The key task of this research was to link the macroecology of ant assemblages—properties like average size, abundance, and diversity—to abiotic gradients sensitive to climate change. Repeatedly, NPP and temperature were identified as major players underlying the biogeography of community structure. To get a more complete picture of our work, scan our pdf reprints.

Body size is an evolutionary variable that responds to a variety of ecological gradients and in turn shapes ecological interactions. Both aspects of size in a social insect colony--the size distribution of individuals, and the number of individuals--vary across species (4 and 8 orders of magnitude respectively). The challenge is to identify the origins of size gradients and use them in community models.

Our 49-site survey of ant communities has allowed us to build the best size dataset for a terrestrial arthropod taxon. Gradients of worker size supported von Bertalanffy’s hypothesis that increases in catabolism with temperature produce smaller ectotherms in warmer environments.

In contrast, the average worker number in a colony was a unimodal function of NPP. This is consistent with the twin hypotheses that hypo-NPP constrains colony size and hyper-NPP favors small colony species by relaxing a key weakness when they compete with larger species: their relatively high metabolic rates.

For more on size in social insects, check out
11.  Kaspari M and E Vargo  (1995) American Naturalist
16.  Kaspari M, M Weiser (1999) Functional Ecology
24.  Kaspari, M and M Weiser. (2000) Biotropica
34. Kaspari M. (2005) PNAS

Gradients of Ant Abundance

One great advantage to the study of soil arthropods is that they can be studied with quadrats. Elton in 1927 bemoaned the lead that botanists then had over their zoological counterparts by arguing that "Their study organisms don't jump up and run away whhen you try to count them." Ant colonies in the soil are relatively sessile, and this simple fact has allowed us to explore of how ant abundance is regulated at scales from local to global.

Gradients of Ant Diversity

Finally, a convincing explanation for diversity gradients remains the grail of community ecology. My lab is helping to build a scalable theory in which each process that regulates biodiversity has a particular scale at which it dominates.

Our 49-site study represents our first attempt to test diversity models for scale-dependence. It yielded 741 taxa with about one third unknown to science at the time. At small scales, the ecological factors that regulate abundance best allowed us to predict patterns of ant diversity. As plot size increased, however, tropical sites begin to diverge (there is little evidence for a latitudinal gradient within the 48 states). Two hypotheses that link high tropical diversity with warm temperatures (=high mutation and recombination rates) and the large area (=high balance of speciation to extinction) account for three quarters of the variation in ant diversity at the largest scales Metabolism based theories accurately predicted the slope of the temperature-diversity relationship.