There are a number of ideas for how people might intentionally alter the planet’s climate system – an approach called geoengineering. One of the most frequently discussed ideas is solar geoengineering, blocking some of sun’s energy by, for example, injecting tiny particles called sulfate aerosols into the atmosphere. But solar geoengineeering remains a controversial method of addressing climate change.
And while much has been written about its potential benefits and its potential drawbacks, relatively little work has been done systematically evaluating those costs and benefits.
We find that solar geoengineering can play an important role in the near-term in keeping both costs and temperatures low. But in the long run, optimal policy must involve eliminating greenhouse gas emissions.
The Earth is getting warmer because of the greenhouse gases like carbon dioxide that humans are emitting into the atmosphere. The most direct way to solve this problem is to stop emitting those gases.
There are two problems with that solution, though. First, it isn’t cheap or easy to do so right now, since so much of our economy depends on fossil-fuel-fired, greenhouse-gas-emitting energy. Second, even an immediate halt to all such emissions wouldn’t stop the warming that is already “baked in” to the planet, since greenhouse gases remain in the atmosphere, warming the planet, for decades.
There is a solution to global warming that avoids both of those problems: solar geoengineering (SGE). SGE can reduce temperatures by blocking a fraction of incoming sunlight in order to offset the warming caused by greenhouse gases. The sunlight can be blocked by creating sulfate aerosols (sulfur-rich particles) in the stratosphere to act as tiny reflectors. This mimics what happens to the climate naturally after large volcanic eruptions.
SGE is cheap – up to 100 times cheaper than reducing greenhouse gas emissions. It can work fast – reducing temperatures nearly instantaneously.
However, this technique comes with drawbacks. It does not reduce greenhouse gas concentrations in the atmosphere or the oceans, and thus it does nothing to address ocean acidification. It may also have unexpected negative side effects, like affecting tropical monsoons.
Models and scenarios
To evaluate these costs and benefits, we use a tool called an integrated assessment model (IAM). An IAM includes both an economic model and a climate model to analyze the effect of the climate on the economy, the effect of the economy on the climate, and of policy interventions.
For example, an IAM will tell you what the optimal level of emissions reductions is once you input the costs of reducing emissions and the damages caused by temperature increases. IAMs typically demonstrate a “ramping-up” of climate policy – modest levels of emissions reductions now, with increased intensity in the future.
However, IAMs typically do not model solar geoengineering. We modified a commonly used IAM called DICE to include SGE as an alternative policy option. In the original version of DICE, the only policy available is abatement – reducing carbon dioxide emissions. In our modified DICE model, policymakers can also choose to use SGE. SGE will reduce temperatures, and it will do so more quickly than will abatement, but it will not reduce carbon concentrations.
Although not the first to incorporate SGE, our model is the first to systematically solve for the optimal mix of abatement and SGE. Also, we carefully calibrate the costs, benefits, and risks of SGE based on the most recent state-of-the-art scientific analysis.
This is important because substantial uncertainties remain, in particular over the side-effect damages from SGE, like the effect that a shortened monsoon season could have on food production. For these highly speculative damages, our model is more conservative. It assigns even higher values to the damages from SGE than previous models.
Our model also captures the crucial fact that, because SGE does not reduce carbon concentrations, it cannot remedy any damages that come directly from rising levels of carbon, including ocean acidification. In other words, we modeled SGE as a substitute for abatement, but recognized it as an imperfect substitute.
The results from our base-case analysis confirm what others have written about optimal SGE deployment: it can “buy time” by allowing us to defer costly abatement while keeping temperatures in check. But it cannot fully replace the need to eliminate carbon emissions.
In the image above, we simulate optimal policy far into the future, over the next 500 years. Two different scenarios are represented. In the first scenario, SGE is omitted, and the only policy option is abatement. This is shown in the red curve, which gives the optimal abatement intensity over time.
The second scenario is our base case, which includes both abatement and SGE as policy options. The blue curve represents optimal abatement in this scenario, and the green curve is the optimal intensity of SGE.
The graphs show that when allowed, SGE is used in the near term, and as a result, the amount of abatement that is required is slightly less than in the case where SGE is omitted. When SGE is used, the optimal abatement intensity (expressed as a percentage of carbon emissions that are abated) is up to 25% lower compared to the scenario where SGE is omitted (the difference between the red curve and the blue curve). However, even in the base case where SGE is used, abatement is also used and eventually reaches 100% – all carbon emissions are eliminated. The date at which this occurs is pushed back by about three decades when SGE is used – this is the “buying time” of SGE.
Although more carbon is being emitted into the atmosphere when SGE is allowed, temperature increases are kept substantially lower. This is shown in the above graph, which compares temperature increases (in degrees Celsius relative to preindustrial temperature) under our base case scenario and under the no-SGE scenario. Without SGE, temperature peaks at about 3.5 degrees Celsius above pre-industrial levels, while SGE keeps this increase to just 2C (3.6 degrees Fahrenheit). (Note that the two-degree optimal temperature increase is derived from the parameters of the model, and it is only a coincidence that it matches the two degrees target that is often discussed by policymakers and the media.)
SGE can also be an effective way to prevent the climate from reaching a tipping point, an irreversible and very costly change in the climate system caused by global warming. Like much about the climate system, there is a great deal of uncertainty about these tipping points: whether they exist, at what temperature they are reached and how costly they would be once reached.
In an extension to our main analysis, we include the possibility of reaching a tipping point. We model three different types of tipping points and consider three different policy regimes for SGE: ban it, allow it or allow it but only after the tipping point is reached. We find that allowing SGE from the onset is an effective way of managing the risk of reaching a tipping point. Banning SGE or allowing it only once the tipping point is reached only increases society’s exposure to tipping point risks, since by definition the tipping points are irreversible.
Our analysis provides concrete, numerical justification for many of the policy ideas that have been merely speculative until now: solar geoengineering can reduce costs, suppress temperature increases, and lessen tipping point risks. But, SGE is not a perfect substitute for abatement, so eventually we must transition to a carbon-free society.
We are not nearly ready to begin solar geoengineering on a large scale, but it seems likely that solar geoengineering will be an important part of our optimal policy toolkit to combat global warming.