« Sweden To Go Oil-Free (Revisited) | Main | The Wave Hub »

A New Model for Understanding the Climate

Duke University's Adrian Bejan, along with colleagues from the University of Evora in Portugal, has discovered something potentially quite important: a recently-developed theory of optimizing flow configurations over time called Constructal theory can be used to model key parts of the global climate, and do so using only a small number of well-known inputs. Moreover, this theory could be used to build models of changes to weather patterns resulting from greenhouse gas accumulation. What makes this notable -- and possibly worldchanging -- is that Constructal theory is shaping up to be a universal physics principle applicable to everything from traffic flows to the evolution of the circulatory system. This is kind of abstract, but bear with me -- this could be a major discovery.

Constructal theory, developed in the late 1990s by Dr. Bejan, is the formalization of a superficially obvious notion: "For a flow system to persist in time (to survive) its configuration must evolve (morph) in time in such a way that it provides easier flow access." That is, over time, physical structures and processes evolve towards optimization of the imperfections of a system; because conditions of a system can change, such optimization may never be complete. Constructal theory takes this concept and turns it into a series of mathematical and geometric principles. The principles, in turn, have been able to predict a variety of well-known physical and biophysical laws, from the proportion of metabolic rate and size to the relationship between optimal cruising speed and mass of flying bodies. Constructal theory isn't just predictive, though; it can be used as a design application, and has been employed as a tool for optimizing travel time for people in buildings such as airports.

What Bejan and his colleagues have now done is to demonstrate that Constructal rules apply to the flows of heat in the atmosphere, and that fundamental climate systems can be derived directly from the theory. The researchers needed just four inputs: temperature of the sun, the solar constant, cloud cover and the Earth’s greenhouse factor. They believe that they will also be able to use the model to predict the effects of changes to these inputs:

The researchers found that the “constructal theory” can predict the global circulation that determines the boundaries between desert and tropical forests as well as between temperate zones and the poles. Based only on the optimal flow of heat from the sun, the theory also predicts other climate characteristics, such as average wind speed and the average temperature difference between night and day. [...]

“While other very complicated empirical models predict the same basic features, constructal theory does this in a much simpler way,” said geophysicist A. Heitor Reis of the University of Evora. “This is an entirely new kind of approach to climate.” [...]

To apply the constructal theory to global climate, the researchers treated Earth as if it were a heat engine that, rather than doing work, dissipates all the power it produces through air and water currents.

“The Earth with its solar heat input, heat rejection, and wheels of atmospheric and oceanic circulation, is a heat engine without shaft,” Bejan said. “Its maximized mechanical power cannot be delivered, but is instead destined to dissipate through air and water friction and other forms of heat loss. It produces maximum power, which it then dissipates at a maximum rate.”

By applying this analogy, the researchers predicted the main characteristics of global circulation and climate based on very few inputs – namely the temperature of the sun, the solar constant, cloud cover and the Earth’s greenhouse factor. The solar constant refers to the amount of incoming solar radiation measured on the outer surface of Earth's atmosphere. The greenhouse factor takes into account the concentrations of aerosols and greenhouse gases to determine the amount of heat energy trapped by the atmosphere.

“To my surprise, a simple theory anticipates the latitudinal boundaries of the three zones – the Hadley, Ferrel and Polar cells – which comprise the main global circulation on Earth,” Reis said. The theory also predicted the average speed of atmospheric and oceanic flow and the average temperature on Earth, among other climate features, they reported. [...]

The new climate theory may also aid in predicting the consequences of global change, the researchers said.

“If the properties of the atmosphere change as people say they will, we could anticipate what that might mean for global climate,” Bejan said.

“By playing with the Earth’s greenhouse factor, we could determine what it would take to get another result,” Reis added.

So what does all of this mean?

The most directly relevant application of this finding is that we could see how changes to the greenhouse factor trigger larger changes in the global climate. It may not give us precise predictions as to the effects on a particular location, but if this observation is borne out, it could give us a much stronger sense of how close we are to a climate "tipping point." Climate scientists worry that we could be getting close to a level of greenhouse gas accumulation in the atmosphere that would lead to a major shift in the overall global climate to a new (and likely unpleasant) stable configuration. If the Constructal model of the global climate shows a phase shift resulting from increased greenhouse gases, we'll have a better lead on how much more CO2 we can add to the atmosphere before things fall apart. In turn, this will tell us whether a slow/steady mitigation plan will work, or whether we need to do a major try-anything campaign to get greenhouse gas levels down.

