NYT: Geoengineering: Testing the Waters- Naomi Klein

111 views
Skip to first unread message

Andrew Lockley

unread,
Oct 27, 2012, 6:04:50 PM10/27/12
to geoengineering

http://mobile.nytimes.com/2012/10/28/opinion/sunday/geoengineering-testing-the-waters.xml

Geoengineering: Testing the Waters

By NAOMI KLEIN

Published: October 28, 2012

FOR almost 20 years, I've been spending time on a craggy stretch of British Columbia's shoreline called the Sunshine Coast. This summer, I had an experience that reminded me why I love this place, and why I chose to have a child in this sparsely populated part of the world.It was 5 a.m. and my husband and I were up with our 3-week-old son. Looking out at the ocean, we spotted two towering, black dorsal fins: orcas, or killer whales. Then two more. We had never seen an orca on the coast, and never heard of their coming so close to shore. In our sleep-deprived state, it felt like a miracle, as if the baby had wakened us to make sure we didn't miss this rare visit.The possibility that the sighting may have resulted from something less serendipitous did not occur to me until two weeks ago, when I read reports of a bizarre ocean experiment off the islands of Haida Gwaii, several hundred miles from where we spotted the orcas swimming.There, an American entrepreneur named Russ George dumped 120 tons of iron dust off the hull of a rented fishing boat; the plan was to create an algae bloom that would sequester carbon and thereby combat climate change.Mr. George is one of a growing number of would-be geoengineers who advocate high-risk, large-scale technical interventions that would fundamentally change the oceans and skies in order to reduce the effects of global warming. In addition to Mr. George's scheme to fertilize the ocean with iron, other geoengineering strategies under consideration include pumping sulfate aerosols into the upper atmosphere to imitate the cooling effects of a major volcanic eruption and "brightening" clouds so they reflect more of the sun's rays back to space.The risks are huge. Ocean fertilization could trigger dead zones and toxic tides. And multiple simulations have predicted that mimicking the effects of a volcano would interfere with monsoons in Asia and Africa, potentially threatening water and food security for billions of people.So far, these proposals have mostly served as fodder for computer models and scientific papers. But with Mr. George's ocean adventure, geoengineering has decisively escaped the laboratory. If Mr. George's account of the mission is to be believed, his actions created an algae bloom in an area half of the size of Massachusetts that attracted a huge array of aquatic life, including whales that could be "counted by the score."When I read about the whales, I began to wonder: could it be that the orcas I saw were on their way to the all-you-can-eat seafood buffet that had descended on Mr. George's bloom? The possibility, unlikely though it is, provides a glimpse into one of the disturbing repercussions of geoengineering: once we start deliberately interfering with the earth's climate systems - whether by dimming the sun or fertilizing the seas - all natural events can begin to take on an unnatural tinge. An absence that might have seemed a cyclical change in migration patterns or a presence that felt like a miraculous gift suddenly feels sinister, as if all of nature were being manipulated behind the scenes.Most news reports characterize Mr. George as a "rogue" geoengineer. But what concerns me, after researching the subject for two years for a forthcoming book on climate change, is that far more serious scientists, backed by far deeper pockets, appear poised to actively tamper with the complex and unpredictable natural systems that sustain life on earth - with huge potential for unintended consequences.In 2010, the chairman of the House Committee on Science and Technology recommended more research into geoengineering; the British government has begun to spend public money in the field.Bill Gates has funneled millions of dollars into geoengineering research. And he has invested in a company, Intellectual Ventures, that is developing at least two geoengineering tools: the "StratoShield," a 19-mile-long hose suspended by helium balloons that would spew sun-blocking sulfur dioxide particles into the sky and a tool that can supposedly blunt the force of hurricanes.THE appeal is easy to understand. Geoengineering offers the tantalizing promise of a climate change fix that would allow us to continue our resource-exhausting way of life, indefinitely. And then there is the fear. Every week seems to bring more terrifying climate news, from reports of ice sheets melting ahead of schedule to oceans acidifying far faster than expected. At the same time, climate change has fallen so far off the political agenda that it wasn't mentioned once during any of the three debates between the presidential candidates. Is it any wonder that many are pinning their hopes on a break-the-glass-in-case-of-emergency option that scientists have been cooking up in their labs?But with rogue geoengineers on the loose, it is a good time to pause and ask, collectively, whether we want to go down the geoengineering road. Because the truth is that geoengineering is itself a rogue proposition. By definition, technologies that tamper with ocean and atmospheric chemistry affect everyone. Yet it is impossible to get anything like unanimous consent for these interventions. Nor could any such consent possibly be informed since we don't - and can't - know the full risks involved until these planet-altering technologies are actually deployed.While the United Nations' climate negotiations proceed from the premise that countries must agree to a joint response to an inherently communal problem, geoengineering raises a very different prospect. For well under a billion dollars, a "coalition of the willing," a single country or even a wealthy individual could decide to take the climate into its own hands. Jim Thomas of the ETC Group, an environmental watchdog group, puts the problem like this: "Geoengineering says, 'we'll just do it, and you'll live with the effects.' " The scariest thing about this proposition is that models suggest that many of the people who could well be most harmed by these technologies are already disproportionately vulnerable to the impacts of climate change. Imagine this: North America decides to send sulfur into the stratosphere to reduce the intensity of the sun, in the hopes of saving its corn crops - despite the real possibility of triggering droughts in Asia and Africa. In short, geoengineering would give us (or some of us) the power to exile huge swaths of humanity to sacrifice zones with a virtual flip of the switch.The geopolitical ramifications are chilling. Climate change is already making it hard to know whether events previously understood as "acts of God" (a freak heat wave in March or a Frankenstorm on Halloween) still belong in that category. But if we start tinkering with the earth's thermostat - deliberately turning our oceans murky green to soak up carbon and bleaching the skies hazy white to deflect the sun - we take our influence to a new level. A drought in India will come to be seen - accurately or not - as a result of a conscious decision by engineers on the other side of the planet. What was once bad luck could come to be seen as a malevolent plot or an imperialist attack.There will be other visceral, life-changing consequences. A study published this spring in Geophysical Research Letters found that if we inject sulfur aerosols into the stratosphere in order to dial down the sun, the sky would not only become whiter and significantly brighter, but we would also be treated to more intense, "volcanic" sunsets. But what kind of relationships can we expect to have with those hyper-real skies? Would they fill us with awe - or with vague unease? Would we feel the same when beautiful wild creatures cross our paths unexpectedly, as happened to my family this summer? In a popular book on climate change, Bill McKibben warned that we face "The End of Nature." In the age of geoengineering, we might find ourselves confronting the end of miracles, too.Mr. George and his ocean-altering experiment provides an opportunity for public debate about an issue essentially absent during the election cycle: What are the real solutions to climate change? Wouldn't it be better to change our behavior - to reduce our use of fossil fuels - before we begin fiddling with the planet's basic life-support systems?Unless we change course, we can expect to hear many more reports about sun-shielders and ocean fiddlers like Mr. George, whose iron dumping exploit did more than test a thesis about ocean fertilization: it also tested the waters for future geoengineering experiments. And judging by the muted response so far, the results of Mr. George's test are clear: geoengineers proceed, caution be damned.

