt to these changes. This isn’t a statement from me of a preference for adaptation or mitigation (both are needed), this is a situation where not understanding or misstating the science could lead to the wrong policy prescriptions.
Returning to monsoon Asia — the setting of Lomborg’s exhortation for adaptation set in contrast to the efficacy of emissions reduction — uncertainty about the future course of the onset, strength, and intensity of the monsoon still reigns. While the AR4 models above suggest wetter summer monsoons (but dry winter monsoons), there is less certainty and more spatial variability. And regional climate models potentially give a different picture. Recently, researchers from Purdue University found that, in their climate model experiments, anthropogenic climate change ‘resulted in overall suppression of summer precipitation, a delay in monsoon onset, and an increase in the occurrence of monsoon break periods’
So, robust projections from climate models project future subtropical drying, including many areas with considerable vulnerability to climate change and water scarcity. For other regions, including monsoon Asia, uncertainty is still high.
I maintain that science matters for making sensible climate policy. If you get the science wrong, you might still accidentally stumble your way into good climate change policy, but adaptation and mitigation policies are more likely to be successful if they address accurately the most likely sources of vulnerability (fundamentals, like will a region get wetter or drier?). Therefore, undermining or mischaracterizing science in the name of one’s own ideological or policy goals seems likely to, in fact, undermine the chances of success of prescriptions for dealing with the consequences of anthropogenic climate change. Clean drinking water, sanitation, access to health care — all of these and more are good things irrespective of future climate change. The scientific details — in space and time and the sign and magnitude of uncertainty — matter, and no one is served well by getting them wrong.
UPDATE: I see Daniel Collins was thinking along similar lines.
The few comments this blog has received thus far have been largely positive and generally inquisitive. But I want to establish from the outset that I have no interest in letting the comment section of this blog resemble the ‘Lord of the Flies’. As I related to commentator Eric from Climate Audit in the thread below,
I have no intention of allowing comments here to become the platform for hasty or ill-informed mudslinging (well-informed and creative mudslinging, perhaps … ). Useful technical issues, questions, and comments all posed with a modicum of politeness and genuine interest are welcome and indeed encouraged– if you want to denigrate working scientists, you’re going to have to find another venue which tolerates such things, sorry.
And this is pretty much how things are going to go. I will not, from now on, publish off-topic or mean-spirited comments. Full stop.
If you want to participate in the technical discussion of the science here, I welcome you to jump in. If you’ve got questions — even very basic ones — I encourage you to speak up. But I will not create a haven or venue for attacks on working scientists or other commentors who have a genuine interest in the science. I will endeavor to maintain the future tone and content, from both myself and any commentors, focused on the science.
Other things I will take a dim view of and I feel will lower the level of inquiry include: concern trolling, appeal to false authority, begging the question (in the proper sense), lies, and soft focus photos of unicorns.
Science is cool. Lets not lose track of that in the race to fill the internet with bile.
As I mentioned in my previous post, the Khadyta River chronology appears to suffer from the classic signs of what dendrochronologists have come to call ‘the divergence problem‘. Before I continue, however, I should emphasize a few points.
 I didn’t collect the Khadyta River chronology. Neither did Steve McIntyre. The problem with treating tree ring chronologies as nothing more than received time series downloaded from the internet to be manipulated in various ways is that the context of the original investigators can be lost. Moreover, there are several subdisciplines within dendrochronology that collect tree ring data for different reasons, and in their fieldwork emphasize different site or individual tree characteristics during sampling.
 In any case, the concept of what comprises a single ‘site’ is ill defined (Wikipedia covers this in the context of archaeology). There is no hard-and-fast rule for how to geographically delimit which group of individual trees belong to a given chronology site.
Let’s return to Yamal and Khadyta. Remember that adding the Khadyta raw data to that from Yamal gives us the following situation (zooming in on the last century and a half now):
What becomes clear immediately is that the two chronologies diverge in approximately the early 1970s. Why does this occur?
We can look at the raw data from each site separately. First, I’ll graph the standardized raw data (to account for differences in tree size and mean growth rate) and their mean from Khadyta:
Again, what jumps out here is that the trees at this size show a decrease in growth in the late 1960s and early 1970s. The step change that the mean takes jumps out at me immediately.
Lets look at the same type of plot for the Yamal data:
Here, the persistent low growth years seen at Khadyta after 1970 are not present in the majority of trees (although, interestingly, perhaps in a few of them). As a consequence, the mean raw value continues to climb through the 1980s and 1990s.
We can now look at the two mean raw value time series together, as well as compare them to the gridded temperature data from the region. For this purpose, I’ve extracted summer (JJA) temperature for the grid box [65-70N, 65-70E] from CRUTEM3— a bit crude, but for the comparison to the raw mean data, it will serve our purpose:
Considered individually, and remember here with no detrending, both mean time series track each other — and the gridded summer temperature — well at both low and high frequencies. The two mean tree ring series diverge starting in the 1960s. Both track years with lower temperatures in the late 1950s and 1960s, but by the 1970s, Khadyta fails to mirror regional warming in the gridded summer temperature. Yamal continues to track increasing temperatures from the 1970s through the 1990s.
