« Famous Economists Quiz | Main | Dolphin markets in everything, Gresham's Law edition »
Geoengineering with Iron Fertilization
As even their critics admit, Levitt and Dubner have performed a useful service in drawing greater popular attention to geoengineering. Garden hoses to the sky,however, are not the only approach. Iron fertilization is simpler, cheaper and much more easily testable.
Most people are aware that CO2 and temperature are positively correlated in the long historical record but fewer people know that iron dust correlates negatively on the same scale - that is, temperature and CO2 levels are low when iron-dust is high. The graph illustrates.
The basic mechanism that appears to drive the association between low temperature, low CO2 and high iron-dust levels is that iron-rich dust sometimes sweeps off the continents into the oceans where it creates a plankton bloom. Phytoplankton take up CO2 in order to grow and as they die and produce fecal matter (I kid you not) carbon sinks to the lower depths or bottom of the ocean where it may remain for 100 to a 1000 or to even to millions of years (in the latter case eventually becoming oil).
A big advantage of iron fertilization as a way of reducing CO2 is that this process occurs naturally all the time and thus may be studied. It is also possible to run experiments. Indeed a dozen small-scale experiments over the past decade have already been run with all showing that iron fertilization does create phytoplankton blooms and some showing carbon sequestration. Interestingly, private firms looking for future carbon offset sources are driving much of the research into iron fertilization.
Of course, all the usual caveats about uncertainty and unintended consequences apply. Oceanus, the magazine of the Woods Hole Oceanographic Institution has an excellent issue on this topic.
Posted by Alex Tabarrok on November 4, 2009 at 07:40 AM | Permalink
Comments
Careful! You know what happens to heretics.........
Posted by: Jean at Nov 4, 2009 8:12:51 AM
Geoengineering is the left's Star Wars Missile Defense.
Posted by: Ryan at Nov 4, 2009 8:21:31 AM
get with the program Alex, it's all about cheap desalination, drip irrigation and biochar.
oops
Posted by: Luis Enrique at Nov 4, 2009 8:48:56 AM
It's the main advantage of garden hoses to the sky is that we already have reliable large-scale data from past volcanic activity? With other methods we only have small scale data and we need to rely on extrapolation - it would be really bad if we tried large scale iron fertilization and it turned out not to work as well as small scale experiment indicated. With garden hoses it's not that much of a problem (and they seem ridiculously cheap compared to carbon limits anyway).
Posted by: Tomasz Wegrzanowski at Nov 4, 2009 8:56:56 AM
See this highly critical article from a recent issue of Nature.
Opinion
Nature 461, 347-348 (17 September 2009) | doi:10.1038/461347a;
Published online 16 September 2009
Ocean fertilization: time to move on
Aaron Strong1, Sallie Chisholm2, Charles Miller3 & John Cullen4
Top of page
Abstract
Adding iron to the ocean is not an effective way to fight climate change, and we don't need further research to establish that, say Aaron Strong, Sallie Chisholm, Charles Miller and John Cullen.
(gated)
http://www.nature.com/nature/journal/v461/n7262/full/461347a.html
Posted by: jdm at Nov 4, 2009 8:59:54 AM
I don't think Krugman said what you think he said, Alex.
The conversation Krugman thought was good was, essentially, that the alternatives to cap and trade look good only because they haven't been thought through. And L&D's treatment of geoengineering is listed as an example of a suggestion that looks good because you didn't actually think about the consequences. That's a far cry from L&D did a useful public service by bringing public attention to geoengineering.
Posted by: Joe at Nov 4, 2009 9:20:48 AM
"Of course, all the usual caveats about uncertainty and unintended consequences apply."
Somehow these words ring a bit hollow. I sense enthusiasm for a techno-gimmick.
Posted by: City Comforts at Nov 4, 2009 9:47:08 AM
Just eyeballing it, the third chart looks much less correlated than the first two. For example the spike around 50k years has no real temperature equivalent, in spite of being the second-highest spike. Just not drawing a dashed line through it doesn't make it irrelevant.
Also it sure looks like iron dust levels lag temperature changes, rather than causing them.
I am not sure the data says what you think it says....
I'm willing to accept less reliable theories when they're highly valid, but phrases like "appears to drive" is not the level of confidence you want to see when considering dumping a whole bunch of crap in the ocean without really knowing what the unintended consequences might be.
Posted by: Andrew Edwards at Nov 4, 2009 10:12:00 AM
Carbon ppm tends to lag temperature changes by about 800 years. This is a point of contention in the climate science community: since the historic record suggests that temperature increases before carbon content in the atmosphere increases, and temperature dips before carbon content dips, it would appear that temperature drives carbon content in the atmosphere, and what's left to debate is
A)if they are directly related, how much of a positive feedback does the carbon content introduce for temperature.
B)if they are not directly related, what is the common forcing mechanism that drives them both.
