McGARVEY ECOLOGY — Ecological Scaling

The number of biological phenomena that can be quantified with simple allometric or power-law equations is astonishing. My interest in these scaling relationships began when I was a post-doc at the USEPA's modeling lab in Athens, GA, at about the same time Jim Brown's 'Metabolic Theory of Ecology' really took off. I was helping a large team of ecologists and computer scientists to apply very complex, process-based models of freshwater ecosystem services and I found myself using the simpler ecological scaling models as a kind of model validation tool.

A good example of my approach is illustrated below. I began with empirical estimates of net primary production in representative streams, then used some basic structural/topological food web information and trophic transfer efficiency estimates to predict annual production at higher trophic levels (i.e., fishes or tertiary consumers). Production:biomass ratios were also used to estimate standing stock fish biomass. Finally, the classic self-thinning relationship, where population density is a power function of average individual body mass, was used to predict the density and regional abundance of select fish species (mostly sport fishes).

Currently, I am most interested in quantifying and applying individual-level scaling relationships, rather than species-level averages. Scaling models that incorporate individual body size data have proven useful in pelagic environments, where the high abundance of tiny planktonic organisms can make species-level identification impractical. Treating all individuals as ‘particles’ of given size, rather than discrete species, can greatly facilitate sample processing and has revealed robust, community-level scaling relationships between body size and the standing stock abundance or biomass of individuals (i.e., 'size spectra'). Individual-level data may also be preferable to species-level averages because many freshwater taxa exhibit ontogenetic shifts in feeding behavior.

I am now using a West Virginia streams survey as a pilot project to test individual-level scaling relationships for fishes and invertebrates. Preliminary analyses indicate that the slopes of biomass size spectra from these streams are very similar to characteristic slopes from pelagic ecosystems (slope ≈ -1 on log-log plots; see plots of 'biomass domes' and biomass size-spectra for Lake Superior and Panther Creek/Slaunch Fork, WV below). This suggests the remarkable possibility that biomass accumulation within standardized body size increments may be a nearly constant function of the total biomass at smaller size classes. Needless to say, I'm very excited to see where this line of inquiry leads. . .

Return to top of page