This quick-and-dirty analysis emphasizes a few points. Khadyta River does display the divergence problem, in that it ceases to track temperatures as it did from 1883 to the 1960s. At first glance, in this case, the divergence doesn’t seem to be related to detrending [PDF], and really does seem to reflect a decline in growth in the most recent decades. Second, adding Khadyta River data to Yamal is unlikely to reflect temperatures in the region more accurately.
I want to emphasize that neither Yamal nor Khadyta River are ‘the problem’ — divergence is the problem, and for this reason is a major [PDF] area of focus in dendrochronology. Relatively little more can be said about the specific case of Yamal at this point — I’ll leave that to the scientists working in this part of the world — but even my quick review of these data here shows that including Khadyta River raw data in the Yamal chronology does not result in a more accurate nor precise understanding of past temperatures in the region. This isn’t to say that some time in the past that Yamal didn’t experience divergence (this after all is a large part of the concern about divergence), but we can clearly see that Khadyta River does exhibit modern divergence.
A very interesting paper with relevance to this issue appears to be in press in Global Change Biology:
Esper, J., Frank, D., Buntgen, U., Verstege, A., Hantemirov, R., and Kirdyanov, A. (2009). Trends and uncertainties in Siberian indicators of 20th century warming. Global Change Biology, in press [subscription wall]
Check it out if you can.
UPDATE: Not following? Try Yamal III, a summary and update.
Steve McIntyre has once again stirred the hornet’s nest of online climate change denial with a hasty modification of the Yamal tree ring data published by Keith Briffa and colleagues in 2008 as part of a paper in Philosphical Transactions of the Royal Society (Phil. Trans. R. Soc. B (2008) 363, 2271–2284). Normally, I ignore McIntyre’s blog because of the juvenile name calling, repetitive nonsense, and the general misunderstanding of huge swaths of proxy paleoclimatology. However, I knew when Roger Pielke Jr. jumped in with support for his collaborator, it merited some attention [insert smiley face emoticon here].
Here, I’m actually interested in the data and the science. The first thing was to emulate the steps that McIntyre had performed (an audit, if you will), leaving aside for the moment whether they are even proper steps from a data point-of-view. McIntyre has rolled his own Regional Curve Standardization code in R, strangely eschewing the freely available software used by dendrochronologists, so I wanted first to ensure there was no significant error in his approach.
I downloaded the original Yamal data from here, and the Khadyta from the ITRDB here. I used ARSTAN to first emulate the original chronology used in Briffa et al. 2008. My regional curve standardized chronology differed slightly from the published version available here, probably because Briffa et al. 2008 used a time-varying spline for the regional curve, but the essential features, including the increasing values in the 20th century, are essentially the same. All these data and programs are publicly available. You can check these results for yourself.
I then added the Khadyta River raw data (which shows evidence of the ‘divergence problem‘) to the set of raw Yamal data, and recalculated the master chronology using regional curve standardization (because I am positive that McIntyre would insist on using all the data). Again, I am not yet addressing here whether it is  appropriate to add these data or  appropriate to not also add other or different data. Here is a comparison of the two versions:
Devastating, I know.
The real differences of course arise at the end, where the modern, relatively short series from Khadyta influence the final chronology.
Adding the Khadyta River series reduces the the level of the chronology though the 1970s and 1980s and into the early 1990s, when those data end. But if one includes both data sets, the series terminates similarly to the original Yamal chronology, of course (because the last few years are only present in the modern trees from Yamal). These changes are potentially important, and the actual scientific questions are interesting (as opposed to the political expedience of selecting certain findings to attack one’s political enemies). But the actual impact on the chronology is still far less than being implied by non-scientist partisans on one side. Why is that?
Part of the difference appears to be McIntyre’s use of a 21 year Gaussian low pass filter. The issues of how to smooth data series to avoid misleading end effects is not a trivial one. I can replicate the strong upturn in the modern era in McIntyre’s graph by using reflected end points. This creates the illusion of a massively unprecedented rise in ring width:
But as the close up view shows, one influence of the filter is such that it helps create the appearance of a massive rise, when annual values in mid-century are actually similar to those in the late 20th century.
There are actually interesting scientific questions (as opposed to the utterly uninteresting partisan griping) at play here that deal with the ‘divergence problem‘. I’ll address those in the next post.
UPDATE [10/06/09]: I should emphasize that this isn’t a comparison of standard RCS software vs. McIntyre’s home grown code. I might fire up R and do that comparison at some point, but I expect any differences to be minor.
UPDATE [10/09/10]: In case it still isn’t clear, my point about smoothing is not that there is anything wrong per se with a 21 point Gaussian filter using reflected endpoints. Rather, I’m pointing out one of the reasons that the initial graphs, posted at Climate Audit but that I first saw at Deltoid, convey such a dramatic rise in the last several years compared to mid-century is the behavior of this particular method.