[MUSIC] So in thinking about those temperature targets, it turns out that a lot of the complexity of the carbon cycle goes away. And that's illustrated by these sort of schematic cases of releasing a slug of carbon, either doing it very quickly, all at once, or doing it slowly, but for a longer time, so that the total amount of carbon released is the same in either case. So, there's a fast slug of carbon released and a slow slug of carbon released. So if you release the carbon quickly that reaches the CO2 peak, by the time you, at the end of the slug, and then it sort of slowly drifts down as it's being neutralized by the oceans and what not. If you release it more slowly it reaches a peak that's later, and probably a lesser peak because as you've been releasing it, it has had time to dissolve in the oceans. But the temperature response goes as the CO2, but there's a lag time for the change in temperature of another 1,000 years, because you have to heat up the ocean. So it turns out that the temperature sort of rises up to a plateau, and then it basically just stays there. As far as we're concerned, it stays there forever. But the crucial part to this is that the plateau temperature that it reaches for these two scenarios is about the same in either case. It turns out that the higher CO2 is counter-balanced by the fact that it started dropping, by the time the Earth's temperature is catching up. And so the time dependence of the absorption of CO2 is kind of compensated for by the time dependence of the warming. What this means, what it boils down to is that in terms of the temperature that you're going to reach, it doesn't matter if you release it quickly or slowly, you're going to reach the same temperature in either case. You'll reach it more quickly or more slowly depending on how quickly your release it. But it's a much simpler picture now because we can just draw a one to one relationship between the amount of carbon ever burned and the temperature that the Earth will reach. So it turns out that about 1,000 gigatons of carbon would take the Earth about two degrees C threshold. 500 gigatons of carbon would take it about to one degrees C. So for comparison, we've already burned about 300 gigatons of fossil fuels, and cut down about 200 gigatons of trees. So we're kind of in a one degree C area already and we're about half way to this two degrees C mark. So what would it look like to actually try to cut emissions and stick to one of this say 1,000 gigatons of carbon as a target. Well, here is a plot of emissions of carbon per year. And business as usual, is this line here, it's growing at a rate of 3% per year. That's the trajectory that we continue to be on. And the total amount ever burnt goes as the area under this curve. So let's say in 2010 we had decided instead of growing at 3% per year we would cut emission by about 2% per year. That way the emission flux would sort of drop in this gradual sort of way, and the total area under this curve would come in at about 1,000 gigatons of carbon. If we wait longer, we have to cut more quickly in order for the total area under the curve to stay under 1,000 gigatons. So here's a plot of the percent per year cuts that would be required in order to keep to 1,000 gigatons of carbon for the total slug as a function of the year that they start. So, starting in 2010 cuts of about 2% per year would have done it. 2020 we would have to cut by 4% per year, 2030 about 10% per year and pretty soon by around 2040 or so, basically you would have to cut cold turkey to keep under 1,000 gigatons of carbon. So it's kind of like, you're in an army and your sergeant, he's caught you doing something wrong, he says, drop and give me 20 push-ups. Or, if you don't want to do them today, you can do them tomorrow but you've gotta do 40, or you can do them this weekend, but you've gotta do 100. So what are you gonna do? It's much easier, cheaper to convert our energy infrastructure at a slow rate of a few percent per year, because that's kind of the rate at which power plants become obsolete and need to be replaced anyway. Whereas if you wait until the end, you've got to basically build everything all from scratch all at once. It's a much bigger and more expensive proposition. [MUSIC]