The author, most recently, of "The Shock Doctrine: The Rise of Disaster Capitalism."

Ken Caldeira

unread,
Oct 28, 2012, 8:37:22 PM10/28/12
to josh...@gmail.com, geoengi...@googlegroups.com
Of course, this statement of Naomi Klein's is false (unless you are willing to stretch the meaning of the word 'could' to encompass everything that is not a logical impossibility):

The scariest thing about this proposition is that models suggest that many of the people who could well be most harmed by these technologies are already disproportionately vulnerable to the impacts of climate change. 

Surely, models suggest the contrary, that solar goengineering may allow those who are disproportionately vulnerable to the impacts of climate change to avoid some of the harm.

A robust result of solar geoengineering simulations is that these methods, at least in the models, reduce the amount of climate change for most people in most places most of the time.  Although there is always a chance that someplace might be negatively impacted, the robust results are that solar geoengineering tends to increase food production by diminishing heat stress (see attachment).
 
By working to remove an option that vulnerable communities might use to reduce harm caused primarily by CO2 emissions from developed countries, Naomi Klein, ETC, etc are increasing the potential for damage to "the disproportionately vulnerable".  In their effort to be politically correct, they are exposing to increased risk the very communities they paternally (maternally?) claim to be protecting.



On Sat, Oct 27, 2012 at 10:54 PM, Joshua Jacobs <josh...@gmail.com> wrote:
Respect, Naomi Klein as I may, I am befuddled by the spin that seems to have been swallowed by her, multiple media outlets, and researchers alike.   How do the HSRC activities to restore a marine ecosystem constitute an act of geoengineering (or eco-terrorism to some) any more or less than "native" ecosystem restoration or conservation projects around the world?   Furthermore,  why is "geoengineering" such a reviled word when used in reference to these projects while "conservation" and "restoration" are revered...even when they fundamentally apply to the same process?  That is, imposing our imperfect idea of what Nature would do without us.  In addition, to what prehistoric ideal state can we possibly "restore" a constantly evolving ecosystem to in lieu of a changing climate (now and millennia in the past)?  

