Leave the clouds alone
Might as well be blocking out the Sun: If you're willing to take the threat of rising global temperatures seriously, then you'll want to understand the full spectrum of solutions available. As Bill Gates put it in a 2014 interview with Rolling Stone: "[W]e have a real problem, and so we should pursue many solutions to the problem. Even the Manhattan Project pursued both the plutonium bomb and the uranium bomb -- and both worked!"
■ It's broadly agreed that the root cause of the problem is atmospheric carbon dioxide levels, which have risen dramatically since the Industrial Revolution. As long as we keep on burning carbon-based fuels, the combustion is going to produce carbon dioxide.
■ De-carbonization is the ultimate goal -- if we perform less combustion, there ought to be commensurately less carbon dioxide produced. Animals are still going to breathe, but hypothetically, we can basically zero out artificial carbon dioxide production by converting from combustion engines to electric motors and powering those motors with energy that comes from non-carbon sources. Some emissions will still come from agriculture, but more than three-quarters of US greenhouse gas emissions are from transportation, electrical generation, and industrial use.
■ Here's the problem: The United States is already decreasing per-capita carbon dioxide emissions, and if we start combining renewable electricity with better batteries and expanded nuclear power generation, we can get to a fairly near-zero state. Expanding nuclear power generation remains a big "if" -- capacity hasn't increased in 35 years -- but serious people must take the solution seriously. But even if the USA went to zero-carbon tomorrow, China's use exploded over the last two decades, and India's use is growing. With many more people, their per-capita effects are magnified relative to the cuts the United States can make.
■ There is also the question of equity: If it's more expensive to de-carbonize (or to grow into a low-carbon advanced economy from the start), then can the willpower be found to join in an international effort that might (at least at the outset) appear to put poorer economies on a slower growth rate? Equity is even a contested matter within countries, too.
■ A "many-solutions" approach to global warming requires us to consider harm-mitigation measures on top of efforts to avoid the problem. We may be able to slam the metaphorical brakes, but the collision could occur anyway and we'll want our seat belts fastened. Bjorn Lomborg has famously argued that climate-change consequences are bad but manageable, and that we're better off investing in mitigating the effects rather than preventing it altogether. Other smart people say we have to do both, and urgently.
■ Among the ideas floated to manage the situation is the radical proposal to try geoengineering the planet to create more clouds to block the incoming sunlight. All choices tend to come with unintended consequences, but that one in particular seems fraught. Among other problems, it's quite possible that miscalculating or carrying out the geoengineering in the wrong way could actually make the problem worse.
■ After all, the idea of trapping solar heat with the help of human-made clouds is the backbone of proposals to terraform Mars to move it from "cold and uninhabitable" to "marginally Earth-like". (The technology, alas, is not in our hands yet.) But that's what clouds do -- they reflect off their tops (albedo), but they also trap existing heat. And, of course, if an experiment goes wrong high in the atmosphere, the effects could easily become global. (Again, if the consequences weren't far-reaching, they wouldn't be worth trying against a problem the size of global warming.)
■ Considering the global risk we already face from unpredictable volcanic eruptions devastating agriculture across most of the planet, it's probably best to keep geoengineering behind a glass to break only in case of the most dire emergency. Other approaches may be difficult, costly, and hard to mobilize on a planetary scale -- but they also tend not to come with a non-zero level of existential risk.
