Understanding Ecological Interactions Across Geographic Ranges

Across their geographic ranges, potential symbionts may interact any place they co-occur, or they may interact in only a subset of these sites. As a result, the actual presence of an interaction may be the result of local environmental or ecological contexts. The occurrence of an interaction may be due to the presence or absence of alternative partners (e.g., species may only interact with low-quality partners only if high-quality partners are absent), or directly due to abiotic conditions (e.g., one species’ preferences for or phenological overlap with a potential partner may change as a function of temperature). Given the tight coupling between their interactions, we expect that phylogenetically specialized symbionts to have little deviation between the potential and realized geography of their interactions, while we expect phylogenetically generalist species to realize a proportionally smaller portion of their potential geography. To test this hypothesis, we use a distribution modeling framework to predict the geographic overlap of symbiotic species, and then compare that overlap to models explicitly trained on geographic records of their interactions. We apply this framework to large-scale floral visitation, frugivory, and host-parasite networks. We find that congruence between geographic overlap and realized interactions differs across network types, but is generally higher for more phylogenetically specialized symbionts. Our results highlight the role that local ecological and environmental contexts play in mediating facultative interspecific interactions, and ultimately the potential for changing environments to alter ecological network structures.

Grant Foster
Grant Foster
Quantitative Ecologist & Educator