Either way, the fact that carbon lags temperature is why Gore was criticized for presenting the two graphs similarly in An Inconvenient Truth, without the disclaimer of the 800 year lag. It made it seem to the audience that carbon increased in tandem with temperature and therefore directly drove it.
Posted by: Jim O'Toole at Nov 4, 2009 10:37:19 AM
The whole Nature article is worth reading but the mains points are a)
model based analysis indicate that iron fertilization would not alter the course of climate change, even if done on a massive scale; b) but it would totally change the ecology of the oceans; and c) despite a) and b), interest in this approach is being fueled by private for profit companies that hope to profit from selling carbon credits:
...
"Given that the efficacy and indirect effects of ocean fertilization for geoengineering cannot be tested directly without altering the ocean on unprecedented scales, we must resort to using global-ecosystem models to predict its promise and pitfalls. Many modelling analyses have shown that iron fertilization cannot reduce atmospheric CO2 enough to significantly alter the course of climate change. A model published in 2008 (K. Zahariev et al. Prog. Oceanogr. 77, 56–82; 2008), which is as convincing as any available, found that even if the entire Southern Ocean were fertilized forever with iron sufficient to eliminate its limitation of phytoplankton production, less than 1 gigatonne of carbon a year of CO2 of probable future emissions (currently about 8 gigatonnes a year) would be sequestered, and that amount for only a few years at best. This level of effort is simply not going to happen.
We think the idea of geoengineering by iron fertilization persists because of the blurry line between it and small-scale ocean fertilization experiments that address specific hypotheses.
The original goal of iron fertilization experiments was to answer fundamental questions about how iron and carbon are used and cycled in marine ecosystems. They began with a project called IronEx I in 1993. Since then, 11 experiments have been conducted in regions of the equatorial Pacific, subarctic Pacific, and Southern Ocean that have high levels of most nutrients but a relatively low phytoplankton biomass. The experiments confirmed the testable hypothesis that iron is the crucial limiting nutrient in these regions, and they provided insights into plankton community dynamics. They were conducted on a relatively small scale (less than 300 square kilometres) and over relatively short periods of time (less than 40 days), and had ephemeral effects on surface ocean ecosystems. These experiments fuelled the idea that iron fertilization, scaled up, might be used for geoengineering — even though they were not designed to investigate that possibility. This in turn prompted commercial interest in fertilizing the ocean to sell 'carbon credits'.
...
We already know enough about how ocean systems function to say that iron fertilization on large scales will be disruptive to ocean ecosystems and is unlikely to be effective for climate mitigation. Continuing to justify small-scale iron fertilization experiments in the context of global-scale geoengineering distorts the focus of oceanographic science, and encourages for-profit companies to continue pursuit of this strategy. It is time to move on."
Posted by: jdm at Nov 4, 2009 10:39:06 AM
see also the results of this iron fertilization experiment from this year. basically, the phytoplankton blooms that were supposed to suck up CO2 were promptly consumed by zooplankton, so very little carbon was actually sequestered.
http://www.awi.de/en/news/press_releases/detail/item/lohafex_provides_new_insights_on_plankton_ecology_only_small_amounts_of_atmospheric_carbon_dioxide/
net result: "Based on current knowledge, the Alfred Wegener Institute opposes large-scale iron fertilisation with the aim to reduce CO2 to regulate the climate."
Posted by: curious at Nov 4, 2009 10:58:19 AM
ooh, profit. Must be bad.
Posted by: anon at Nov 4, 2009 11:05:09 AM
Have to put in my obligatory "wtf does that axes label mean?" and remind people that the purpose of graphs is to illustrate.
Posted by: Andrew at Nov 4, 2009 11:11:58 AM
> ooh, profit. Must be bad.
I don't think the authors are addressing the question of whether profits are good or bad. They are addressing the question of whether this approach is likely to work (no), what its side effects are likely to be (bad), and what the motivations are for people who are promoting this approach (profit). Presumably these are all relevant issues for the public and policy makers to think about.
Posted by: jdm at Nov 4, 2009 11:18:58 AM
Well, before Republicans support testing this, the will need a definitive statement from the EPA that this is cost effective as well as a complete analysis by the EPA of the economic impact of the cost of mitigating climate change by the massively expensive spreading of iron dust in the oceans, the massive job losses that will result, and the much higher cost of steel as the iron ore is diverted from steel making.
Look, let's be clear, spending money to pay people to build sustainable energy capital or paying people to spread iron ore to mitigate unsustainable energy policies just results in job losses because higher taxes always kills jobs, because workers paid by taxes are not real jobs.
Look, the record of growth from tax cuts is clear - the economy today is far superior today now that the IRS takes only 14.9% of GDP than in 1999 when it took over 20% of GDP. Increasing taxes like was done in 1982, 83, 84, 85, 86, or in 1990, 91, 93 just kill economic and jobs growth as the depressions of the 80s and 90s demonstrates.