Despite my bewilderment in the overuse of an Appeal to Nature Argument in Naomi's article, I see great value in supporting the rich biodiversity of both native and novel ecosystems (see Emma Marris' "The Rambunctious Garden").  With the enormous carbon exchange that goes on between global ecosystem and atmosphere each year(~210 Gt taken in by photosynthesis, ~210 respirated/decomposed back, plus ~9 Gt anthropogenic), it seems foolish not to utilize the capacity of ecosystems to store atmospheric carbon in organic, mineralized, or re-fossilized forms.  Furthermore, it is necessary to have ecosystem management (of any scale) be financially and politically, as most certainly ecologically, viable.  This is what I believe that Russ George has been, albeit clumsily, aiming for.  We would do well to improve on his model.

Thoughts on this?  

--
You received this message because you are subscribed to the Google Groups "geoengineering" group.
To view this discussion on the web visit https://groups.google.com/d/msg/geoengineering/-/5izNoMepvS8J.

To post to this group, send email to geoengi...@googlegroups.com.
To unsubscribe from this group, send email to geoengineerin...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/geoengineering?hl=en.

Pongratz_et_al_NatureCC2012.pdf

Andrew Lockley

unread,
Oct 28, 2012, 9:09:21 PM10/28/12
to Ken Caldeira, geoengineering

Ken

I accept partly the logic of your argument, but not its conclusions.

You may sensibly argue that the average impact is reduced, or that global food production is not unfavorably impacted.

However, that asssumes efficient allocation of resources. As there's already evidence that endemic corruption in India is causing malnutrition on a grand scale, is it reasonable to assume any compensatory allocation of resources to address monsoon failure will a) actually happen and b) reach the people who need it?

I'm far more worried about a small number of People starving a lot than a large number of people starving a little - especially when the small group has nuclear weapons.

We need to look at the realpolitik, not just the models, when addressing risks.

A

Doug MacMartin

unread,
Oct 28, 2012, 11:03:56 PM10/28/12
to andrew....@gmail.com, Ken Caldeira, geoengineering

Andrew, others,

 

Are there any modeling results that support the hypothesis that there exists some region on the planet for whom the slightest amount of solar geoengineering will shift their climate even further away from whatever baseline you pick (current or pre-industrial) than it will be under greenhouse gases? 

 

It is unequivocally true that one could choose to introduce a large enough reduction in insolation so that some regions would indeed be worse off, and there’s no question there’s a concern here about who gets to decide (that was one motivation for our latest paper on optimizing the distribution of radiative forcing, posted last week).  But in the models I’ve seen, some amount of geoengineering appears to improve the climate everywhere.  (Yes, solar geoengineering is likely to reduce rainfall in India, but part of that is just offsetting the increased rainfall due to greenhouse gases, right?)

 

See, for example, paper last year by Juan Moreno-Cruz, Kate Ricke, and David Keith, in Climatic Change

A simple model to account for regional inequalities in the effectiveness of solar radiation management

They looked at the Pareto-optimal strategies and found that the region that “wanted” the least solar geoengineering still wanted 78% as much as what the global optimum would say.  Of course, that depends on what damage function you define, it depends on how much spatial averaging you do (they looked at Giorgi regions), and it depends on the model you use.  So I certainly agree that this is an issue that we need to be on the lookout for, I just haven’t seen any evidence implying that there is any region where things actually get worse.

 

(If there is, I’m assuming that someone will correct me and point to the paper that shows this.)

 

doug

David Lewis

unread,
Oct 29, 2012, 8:18:31 AM10/29/12
to geoengi...@googlegroups.com, kcal...@carnegiescience.edu
Naomi has invented her own distorted version of an idea Bill McKibben first advanced in the late 1980s in his book "The End of Nature".  McKibben wrote at that time that he felt differently about being in what he formerly regarded as the pristine wilderness now that he realized that human activity had changed the composition of the atmosphere which had changed global climate which must have changed every ecosystem on the planet.  

Naomi's use of this McKibben idea requires her to define everything as fine until she heard all the fuss about a geoengineering experiment out in the Pacific.    

Now she can't look at an orca swimming in the Gulf of Georgia in front of her home without worrying that it wouldn't be swimming there unless that 120 tonnes of fertilizer had been dumped in the Pacific hundreds of miles away.  She feels strange.   She writes: "once we start deliberately interfering with earth's climate systems - whether by dimming the sun or fertilizing the seas - all natural events can begin to take on a sinister tinge.  ....as if all of nature were being manipulated behind the scenes".  

1,000,000 tonnes per hour of the CO2 that is emitted into the atmosphere by civilization is absorbed by the ocean every hour, but a one time application of 120 tonnes of fertilizer, because it is "deliberate" in a way that the CO2 Naomi emits while flying around the world on her speaking tours isn't, bothers her.

Jane Long

unread,
Oct 29, 2012, 8:28:41 AM10/29/12
to jrando...@gmail.com, geoengi...@googlegroups.com, kcal...@carnegiescience.edu
Letters to the editor?
To view this discussion on the web visit https://groups.google.com/d/msg/geoengineering/-/VmGjA_AZGtwJ.

