Ice Age Modeling: Major Step Forward in Climate Change?

[phyzguy]

OK, I'll dive in and start a climate change thread. I'm fascinated by the Nature paper by Abe-Ouchi, et.al., which I already uploaded in post #21 of this thread. If you look at Figure 1, the model does an outstanding job of modeling the last several ice ages. I have been somewhat skeptical of past climate models because they were unable to model the ice ages, but it appears that now they can. So I have two questions:

(1) Do others agree that this model is a major step forward?

(2) Figure 2 shows that, according to their model, we should already have entered into a period of cooling and sea level decrease. Does the fact that we are continuing to warm and the sea level is continuing to increase help confirm the AGW hypothesis?

 

[D H]

(1) It appears to be a major step forward. This paper purports to solve the 100,000 year problem, which is that glaciation has followed a 100,000 cycle for the last million years and that eccentricity is the orbital element with the 100,000 year cycle. The 100,000 year problem is that eccentricity was perceived as a weak forcing. For example, eccentricity has a much smaller impact on seasons than does inclination.

(2) I think you're reading the figure wrong. Look at 2b. It's the North American ice sheet that's the main driver in glaciations. We are currently in the gray area of that graph, the point labeled 0 at insolation=478 (280 ppm CO2) / 498 (220 ppm) and ice volume=5e15. Think of that point as more or less equivalent to the initial point, the one marked 122. We've got a long ways to go before we have to worry about another glaciation.

That is a neat hysteresis process.

 

[D H]

Getting back to point (1), this shows the power of the global circulation models, particularly when coupled with accurate forcing models.

If I'm reading the paper correctly, it even suggests a reason for the switch from a ~40 kyr cycle that used to be dominant before a million years ago to our current ~100 kyr cycle. "In contrast to the North American ice sheet, the dominant cycle of the volume of the Eurasian ice sheet has a period of ~40 kyr, and the volume never grows beyond 40 m sea-level equivalent." Right now it's the North American ice sheet that drives the cycle. Presumably it was the Eurasian ice sheet that dominated before a million years ago.

 

[phyzguy]

(2) I think you're reading the figure wrong. Look at 2b. It's the North American ice sheet that's the main driver in glaciations. We are currently in the gray area of that graph, the point labeled 0 at insolation=478 (280 ppm CO2) / 498 (220 ppm) and ice volume=5e15. Think of that point as more or less equivalent to the initial point, the one marked 122. We've got a long ways to go before we have to worry about another glaciation.

Clearly a major re-glaciation is thousands of years in the future, because it takes several 40,000 year cycles for the ice to build up. But in Figure 2b, if you look at the trajectory in the left hand graph, the track is moving to the left and (slowly) upward. This says the temperature should have begun falling about 10,000 years ago, and the ice volume should have begun increasing (albeit very slowly) about 5,000 years ago. What am I missing?

 

[D H]

That's the insolation at the top of the atmosphere on June 31st at 65N latitude, not the mean annual temperature worldwide. Temperatures in the far north reached a maximum 8 to 10 thousand years ago. The world as a whole reached a maximum somewhere between 5 to 8 thousand years (the Antithermal, or Holocene Thermal Maximum), about a degree warmer than now. Sea levels have been fairly flat over the last 8000 years.

Keep in mind that successive dots in the curves in figure 2b represent the passive of two thousand years.