Look, unless the iron dust is a waste hat is more cheaply dumped in the oceans, this is no more acceptable to those opposing energy sustainability, and is merely just a way of kicking the problem down the road, a conservative tactic that has resulted in the world economy today.
Proposing the dumping of garbage and toxic waste in the rivers as a means of mitigating burning fossil fuels will get instant conservative support, but you won't get any conservative support for taxes to pay for dumping iron dust, nor even support for government funding for research.
Posted by: mulp at Nov 4, 2009 11:20:54 AM
There is also biochar, which may have longer legs than iron fertilization.
Posted by: Barkley Rosser at Nov 4, 2009 11:30:13 AM
Why not view global warming as we view many other complex systems here are the options:
1) Global Warming Not That Bad--We do nothing--Not a bad outcome
2) Global Warming Not That Bad--We do something--Not a bad outcome (could argue cleaner air/water/better outcome)
3) Global Warming Bad---We do nothing---Oh No...
4) Global warming Bad---We do something----Yeah!!!
The only disaster is doing nothing, so best approach is to do something...
Posted by: Joe at Nov 4, 2009 11:32:19 AM
On the topic of geoengineering - isn't it in some ways like a child safety cap on drugs? Rather than keeping drugs in safely out of reach, parents didn't worry so much about them and more kids died. If we know geoengineering is our "ace-in-the-hole," won't we lose most of our incentives to maintain a naturally well-functioning climate? At least one person thinks so.
All the usual caveats about uncertainty and unintended consequences seem to loom at lot larger once we've done permanent damage to the planet. Just saying...
Posted by: Pete at Nov 4, 2009 11:37:30 AM
Joe: "we do something" is too vague. Your calculation seem to implicitly assume that the "we do something" option under consideration doesn't (a) make the problem *worse*, (b) cause new problems that are worse than the original problem, or (c) have no effect.
In truth we have many different "somethings" we might consider doing, most of which don't pass any reasonable cost/benefit analysis. In particular, de-carbonizing the world economy at a cost of reducing economic growth even by a very small amount is probably not a good tradeoff because economic growth saves lives and improves our ability to address *every* possible future threat, not just the one under consideration.
So if we have to do "something", some form of geoengineering might be a good "something" to do.
Posted by: Glen Raphael at Nov 4, 2009 12:30:07 PM
Unintended consequences, yes. Our research group read a paper about modelling iron fertilization a year and a half ago. While it might reduce the amount of CO2 in the atmosphere, iron fertilization may have some unintended consequences. If I remember the paper correctly, when the algae decompose and sink to the bottom of the ocean the bacteria that decompose them emit N2O, which is a much worse greenhouse gas than CO2. Unfortunately, I can't find the reference, so I'll try to find it again today when I have time. Of course, this doesn't mean that we shouldn't explore geoengineering options as a supplement to reduced CO2 emissions, but this one might be a dead end.
Posted by: Aaron at Nov 4, 2009 12:58:49 PM
Glen,
I think geoengineering is a cool idea, and like anything else it will have its benefits and costs (mot of teh costs of this approach are hard to dollarize). Right now the costs of doing this seem small, but what are the long term consequences of pumping our air full of sulfur or iron dust.
We do know that cutting our CO2 emmisions would be a plus, but it to has a cost that we say we are not willing to pay. Tough Choices Abound. I think we are too scared as a planet to do anything so we are actually screwed. I hope I am wrong.
Posted by: Joe at Nov 4, 2009 2:59:47 PM
It seems to me that Biochar would eb much better.
Posted by: Floccina at Nov 4, 2009 3:33:28 PM
It seems to me that biochar would be better.
Posted by: Floccina at Nov 4, 2009 3:34:48 PM
Pete,
If geo-engineering actually works, I am all for "lulling" us into relatively more carbon emissions. Recall that carbon mitigation is hugely expensive in terms of lost GDP for amount of temperature reduction. (Note that the second-order increase in poisonings from easier access to meds by children was smaller than the first-order decrease from keeping caps on.)
Posted by: MW at Nov 4, 2009 3:38:43 PM
Most commentators underestimate how bad ineffective enviro policy can be. Hence the soundness of trying out different techniques before focusing on large taxes and cap and trade which destroy gnp growth.
A good example of how European green policy can go wrong is their encouragement of palm oil. By all accounts it was a disaster with up to 1400m tons of CO2 per annum being spewed into the atmosphere. This is comparable to some of the worst national car pollution and is the byproduct of a single green policy that the EU now admits is a disaster.
http://www.guardian.co.uk/environment/2007/apr/04/energy.indonesia
Are we so sure cap and trade won't lead to other such problems?
Posted by: chlorine gas at Nov 4, 2009 3:56:58 PM
