21 April 2010

A Few Questions from Jonathan Gilligan

Jonathan Gilligan, Associate Director for Research, Vanderbilt Climate Change Research Network and Associate Professor, Department of Earth and Environmental Sciences, at Vanderbilt University, has written in with some interesting questions on the thought experiment I posted earlier today. With his permission I reproduce his email below, along with my replies.
Dear Roger,

A couple of years ago, you wrote a very nice paper reviewing the prospective cost of air capture technology. You post today on your blog arguing that decisions whether to accelerate decarbonization of the energy supply don't depend in any practical way on climate science.

If you have time to think about them and reply, I would like to ask three related questions:

(1) Do you similarly think that decisions whether to spend lots of money researching and developing air-capture technology don't depend on climate science?

REPLY: The decision to deploy air capture technology certainly depends upon a judgment that we should "tune" atmospheric carbon dioxide levels to some precise value. My view is that air capture is not a technology that would be deployed any time soon, but we may wish to develop it to keep our options open decades hence. This judgment is based on the possibility that we may want to have a back-up plan in canse conventional mitigation does not work out and we wish to adopt a "brute force" approach.

So, any decision to deploy air capture will certainly require a judgment that doing so is beneficial, and this would come from information produced by climate science. Just as we will want to invest in air capture research in coming decades, so too will we want to invest in continued climate science research. The state of climate science today is already sufficient to justify investing in air capture research.

My post today was about accelerating decarbonization, and to start on that path, we need not depend upon climate science, and those steps make sense for reasons well beyond science. Just because we can justify accelerated decarbonization without depending on climate science does not mean that we may not wish to rely on climate science for future decisions.
(2) Do you think decisions how much foreign aid to give LDCs to help them adapt to climate change depend on climate science? I can see arguments either way: much of the adaptation aid would help reduce vulnerabilities to current natural hazards, as you have pointed out many times; but on the other hand, the magnitude and timing of the aid could be very different if we were looking at sea-level rise displacing 18 million Bangladeshis in the next 50 years vs. no real change from current circumstances in the coming century.

Just so I'm clear: when I talk of the relevance of climate science, I'm talking mostly about what we know today---that AGW is real and that within the envelope of possible/plausible futures, there are truly catastrophic outcomes, although the science can't assign meaningful probabilities to those outcomes and (as you and Sarewitz have argued) is unlikely over the next decade to improve its predictive power significantly. Despite what you write, I can't help concluding that this limited scientific understanding is indeed very relevant to energy and adaptation policy because the quasi-irreversible nature of climate change (unless we quickly perfect air capture) would strongly affect the speed with which we choose to act.
REPLY: No. I do not think that adaptation funding should in any way be tied to climate science. The reason for this is that is virtually every location that experiences climate impacts, climate (and projected future changes in climate) is far less important than the societal factors that drive losses. Simply focusing resources on the places that are most vulnerable to impacts will have large benefits irrespective of how the future evolves over the next half century or so.

Consider your Bangladesh example. Bangladesh is projected to gain land area by 2050. But whether it does or does not I think is less relevant than the certainty that Bangladesh is a poor country with considerable exposure to climate events. Developing adaptive capacity in Bangladesh makes sense on its own merits. As I have argued, with Dan Sarewitz and others, adaptation policies can stand on their own, and recently with Mike Hulme and Suraje Dessai, such polices do not depend upon accurate climate predictions.
(3) Could research on public opinion test your hypothesis and perhaps settle some disagreements surrounding it? It seems to me that many questions about or disagreements over your hypothesis on the relevance of climate science come down in large part to empirical questions of what matters to voters and decision-makers. Do you know whether anyone's doing good public opinion research to test whether voters' (or their elected representatives') support for decarbonizing energy (or the related hypotheses about support for air capture or adaptation) depend on questions pertaining to climate science? And would such research be relevant to the questions you raise?
REPLY: A good part of Chapter 2 of my new book makes the case that there is ample political will for action on climate change, and has been for more than a decade. Public support in the US has been remarkably consistent over that time, even with its periodic ups and downs. At the same time, the public has grown increasingly skeptical about the claims made about climate science, believing that it has been exaggerated by climate scientists. Yet, support for action remains strong. There is plenty of polling data to back up these claims, which I cite in the book. And there is also literature which suggest that appeals to alarm or fear just "won't do it."

