The desert grasslands of southeastern Arizona are home to 30 species of sparrows. These mid-elevation grasslands, tucked in between higher elevation oak forests and lower elevation desert scrub, are characterized a diverse flora dominated by late winter and early spring annuals, that grow in response to winter rains, and  perennial grasses, that grow in late summer following summer rains. Summer precipitation is highly variable but grass production is usually sufficient to provide plentiful seeds and dense ground cover favored by many species of sparrows. The ground cover is also dense enough to carry wildfires, resulting in a patchwork open grasslands interspersed with scattered stands of mesquite and oaks.

I address the question "How is the local density and diversity of sparrows related to local vegetation and seed production?" Eight species of sparrows breed in these desert grasslands, and they are joined by a large number of migratory sparrows that come in fall just as seeds from the summer rains are ripening.  Grass seed production varies over 3 orders of magnitude, from as low as 0.1 kg ha^-1 in years of low summer rainfall to over 100 kg ha^-1 in wet years. In years of low seed production, sparrows consume ~95% of the seeds produced, despite relative low sparrow densities. In years of high seed production, sparrow numbers increase 10- to 100-fold but since seed production may increase 1000-fold, sparrows consume a smaller fraction of the total seed production.
Results/Conclusions

I will describe the biology of the desert grasslands and grassland sparrows in more detail and present the results of a quantitative model that attempts to explain the coexistence of this diverse assemblage of sparrow species. Sparrow species differ in their seed-size and habitat preferences, and the model focuses on how characteristics of place, e.g. rainfall patterns, topography, tree density, and seed size, determine local sparrow species composition. The model also addresses the question of limiting similarity of coexisting species and concludes that there are stronger constraints on the on the similarity of resident species than on migratory species. A more general conclusion is that ecological theory needs to be tailored to specific systems and that incorporation of place-based knowledge greatly enhances the utility of more general ecological models.