Habitat distributions of wintering sparrows: foraging success in a transplant experiment

Ecology, March, 1996 by Richard R. Repasky, Dolph Schluter

INTRODUCTION

Temperate birds are frequently habitat specialists, and adjacent habitats are often occupied by closely related species. Lack (1944) argued that this pattern of distribution results from interspecific competition. He reasoned that morphological differences between species have only minor consequences on fitness in different habitats and that species ought to be more widely distributed among habitats in the absence of congeners. Lack rejected the alternative hypothesis that habitat distributions result from species differences in foraging success in different habitats. Although the competition hypothesis has frequently been tested (e.g., Terborgh 1971, Pulliam 1975, Terborgh and Weske 1975, Noon 1981, Schluter 1982, Garcia 1983), the foraging rate hypothesis has seldom been tested explicitly (Schluter 1982, Price 1991, Repasky and Schluter 1994).

Habitat specialization might result from food alone if species are adapted to feeding conditions in alternate habitats (e.g., Smiley 1978, Futuyma and Wasserman 1981, Schluter 1982). The trade-off in feeding ability between habitats must be large enough to restrict species' distributions. Hence, there are two testable predictions from the foraging success hypothesis. First, each species should achieve its highest food intake rate in the habitat in which it normally occurs. Second, the trade-off in foraging ability between habitats should be significant.

Similar predictions of foraging success result from a second hypothesis. Under exploitative competition, competitors shape the foraging success of one another as they deplete food supplies. Habitat partitioning results if species deplete food supplies in such a way that each habitat becomes suitable for only one species (Pimm and Rosenzweig 1981).

The foraging success and exploitative competition hypotheses can be tested by transplanting species among habitats and measuring foraging success. Both hypotheses are tested if foraging success is measured using natural levels of food supply that have been subjected to depletion by potential competitors. If species are distributed as one would predict from foraging success, the hypotheses are confounded and further experiments are necessary to distinguish between them. If not, both hypotheses can be rejected.

We tested the foraging success and exploitative competition hypothesis by comparing the feeding rates of sparrow species transplanted among habitats in the Sonoran Desert of southern California. Sage Sparrows (Amphispiza belli), Black-throated Sparrows (A. bilineata), and Dark-eyed Juncos (Junco hyemalis) spend the winter in different habitats along an elevational gradient (Weathers 1983, Repasky and Schluter 1994). Sage Sparrows dwell in a creosote bush - saltbush (Larrea tridentata - Atriplex spp.) shrubland on the floor of the valley at the base of the elevational gradient. Black-throated Sparrows occupy the creosote bush shrublands located on rocky alluvial fans at the entrances of small valleys and on the rocky, lower slopes of the mountains (other common species include brittlebush [Encelia farinosa], burrobush [Ambrosia dumosa], sweetbush [Bebbia juncea], and cactus [Opuntia spp.]). Dark-eyed Juncos inhabit a woodland of pinyon pine (Pinus monophyla) and juniper (Juniperus californicus), located on a plateau above the other two habitats. In an earlier study, we showed that the habitat distributions of these sparrows are unrelated to food availability (Repasky and Schluter 1994). Here, we experimentally address the foraging success and exploitative competition hypotheses to challenge that conclusion.

We carried out a transplant experiment because there were few natural opportunities to observe foraging success outside of species' typical habitats. The advantage of such a manipulation is that observed food intake rates are the net effect of several factors affecting food availability, such as food abundance, vegetation, and substrate structure and possibly predation risk (Repasky and Schluter 1994, Repasky, in press). By measuring foraging success in the aviary, we tested the predictions from the foraging success and exploitative competition hypotheses that species should occupy the habitats in which they forage most successfully and that large differences in foraging ability should exist between habitats.

METHODS

Experimental design

The experiment was carried out during the winter of 1988-1989 in the vicinity of Deep Canyon Desert Research Center, Palm Desert, California (see Weathers 1983). Study sites in the different habitat types were located on ecological reserve lands along a transect on the north slope of the Santa Rosa Mountains.

Birds used in the study were captured from the wild shortly before the experiment began. They were housed individually and maintained on a mix of seeds commercially available for pet finches, mealworms, water, and a vitamin supplement.

Sparrows generally search for seeds on the ground. Food intake rates were estimated by observing solitary birds foraging on seeds naturally occurring on the ground inside of an aviary. We used solitary individuals to minimize the number of birds required in the experiment. Hence, our experiment rests on the assumption that birds feeding in a habitat that is well stocked with food will achieve higher food intake rates than birds feeding in a poorly stocked habitat so long as flock size is held constant between habitats. Although food intake rate often increases with flock size (e.g., Caraco 1979), our assumption is reasonable in the absence of strong interactions between the effects of flock size on food intake rate and habitat. Indeed, the three species are likely to respond similarly to changes in flock size because they all travel in flocks, flee into woody vegetation to escape from avian predators, and feed near cover (Repasky and Schluter 1994).