Thursday, August 23, 2007

Discussion of laws and macroecology

Hello everyone,
First I want to thank you all for contributing to a great discussion. Please continue to read the papers so carefully. We look forward to next Tuesday's look at human life history and the EEA.

I'll mention a few of the things that came up today and then add a few points that we didn't get to.

We spent a lot of time on Brown chapter 1. Figure 1.1 generated interest and confusion and we just want to point out that while each data point in the figure represents an individual mountain top it would be impossible to conduct a detailed study of the flora and fauna on each mountain because of time and money constraints. It is also probably not feasible to simply raise the temperature or cut down the forest at the tops of mountains to see how species respond to habitat loss. Those sorts of pragmatic features of the approach are what the example was meant to illustrate. By using the species-area relationship we can get a feel for what the average tradeoff, across mountain tops, might be between area and number of species. Note also that this prediction is very imprecise with a range between 9 and 62% of species lost. The loss of detail may often be accompanied by a loss of predictive power or precision. How should we feel about this? However, this prediction was also obtained after spending just a few days to weeks at the library, so as a rough low cost estimate it may be a reasonable place to start. It also is simply an example, good or bad depending on your reaction, of answering a question with an inductive, non-experimental, library-based analysis. (Inductive because it makes a general statement of a population based on the properties of a sample to the effect that we consider the statement potentially valid for the population of all mountain tops in the Great Basin. Also, the premises don't guarantee the truth of the conclusion, as in deductive logic where if the premises are true the conclusion must be true.)

One subject from the Brown chapters that could use further thought and discussion is the link between macroecology and complex systems and the topic of emergence. If you are not familiar with emergence, the wikipedia entry is quite good. In simple terms lets say that the first step in recognizing an emergent phenomenon is experiencing a level of surprise. You observe a pattern that seems to not follow from the characteristics of the entities that comprise the pattern. Given what you know about what you study, you didn't expect there interaction to lead to what you just found. This is often stated as the whole being more than the sum of (or different from) its parts. The important link between emergence and macroecology is seen in Brown's quote on page 11 - in that macroecology seeks to "develop more powerful macroscopes that will reveal emergent patterns and processes."

Brown lists 6 features of a complex adaptive system that would be good to think more about, especially in the context of human systems, that are relevant for thinking about macroecological patterns.

We should also keep in mind Brown's 5 characteristics of macroecology, page 18 - 20.

Brian Maurer, in his discussion of complexity and structure, points out that when the parts of a complex system interact the system may gain structure and inertia. In Brown's example of the gas diffusing through the room, the more the gas particles interact, by bumping into each other in the air, the more diffuse they become and hence the less structured. On the other hand, as with the plot of human territory size we showed in class, the more the humans in those systems interact the more structured they become. Lots of living systems have this property of structure being enhanced by interaction and this feature changes the behavior a lot and it should change our perspective as well. In Maurer's own words, when he talks about structure he means " the entities within the system interact or relate to one another in a consistent and stable manner so that the overall persistence of the system is enhanced.” (quote from his book, "Untangling ecological complexity, page 27).

The main point from Ginzburg is that ecology has laws and we've only been wrong about the way we think about laws and in our conceptions of what they should do. Physical laws are often defined in a vacuum using the concept of a limit myth. With a precise physical law you can exactly explain behavior on a frictionless plain or in a vacuum but the real world is full of frictions and external forces so these 'laws' often have to be adjusted to account for such forces. Sometimes this can be done perfectly but other times there will be exceptions or noisy aberrant behavior to the system. This does not mean that the law is false, only that we haven't accounted for the friction. Is it possible that ecologists and anthropologists study systems that experience a lot of 'friction' in that they are acted upon by an exceptional number of external forces and thus accounting for them all can be quite difficult?

Also recall their example of the Titius-Bode law that ended up not being a law at all. We may have to chase after a pattern for a while before we find out that its just a curious fact and not the law we thought it was. Is there some way to avoid this potential? Plugging the pattern into existing explanatory theories that can make deductive predictions might help.

(ps - its very very unlikely that most of our blogs will be this long...)

See you Tuesday,
Oskar

2 comments:

dtinucci said...

Preparing and presenting research is one thing, but having it supported by weighted references is another. In this respect, I find the two articles complimented each other very well, but was expecially grateful for the approach taken by Ginzburg and Cloyvan and I now look forward to using their argument sustain future proposals.

paul said...

John Wheeler, Feynman's advisor in physics at Princeton said, "There is no law except the law that there is no law."

He also said, "In any field, find the strangest thing and explore it." Sounds good for human ecology too.

 
Locations of visitors to this page