To post to this group, send email to geoengi...@googlegroups.com.
To unsubscribe from this group, send email to geoengineerin...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/geoengineering?hl=en.


--
Jane C. S. Long

janec...@gmail.com
jcslong Skype



Doug MacMartin

unread,
Oct 31, 2012, 11:48:41 AM10/31/12
to p.j.i...@gmail.com, geoengi...@googlegroups.com, andrew....@gmail.com, Ken Caldeira

Hi Pete – that doesn’t actually contradict what I was suggesting.  My point, as Tom also suggested, is that SRM doesn’t need to be a binary choice of either we try to restore global mean temperature back to pre-industrial, or we don’t do any SRM.  The former will certainly cause some regions to be too dry.  But halfway between these might be “better” everywhere.

 

 

(To be concrete, suppose you look at a 2xCO2 scenario, and decrease solar radiation by 1%, enough to offset roughly half of the warming.  Is there anywhere “harmed” by this, in terms of their temperature and precipitation being even worse than it would be with no SRM?  Of course, the answer depends on how you measure climate damage, which is a big, and unanswered question.  But the answer is certainly not obvious.  Kate pointed out that there are regions that get drier with CO2, and drier still with SRM, but the precipitation changes are relatively small compared to temperature changes, so that in their damage function, everywhere would still benefit from this half-way scenario.  Really, to answer this need to be much more careful with defining benefit and harm, I’m being loose with just associating it with temperature and precipitation changes, and even there need to ask how to normalize relative changes.)

 

And of course, whether people judge that they have been harmed by geoengineering is different from whether they have been; it is certain that many people will feel that they have been.

 

And thanks! 

 

doug

 

From: geoengi...@googlegroups.com [mailto:geoengi...@googlegroups.com] On Behalf Of p.j.irvine
Sent: Wednesday, October 31, 2012 12:38 AM
To: geoengi...@googlegroups.com
Cc: andrew....@gmail.com; 'Ken Caldeira'; macm...@cds.caltech.edu
Subject: Re: [geo] Re: NYT: Geoengineering: Testing the Waters- Naomi Klein

 

doug,

 

I think you might be a little hasty in arguing that in model simulations no region has shown any signs of that there will be a shift of their climate further from the baseline. Your paper and the Moreno-cruz paper both use a similar approach that may miss out on some perfectly anticipatable objections to the climate shifts that you simulate. I believe that a robust signal across the geoMIP ensemble and other model simulations is a reduction of precipitation across North America and Eurasia to a level below the pre-industrial, reversing, and then some, the anticipated global warming shift to greater precipitation. A reduction in precipitation is not equal to an increase in precipitation, and in the extreme case, a doubling of precip is very different from precip dropping to zero (not that this has been seen). There are going to be regions where people judge that they have been harmed by SRM geoengineering, Although whether Monsoon affected India and Africa are in that class is uncertain.

 

nice paper by the way,

 

Pete

To view this discussion on the web visit https://groups.google.com/d/msg/geoengineering/-/FZp0y3-1Qu4J.

Ken Caldeira

unread,
Oct 31, 2012, 12:09:14 PM10/31/12
to Doug MacMynowski, Peter Irvine, geoengineering, Andrew Lockley
Two points:

1. Plants do not care much about precipitation but are sensitive to soil moisture. The focus on precipitation may be misplaced. On a global mean basis evaporation = precipitation so decreased precipitation means decreased evaporation. Less precipitation does not necessarily mean drying. Talking about precipitation changes without talking about evaporation changes is like talking about assets without talking about debts.

2. The issue is that solar intensity causes higher ratio of precip-to-temperature changes than do changes in CO2. If the concern is not to have less precip than in pre-industrial days, we could be talking about solar geoengineering not being able to compensate for all of the warming, rather than solar geoengineering causing a decrease in precipitation. It is all scenario dependent.

_______________
Ken Caldeira

Carnegie Institution for Science 

Stephen Salter

unread,
Oct 31, 2012, 1:06:53 PM10/31/12
to geoengi...@googlegroups.com
Doug

What you say may depend on what kind of SRM you do.  Tropospheric  seeding can affect precipitation in either direction a long way from the spray source depending on where, and perhaps when relative to the monsoon you do it.  We hope to be able to make dry places a bit wetter and wet places a bit drier.

Stephen
--
Emeritus Professor of Engineering Design School of Engineering University of Edinburgh Mayfield Road Edinburgh EH9 3JL Scotland S.Sa...@ed.ac.uk Tel +44 (0)131 650 5704 Cell 07795 203 195 WWW.see.ed.ac.uk/~shs
Reply all
Reply to author
Forward
0 new messages