The great moral challenge of our time by David Atkins

1. How much can global average temperature rise before we risk "dangerous" changes in climate? The current consensus answer is: 2 degrees C [3.6 degrees F] above pre-industrial levels.

The 2 degrees C number has been around for over a decade and was reaffirmed by the Copenhagen Accord just last year. Deciding on an "acceptable" level of temperature is a political and somewhat arbitrary judgment, of course, since it lets one number stand in for a wide range of heterogeneous considerations. But it's an important marker. And when it was first developed, it was based on the science of the day...

Right now, global emissions are rising, faster and faster. Between 2000 and 2007, they rose at around 3.5 percent a year; by 2009 it was up to 5.6 percent. In 2010, we hit 5.9 percent growth, a record. We aren't just going in the wrong direction -- we're accelerating in the wrong direction...

As you can see, if we delay the global emissions peak until 2025, we pretty much have to drop off a cliff afterwards to avoid 2 degrees C. Short of a meteor strike that shuts down industrial civilization, that's unlikely.

How about 2020? Of the available scenarios for peaking in 2020, says Anderson, 13 of 18 show hitting 2 degrees C to be technically impossible. (D'oh!) The others involve on the order of 10 percent reductions a year after 2020, leading to total decarbonization by 2035-45.

Just to give you a sense of scale: The only thing that's ever pushed emissions reductions above 1 percent a year is, in the words of the Stern Report, "recession or upheaval." The total collapse of the USSR knocked 5 percent off its emissions. So 10 percent a year is like ... well, it's not like anything in the history of human civilization.

This, then, is the brutal logic of climate change: With immediate, concerted action at global scale, we have a slim chance to halt climate change at the extremely dangerous level of 2 degrees C. If we delay even a decade -- waiting for better technology or a more amenable political situation or whatever -- we will have no chance.

It might seem that, given the extraordinary difficulty of hitting 2 degrees C, we ought to lower our sights a bit and accept that we're going to hit 4 degrees C. It won't be ideal, but hitting anything lower than that is just too difficult and expensive.

It's seductive logic. After all, to hit 4 degrees C we would "only" have to peak global emissions in 2020 and decline thereafter at the relatively leisurely rate (ha ha) of around 3.5 percent per year.

Sadly, even that cold comfort is not available to us. The thing is, if 2 degrees C is extremely dangerous, 4 degrees C is absolutely catastrophic. In fact, according to the latest science, says Anderson, "a 4 degrees C future is incompatible with an organized global community, is likely to be beyond 'adaptation', is devastating to the majority of ecosystems, and has a high probability of not being stable."

Yeeeah. You'll want to read that sentence again. Then you'll probably want to pour yourself a stiff drink.

Obviously, "incompatible with an organized global community" is what jumps out, but the last bit, "high probability of not being stable," is equally if not more important. One of the most uncertain areas of climate science today has to do with feedbacks -- processes caused by climate change that in turn accelerate (or decelerate) climate change. For instance, heat can melt the Arctic permafrost, which releases methane, which accelerates climate change, which melts more permafrost, etc.

Based on current scientific understanding, positive climate feedbacks -- the ones that accelerate the process -- considerably outweigh negative feedbacks. At some level of temperature rise, some of those positive feedbacks are likely to become self-reinforcing and effectively unstoppable, no matter how much emissions are cut. These are the "tipping points" you hear so much about.