Over at The Volokh Conspiracy, Jonathan Adler notes that the NAS is starting to look into the possibility of geoengineering to roll back human changes in the climate. For those of you who haven't heard, geoengineering is the process of deliberately changing the climate to compensate for the effects of greenhouse gases.
There's no shortage of reasons it might not work. The most obvious is that climate is not fully understood as it is, and so massive alterations may have unintended consequences and make things worse. Especially if the methods of geoengineering result in permanent changes, this could be disastrous. It's also true that no global change will impact each climatic region equally and so any geoengineering scheme will likely have at least some countries objecting to its use even in the abstract.
Those are fairly compelling objections, though I should also mention another common objection that I don't find compelling at all: the moral hazard argument. If geoengineering works, that might destroy the incentive to reduce CO2 emissions. Well, yes, probably so. And who cares? The reason to reduce CO2 is that it has bad effects, not that they're little sin molecules that it's intrinsically immoral to release. A weaker version of this argument does have some merit; that is if geoengineering only works partially it may merely delay problems that could be actually prevented by emissions elimination. But that's a reason for simply being very careful to make sure it works completely or almost completely, not a reason for abandoning the entire concept without study.
The methods themselves sound something like science fiction, not least of the reasons why is that at the moment that's what they are. Among the methods proposed are orbital mirrors to redirect a tiny fraction of the sun's light away from the earth, reflective sulfur particles in the upper atmosphere doing much the same thing, using iron to grow large blooms of plankton as a carbon sink, and a number of other clever and slightly alarming ideas.
I have no idea if any of this will turn out to work even in theory, much less whether it will be practical at cost scales below tens of trillions of dollars. If it does, it may well not be practical from a legal standpoint in that the tangle of environmental treaty obligations may not ever allow such a project to get off the ground.
But there's no way to know without studying the problem in detail. I'm glad the NAS is at least looking into the problem. If it doesn't pan out, it doesn't pan out. But if it does, it might be very good news for this planet.
- Log in to post comments
Most of which fail to address ocean acidification (and in fact, the last will probably worsen it). That puts some sharp limitations on the situations in which they might be advantageous. Especially if people want to go on eating many kinds of seafood. Only schemes to bind carbon up in rock (enhanced silicate weathering) or in plants and soil (biochar, forest-planting etc) address ocean acidification - and presently those schemes appear inadequate to cope with current emission levels.
More importantly - none of these methods can be implemented now. CO2 emissions reductions can be implemented now. That's faster to deploy, that's off-the-shelf technology. These other geo-engineering schemes are not (though mesospheric SO2 would be if we weren't picky about results).
Finally - all so for suggested geo-engineering schemes will have fewer risks, be much cheaper, and more effective with lower CO2 emissions.
It's important that the NAS is investigating these methods - they may be needed. But none of them is likely to be a solution in its own right.
Trivial solution: Put the sulfur back in (don't remove it in the first place from) aviation kerosene. FOB (actually less than zero cost - no cost to remove it) SO2 injection into the stratosphere will erect a faint sulfuric acid hydrate mist to reflect a smidgeon of the solar constant.
BUT THAT'S CRAZY TALK! It could be done, it would cost nothing, and if it goes sour go back to clean kerosene and the problem dissipates in five years. Imagine miliiary and civilian air traffic aiding the whole world. Uncle Al says, "ACK! THBBFT!"
The cooling resulting from Pinatubo's eruption wore off quickly. If the SO2 is just injected into the stratosphere, a mount Pinatubo every year (or more) would be required. All that SO2 comes back down into the air we breathe. Most scientists researching SO2 injection now agree it falls out of the stratosphere too quickly. So now they're looking into injecting it into the mesosphere.
In either case - SO2 injection has several weaknesses - (a) it doesn't address ocean acidification, and (b) it will probably (more research needed) result in much dryer weather in the mid-latitudes. Since drying of the mid-latitudes is one of the larger dangers of global warming (probably the most important directly affecting the US) SO2 may not be terribly effective.
There's nothing particularly crazy about the idea; it's one of the few geo-engineering ideas for which the past gives some guidance about what might happen (e.g. historical and paleo-climate records of how volcanoes affected climate). But further investigation has turned up some serious weaknesses.
The fundamental reason why none of this will work is that it's based on pseudoscience. The danger is that we might accidentally stumble on something that actually has some sort of effect. That all these approaches would waste resources is not a danger, it's a certainty.
Indicative of the intellectual level of these proposals is the assumption that anthropogenic aerosols significantly reduce global average temperature (all the sulfur stuff above). This is the explanation given for why observed warming in the past is only half to a third of that predicted by the models: the greenhouse warming is being masked by anthropogenic aerosols. But the great majority of anthropogenic aerosols are generated in the Northern hemisphere, and aerosols are the opposite of well-mixed. So you would expect warming to proceed faster in the Southern hemisphere while lagging in the Northern. The exact opposite is observed. Further, we should see regional effects that would be impossible to miss, but see nothing remotely like that.
But noticing the obvious would leave us with no explanation for the (model-observed) discrepancy. So everyone acts as if anthropogenic aerosols have large cooling effects no matter how obvious it is that they don't.
And you want to make that the basis of geoengineering?
The expectation that SO2 aerosols will produce cooling is not based on the history of the effects of anthropogenic aerosols. Many anthropogenic aerosols are not SO2, and some, such as black carbon, produce warming rather than cooling. Instead, it's inspired by the effects of large volcanic eruptions, like Mt Pinatubo in 1991, or Mt Tambora in 1815. (Volcanoes also produce many other sorts of aerosols, but those have much shorter lifespans in the atmosphere.)
It's true that SO2 aerosols are not well-mixed (unlike CO2) and this does raise some serious problems (which I ignored in previous posts due to length). It requires the injection of the SO2 aerosols be carefully distributed throughout the mesosphere. Fortunately, there is relatively little weather in the mesosphere, so the carefully designed even distribution probably won't require a lot of maintenance - but the maintenance will be non-zero, and worse, it will be proportional to the amount of CO2 in the atmosphere. (Since CO2 is relatively long-lived in the atmosphere, and SO2 aerosols are not, even when injected into the mesosphere, this is potentially a serious problem unless CO2 emissions are reduced to zero.)
However - these problems do not make SO2 injection 'pseudoscience'. On the contrary - the very reason these problems are known to exist is because SO2 injection has been studied scientifically.
Most of the world's land surface area is in the Northern Hemisphere. The Southern Hemisphere, in contrast, is dominated by ocean. Water has a much higher heat capacity than land. (Which is why bodies of water are often cited as moderating influences on climate.) This would make the land-dominated Northern Hemisphere warm faster than than the Southern, if all other things were equal.
People need to call the geo-experimenting. There has never been previous engineering work at this level, they have no idea of all the variables involved. This is not engineering.
> .) This would make the land-dominated Northern Hemisphere warm faster than than the Southern, if all other things were equal
And the regional effects of which no sign is seen?
The point is, not only do we not know what effect aerosols have, we're dealing with people who are willing to assume they have effects they can't possibly have. But in the case of SO2 aerosols, trust them, they know what they're doing.