There is also a much broader literature on the public understanding of science which argues that knowledge of particular scientific facts does not compel particular actions. See this review article, for instance. So I think that the work in this area provides a compelling body of theoretical and empirical reasons to suggest that support for climate policy depends far less on climate science than many in the debate actually might think.
Thanks for your time and attention.
Thank you as well!
Best regards,


  1. Ah the Bangladeshi argument - introduced to tell us all that he truly cares about the poor and vulnerable while the rest of us don't - yet totally ignorant of on-the-ground facts, like much of this debate, such as:
    1. As you say they have actually been gaining ground not losing it.
    2. The sheer numbers moving to the coast belies the idea they are worried about sea-level rise.
    3. Seasonal floods and droughts cause utterly massive and regular climate changes far more important than a few measly cm of sea level rise in 50 years. All river dwellers, especially delta dwellers know that flooding is a persistent danger. They normally build houses on stilts for that very reason.
    3. Plain old-fashioned erosion is the biggest problem, solvable by building sand-banks, not by the displacement of 18 million people. That's what needs done, not the fatalistic computer projections and impact reports. We know it and they know it but nobody ever provides the money for it.
    4. Using more fossil fuels would probably alleviate the need to cut down trees and hence prevent some of the devastating mudslides caused by local deforestation.

    Where do these guys get their ideas anyway? Whenever they consistently get the basics wrong, you can't really trust their judgement on anything more complex.

  2. "REPLY: A good part of Chapter 2 of my new book makes the case that there is ample political will for action on climate change..."

    Push polling the climate change issue reveals little about public support. Ranking priorities is far more revealing, and when climate change is included, it falls way down the list. A Pew poll last November had it at 20 in a list of 20 important issues measuring public opinion.

    Now, when scientist/policy advocates choose the lower priority over the more urgent issues, is it any wonder the distrust that develops when the convoluted misdirection is revealed?

  3. What's a picture of David Dimbleby (http://en.wikipedia.org/wiki/David_Dimbleby)doing adorning this article?

  4. Regarding my motives, jgdes (#1): mind-reading is an inexact art. This was a private letter to Roger, which he asked to post on the blog. It wasn't a public statement. I was not posturing.

    While I'm not an expert on Bangladesh, I have a personal interest in the country because I have friends and colleagues there; my question arises from that.

    I also brought in Bangladesh because Roger has written previously on adaptation and third world countries, so I guessed that he would be knowledgeable and thus could give me an informed and thoughtful answer, which he did.

    Regarding recent trends in deposition: First, I am well aware of these. Steve Goodbred, who has the office next to mine, has done a lot of work on sediment transport in the Ganges Brahmaputra system and the evolution of the delta, and is one of the scientists who first called attention to the fact that deposition could significantly offset the effect of sea-level rise.

    However, I would ask whether anyone here really thinks that extrapolating recent trends in sediment deposition is a reliable way to predict conditions 50 years from now. My take is that the interaction of sea-level, sediment transport, and subsidence is very complicated and that 50-year predictions are extremely uncertain. Roger understands this; hence his, "whether it does or not ... is less relevant..."

    Lots could happen to the low-lying regions apart from sea-level rise, but sea-level rise is an important factor and I think it should not be neglected.

    Sheer numbers moving to the coast? Are you familiar with Bangladesh? First, for the most part, people are not moving to the coastal villages; they're mostly moving to the cities, most of which are not on the coast.

