Thursday, January 27, 2005

Turning Up the Heat

I missed a speech on global warming at OU yesterday, so I'll have to depend on whatever I can learn from the web for information.

Here's a link to a story that says the threat of global warming is twice as bad as previously predicted.

If sea levels are going to rise by 20 feet, I'm glad I left Houston for Oklahoma City. If the tipping point comes in my life time, I think I'd rather be somplace with a climate like the Sahara Desert than to be submerged in a place like the lost city of Atlantis.

3 comments:

Harold Brooks said...

I made it to Kevin Trenberth's talk. It's part of a new University of Oklahoma "dream course" on climate change, where experts come in to give invited lectures for undergraduates. There will be 3 others in the series, the next being on 23 February, when Syukoro Manabe, the "dean" of climate modelling from the Geophysical Fluid Dynamics Laboratory in Princeton will come here.

Trenberth talked about changes in the water cycle of the planet. Even without changing the total amount of precipitation at a location, there's a suggestion that it will be (and, possibly is already) redistributed into fewer rain events, but with heavier precip when they occur. The physical mechanism runs like this:

1. Increased temperature
2. Increased absolute humidity (total water vapor in atmosphere)
3. Increased rain when it rains
4. Longer time to "recharge" moisture

There's observational evidence for step 1, theoretical and observational evidence in the US for step 2, possible evidence for 3, and for 4, it's not clear.
The other interesting point that Trenberth made was the possible impact of changing the phase of precipitation, with, in a warmer climate, more precip falling as rain, rather than snow, and earlier melting of snowpack. In the western US, winter snowpack is desperately important for water resources. If the snowpack decreases or melts earlier, it's likely that summer droughts in the west may become more frequent and critical.
As far as the paper referred to in Bruce's post, I've been amazed at the press coverage, which has missed the main point of the experiment. climateprediction.net is a project dreamed up at Oxford and The Metoffice in the UK to run a reasonably complex global climate model on personal computers around the world, taking up those unused cycles overnight. This allows them to run a large number of simulations with different conditions and different physical packages in the model in order to make an estimate of the probability of things happening. The paper in Nature is the first publication out of the project, using >2000 model runs that go out for 45 years. (They aren't as high of a resolution as the "best" climate model runs and they can't see small features, but the fact that you can run a massive number of them cheaply is a big help.) The primary goal of the work is to try to define the envelope of likely changes and to explore the sensitivity of the model. The current estimates are for warming on the order of 1.4 C to 5.8 C by the end of the century. What the Oxford/Metoffice group has done is to models are more sensitive than previously known. Their median prediction of change is 3.4 C, not that different than the latest IPCC asseessment, but they've got a ~10% change of a greater than 8 C increase. What they've shown is that the distribution of probability of changes is nothing like a normal, bell-shaped distribution. Rather, it's skewed, so that there's very, very little chance of a warming less than 1.9 C.
The emphasis in the paper isn't on the most extreme event, which is what the media coverage has been about. It's about putting bounds on likely changes and I believe this work has increased the confidence we can have in warming on the order of 2.5-4.5 C. I'm sure that at least one of the authors (my former roommate in grad school) is having convulsive fits over the emphasis on the extreme events in the media.

Dr. Bruce Prescott said...

Harold,

One of the articles I read said the tipping point for major climate change was a rise in the average of 2 C. If I?m reading you right, you?re saying there is evidence that the possibility of rise less than 1.9 C by the end of the century is slim and that one of the best guesses is around 3.4 C.

Is the debate just over the tipping point and its effects? If 2 C is too low, what about 3.4 C?

Harold Brooks said...

I'm not convinced that we can identify a temperature that represents a tipping point. It is a nonlinear process and there could be something like a regime change and we go off into previous rarely visited climatic phase space. The problem is that the processes are very nonlinear and, as a result, are likely to have sensitivities that we don't really understand very well. Obviously, that could lead to things being worse than a less nonlinear system (such as the numerical models) suggest, but it could also lead to them being better.

That said, I think the real issue is rate of change. Slow changes can be dealt with reasonably easily. Climate has slowly varied over the history of civilization and, for the most part, people have coped. I have no doubt that we are capable of adapting to the end state of a doubled-CO2 world without trouble, for the most part. The problem becomes if change occurs on time scales of ~50 years. Natural systems (particularly plant life) may have a difficult time, given that they may not be able to move to new locations or adapt to new stresses. Large infrastructure commitments for human societies also become a challenge on that kind of a time scale. If water is going to be a primary stress in the western US, we could build a large number of dams to manage the resource and create non-fossil fuel power. Wind farms in the Plains could become more common. Nuclear energy could return. All of those also have negative environmental consequences and making policy to balance those choices is really, really hard. If we go down certain avenues, such as dam-building in the West, that can't be done in a couple of years. The funding commitment and the time to do the preparation and building work is likely to be on the order of a decade or more. If change is relatively rapid, you have to make decisions in the face of great uncertainty. People do it all the time, but it's not something that society likes to do when the resources required are large. One of the problems with running a large federal deficit is that it limits flexibility to do things. If inconvertible proof arrived tomorrow that southern California would become prohibitively expensive to provide water for in 50 years because of climate change and that mitigation wasn't possible, we'd have a very difficult time finding resources to fund adaptation.

It's quite possible that many aspects of a CO2-enriched planet would be preferable to the current climate in many locations. The Soviets certainly believed that in the 70s and 80s and, as a result, wanted to speed up the transition. The problem is that the transition process may require us to adapt in ways that are difficult to do quickly. Societies tend to move slowly, which is frequently a good thing, but can be disastrous when rapid movement is required.