Wednesday, February 28, 2007

A GUT of ecological scaling?

OK. I'm not going to pretend that I grasp the details of this, but here goes.

A recent paper in Physical Review Letters by Jayanth Banavar et al. presents a general theory of (macro)ecological scaling.

That is, they offer a common explanation for patterns such as that between the area of a place and the numbers of species that live there, area and total biomass, the number of organisms and species of different sizes, the maximum body size of the biggest species in an ecosystem (see here for another view of maximum carnivore size), and relative abundance — the spread of rare and common species.

All these scaling patterns seem to follow power laws, at least over partially. That is, they take the form y=ax^n, and plot as straight lines on a log/log graph.

Here's the abstract:
Scaling provides an elegant framework for understanding power-law behavior and deducing relationships between critical exponents. We demonstrate that scaling theory can be generalized to develop a framework for the analysis of diverse empirical macroecological relationships traditionally treated as independent. Our mathematical arguments predict links between the species-area relationship, the relative species abundance and community size spectra in excellent accord with empirical data.

As far as I can tell, Banavar et al. present a scaling hypothesis, based on a probability distribution of the mass and abundance of different species. Using this, and a set of reasonable assumptions (such as, that an ecosystem's total population and mass is proportional to its area) they derive the relationships described from a single starting point.

Blimey. If all these things can be brought under the one roof — this is the first such attempt that I know of — that's a big deal, I think. These are fundamental ecological parameters. Together, they pretty much sum up most of the questions that community ecology seeks to answer. Even the metabolic ecology models of West et al. (with which I am more familiar) have steered clear of tackling the relationships between species diversity and area, and between body size and diversity and abundance.

But, although much of the criticism of metabolic ecology focuses on its generality, Banavar and his colleagues have an interesting history of seeking yet more general models for the scaling of metabolic rate with body size. Banavar has also worked with Steve Hubbell to develop the neutral ecological theory. It's interesting that this is in PRL. I wonder what a biology journal would have made of it.

Thanks to Phil Ball for bringing this to my attention. I think. If anyone has any thoughts on this, I'd love to hear them.


postblogger said...

Nope; no idea at all.

But I am still alive and does this have any bearing at all on the power-laws which are increasingly in vogue in 'network' theory and which have recently come under fire for being less revealing than previously suspected, for (possibly) similar reasons to those covered here, i.e. certain systems throw them up and they don't necessarily reveal that 'deep' secrets?

Anonymous said...

Thank you for your interesting post!
I thought perhaps you may also find this related post interesting to you:
Body Size and Human Longevity