    Second, the country is incredibly densely populated---half the population of the US in the area of Iowa. People even live on chars---sand bars in the middle of rivers---that flood every summer and eventually erode into oblivion. People live wherever they can find land, regardless whether it's safe or dangerous, permanent or transient.

    There has been some internal migration into marginally habitable coastal lands in the Sundarbans, but these lands are extremely vulnerable to erosion and to cyclones and people move there only because there's nowhere else to go.

    Erosion is not easily stopped by engineering. The multiple failures of the $70 million hard point at Sirajganj are a good example of this. Erosion in the delta would be even harder to control, given the dense network of tidal channels. It would be good to review the reasons the country rejected the Flood Action Plan in the early 1990s to understand the problems of hard engineering solutions.

    On the other hand, there have been very successful engineering projects, such as the Bongobandhu bridge and the cyclone shelters throughout the coastal region. Engineering can work, but it must be done intelligently, and with careful consideration of environmental stresses (which would include climate change).

    Deforestation in the Chittagong Hill Tracts is driven not primarily by demand for fuel but by demand for timber as a raw material and slash/burn agriculture. Apart from the hill tracts, the country is too flat for mudslides to be a big problem.

    It is discouraging that instead of pursuing civil discussion of these complex questions, some people would rather lash out with ad hominem attacks, assuming that anyone who disagrees with them must be both ignorant and malevolent.

  5. Roger: Thanks for taking the time to respond to my questions. I agree with much you say:

    We can set sensible policy on adaptation, decarbonization, and air capture R&D based on the current state of climate science and that detailed predictions of future climate change are not tremendously relevant to the policy process at this initial stage.

    In the future, policy decisions may indeed be strongly affected by future developments in our ability to predict climate but the first steps would basically be the same regardless of the exact climate future we face (within the envelope of reasonable possibilities).

    Where I think there's an interesting thread to pursue is in the role of climate science and adaptation policy: I agree that we shouldn't base adaptation policy on detailed predictions, but I think we should look hard at the range of possible climate futures. Suppose some adaptation measures are more vulnerable to certain possible future climates whereas other adaptation measures are more resilient in the face of a wide range of climate futures. It is only sensible to factor the vulnerability or robustness of different adaptations to the range of possible climate futures.

    So what I'm proposing is that the best use of climate science is to explore the full range of possible climatic conditions and to assist policymakers in choosing policies that have the best chance of producing a desired outcome over that range.

    To me, this could be a useful application of the "Honest Broker" role you describe: there's a rich literature on the problems of policymakers failing to understand the range of possible stresses their policies will face and the possible outcomes under those stresses (I particularly like Slovic, Kunreuther, and White's "Decision Processes, Rationality, and Adjustment to Natural Hazards"). An Honest Broker could provide clear guidance to make sure the decision-maker (voters, elected officials) understood the context and range of plausible outcomes of different choices.

    You recommend a similar approach in the paper with Hulme and Dessai that you cite above. So if I understand that argument correctly, it seems as though we both agree that there's an important role for climate science in setting policy to respond to climate change, but that this role is not to predict specific outcomes but to conduct vulnerability analysis by exploring the full range of possible outcomes.

  6. -5-Thanks Jonathan

    Climate science that attempted to map out the range of possible futures, rather than selected scenarios, projections or predictions, would be a very important role. We agree on this point 100%. Thanks much for the exchange;-)

  7. Jonathan
    My apologies for lumping you in with others who routinely bring up Bangladesh as a convenient totem and try to apply their simplistic model extrapolations without considering the real situation. I'll remain civil, though I detected no ad homs but I was certainly overly harsh. It's the frustration caused by over-simplifications: I'd hoped everyone knew enough not to bring up that false totem again. Though i'm a little perturbed you brush aside the killer mudslides as being of such minor importance. I'm am very much FOR alternative, green fuels but i cannot hide from the facts that fossil fuel use has had an overall effect over here of halting deforestation, by providing better fuel, housing materials and farming methods. One might even say coal use saved our forests.