There's a bigger-picture application, too. This finding is another piece of evidence that the Constructal theory describes a broadly-applicable law; what other applications can we imagine for the model? If it works for understanding traffic (either pedestrian or vehicular), could it be used to optimize trade flows? Would it have an application to the management of global financial markets? What does a Constructal version of urban design look like? What does a Constructal approach say about maximizing energy efficiency in a building, or flows of fuel and exhaust in a vehicle?

Constructal theory may, in the end, turn out to be less useful than it now appears, and -- who knows? -- its predictions may be little more than coincidence. But I find myself drawn to a model that shows an underlying similarity of processes across seemingly disparate arenas, and wondering what else we have yet to discover about the way the world works.

Comments (7)

Constructal theory can be used to model key parts of the global climate, and do so using only a small number of well-known inputs

CNN was going to lead with a story about an attractive white girl who'd been kidnapped, but I think this might trump it!


I am very curious to know what effects such things as radiative forcing due to land cover/land use change has on these models. And, how this theory works at local/regional climate scales.

Jordan Dawe:

This is not quite so important for climate science as you make out; there are already several analytic methods that one can use to predict the (mean) location of the Hadley and Ferrel cells, such as calculating the balance between meridional pressure gradient and transport of angular momentum from the equator poleward. These types of models are generally not useful to examinations of global climate because the actual climate details depend on things like cloud formation rates and ice cover. You certainly can use this method to predict the effect of adding greenhouse gases, but you can do the same with the other methods as well, and they all basically say "The Earth heats up". We are long past this stage in climate science; now we are focused on exactly how much, and where, warming occurs.

While this doesn't do much to help climate science, it definately is a fascinating new idea for thermodynamics, and it makes a lot of sense when you think about it. So from the point of view of this being a new fundamental law of nature, this may indeed be worldchanging.

Jordan, it sounds like you might have more info on this than I do, but the impression that I got from the material at the Constructal theory website (as well as from the journal link) was that this would do more than simply say "it gets warmer." As I understood the research, the method would provide more insight into the system-level responses to greenhouse changes.

I may well be wrong on this, and the Constructal approach may just be another item on the pile of model-based evidence for global warming-induced climate change. Thanks for the comment.

Jordan Dawe:

Well, I just read the paper this morning, but I just went back through it and the more I look at it, the dodgier it appears. For instance, it is easy to show with a full atmospheric circulation model that the earth's rotation speed has a dramatic effect on the atmospheric overturning cells. Run a model with an increased rotation rate and you get more overturning cells, as it requires a greater north/south pressure/temperature gradient to override the coriolis forces that result from moving northward.

The constructal paper basically assumes that there are two maximal heat transfer latitudes, corresponding to the boundaries of the overturning cells, and then derives those latitudes. So it might help to predict, say, whether the locations of the Hadley and Ferrel cells will shift as the earth warms, but that doesn't really give a whole lot of information about the system level response to climate change that you can't derive just by knowing how the atmosphere works.

Again, this isn't a comment on the general applicibility of Constructal Theory, which in general looks very good to me. But the climate science in this paper seems like it needs some work.

Thanks Jordan -- that's very helpful!

Subbarao Seethamsetty:

If it is truly a fundamental law of nature applicable to systems in general and predicts time lapsed outcomes with relatively few primary inputs, it is a very exciting avenue to follow. How wonderful it would be to design and implement changes or deltas in inputs to effect modest and positive global changes - like poverty elimination, peaceful and sustainable cohabitation of all living thingies and most importantly to standardise 6 hour workdays, four days a week for nine months a year for everyone.


This page contains a single entry from the blog posted on February 9, 2006 1:13 PM.

The previous post in this blog was Sweden To Go Oil-Free (Revisited).

The next post in this blog is The Wave Hub.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34