The influence of biotic and abiotic factors on species interactions and overall community structure has long interested ecologists. Despite a legacy of interest, there is still ambiguity into the role of biotic and abiotic factors due to highly nuanced, complex networks of interactions that are difficult to comprehend. Yet, understanding how such nuances is an essential goal to determine how they affect population and community structure. Thus, the goal of my dissertation was to understand how multiple biotic and abiotic mechanisms alter interactions among larval stages of two pond-breeding salamanders. Larval stages of pond-breeding salamanders represent an excellent system for understanding how species interactions vary along abiotic and biotic gradients. Intra-and interspecific interactions are frequently determined by size differences among individuals, where larger larvae are predators of smaller larvae and can out-compete them for shared resources. However, when size differences are minimized, only competition occurs. Such conjoined competition and predation is termed intraguild predation, and is a common interaction in many taxa. The factors that determine size differences among individuals (both within and between species) are critical towards to determining both the type of interaction, as well as the strength of such interactions. The focal species I used were the ringed salamander (Ambystoma annulatum) and spotted salamander (A. maculatum). The former breeds earlier than the latter, creating a larval size advantage which results in predation as the dominant interaction between species. However, factors that influence growth rates of ringed salamanders could result in minimized size differences, resulting in a change to the strength or type of interaction that occurs. For my dissertation, I experimentally investigated three different processes that were expected to affect the relative importance of predation and competition: density dependence, food web structure, and phenological shifts. In my first chapter, I tested whether the density of ringed salamanders influenced their growth rates to such a degree that the interaction type with spotted salamanders would switch from predation to competition. I found that increased intraspecific competition in ringed salamanders reduced their body size and increased their larval period length. However, intraspecific competition did not reduce their size to such a degree that predation on spotted salamanders was precluded. Spotted salamanders showed decreased survival and increased size at higher predator densities, indicative of thinning effects. The period of overlap in ponds also increased at higher predator densities, resulting in a larger temporal window for interactions to occur. In my second chapter, I tested how six different top predator food webs would influence intraguild predation between ringed and spotted salamanders. I also tested whether food web configuration would be simultaneously impacted by increased habitat complexity. I found that ringed salamander body size and survival were unaffected by habitat complexity, and that only certain combinations of predators affected these demographic rates. Spotted salamander body size and survival showed positive and negative relationships with ringed salamander survival, but the strength of these relationships varied depending on the predator and habitat complexity treatment. Thus, pairwise interactions may not exemplify typical interactions when embedded in more complex food webs with other predators. For my third chapter, I investigated whether phenological shifts in both the ringed and spotted salamanders, simultaneous to density dependence in the ringed salamander would influence the type and strength of their interactions. I found ringed salamander survival varied with phenological shifts but only when at high intraspecific densities. Spotted salamanders were relatively unaffected by phenological shifts, and that their interactions were, similar to the previous chapters, influenced primarily by survival of ringed salamanders. As phenological shifts are predicted for many species with climate change, this study highlights that not all species interactions will be subsequently affected, and that other underlying factors (e.g. density dependence) may be more important. Thus, the most important findings of my dissertation include 1) predator density can be a dominant factor in species interactions, 2) pairwise interactions may change when embedded in different habitats or food webs in non-intuitive ways, and 3) simultaneously testing multiple mechanisms can elicit a greater understanding of the relative importance of different ecological processes.
Inferring species interactions from ecological survey data: A mechanistic approach to predict quantitative food webs of seed feeding by carabid beetles