    Surely though, if you reread your sentence you must realize it is overly simplistic - even in private correspondence. As you seem to know, their problems are here and now, not in 50 years, and not at all from a few piddling cm of potential sea level rise that is totally dwarved by seasonal surges and will likely be offset by sediment rise anyway. Yes simplistic linear extrapolation is indeed a dodgy practice, especially using computer models that everyone knows are utterly useless at local scales and especially so for Bangladesh. For these vulnerable areas we need to look at the actual long term records. Hence presumably you also know that there have been zero trends in monsoons, rainfall, droughts and even temperatures since records began there. Indeed there is no climate change there at all. That there are such great problems is because it is a vulnerable area to live in the first place and rank poverty makes it much worse. Happily, however their gdp is steadily increasing - one must presume thanks to technological progress and hence, in large part to fossil fuels.

    This is the crux of environmental concerns. Are we truly worried about the real problems and their roots or are we just trying to simplistically blame everything on a few tenths of a degree of temperature rise? Are we in danger of ignoring these current problems by highlighting shaky and irrelevant scenarios 50 years hence.

    As for sea level rise causing displacement: In fact while the sea level has been rising, the reclaiming of land by engineering has increased the amount of land available in New York, Boston, London, Paris, Rome, indeed anywhere you care to look. Ancient Rome is 20 feet under modern Rome; they coped! Gavin schmidt used to use a bad joke about Battery Park basement flats being soon under water, without noting they were built on reclaimed land in the first place. The Dutch manage to control erosion with engineering and so do the French. Sandbanks are not even hi-tech but they worked brilliantly in the Landes thanks to Napoleanic engineering. So France's largest forest is now sitting where only wind-blown, malarial marshes existed beforehand. There's the solution - absolutely nothing to do with making us all feel guilty about industrial progress. The real truth is the Bangladeshis simply don't yet have enough of what we take for granted.

  8. By the way, here is a more useful prediction model for Bangladesh:

    "Bangladesh’s coast is in constant danger of devastating floods. Tropical hurricanes in the Indian Ocean move towards the coast, pushing a wall of water ahead of them. This creates a great wave called a storm surge, which can reach 12 meters high. In 1970, a five-metre wave flooded a million acres of rice fields, killing at least 200,000 people. More than 100,000 further lives were lost this way in 1991.
    The country is densely populated and lacks the resources to erect defences or carry out evacuations. Also, the land is so low that even flood waves a couple of metres high can have terrible consequences."

    Now what climatic event does anyone suppose that Bangladeshis are realy most concerned about? And does any CO2 mitigation by anyone, anywhere help them one jot? And before anyone helpfully suggests that cyclones will get worse under global warming - well the science doesn't even mildly indicate that and the records negate it.

  9. Regarding the debate over Bangladesh there is this recent news: http://www.google.com/hostednews/afp/article/ALeqM5jxWAlO7hpr2AXkrZMWswKyK39gOA

    "But IPCC's prediction did not take into account the one billion tonnes of sediment carried by Himalayan rivers into Bangladesh every year, which are crucial in countering rises in sea levels, the study funded by the Asian Development Bank said.

    "Sediments have been shaping Bangladesh's coast for thousands of years," said Maminul Haque Sarker, director of the Dhaka-based Center for Environment and Geographic Information Services (CEGIS), who led research for the study.

    Previous "studies on the effects of climate change in Bangladesh, including those quoted by the IPCC, did not consider the role of sediment in the growth and adjustment process of the country?s coast and rivers to the sea level rise," he told AFP.

    Even if sea levels rise a maximum one metre in line with the IPCC's 2007 predictions, the new study indicates most of Bangladesh's coastline will remain intact, said Sarker.

    "Based on the findings of the study, it appears that most of Bangladesh?s coastline, notably the Meghna estuary, which is one of the largest in the world, would rise at the same pace as the sea level growth," he said."

    It's really a matter of ranking threat priorities and choosing the right mix of adaption and mitigation responses.