UC Santa Barbara

Few hosts have many parasites while many hosts have few parasites -- this axiom of parasite ecology is known as parasite aggregation and is so pervasive that it is one of the few general laws in disease ecology. The propensity of parasites to be aggregated has important implications for both making inference about the mechanisms structuring a host-parasite interaction as well as predicting population-level host-parasite dynamics. In this dissertation I ask two questions: 1) How do the dynamics of parasite load and aggregation affect disease transmission, epidemics, and endemics in wildlife disease? 2) When can we make inference about the mechanisms structuring a host-parasite interaction from observed patterns of parasite load? I develop constraint-based theory for host-parasite systems to identify when patterns of parasite aggregation provide information about the mechanisms driving a host-parasite interaction. This approach shows that common patterns of parasite aggregation are highly constrained (i.e. predictable) by a simple set of constraints, providing a system-independent explanation for the ubiquitous pattern of parasite aggregation across host-parasite systems. However, despite the highly constrained nature of parasite aggregation, I show that particular mechanisms, such as parasite-induced host mortality, can lead to deviations from constraint-based theory and I develop statistical procedures to detect these deviations in cross-sectional parasite load data. While constraint-based theory focuses on static patterns of parasite aggregation, parasite aggregation also influences host-parasite dynamics. I show that when parasite aggregation is consistent with the predictions from constraint-based theory, the ability of parasites to regulate the host population and stabilize the host-parasite equilibrium is significantly reduced, compared to the canonical assumption of fixed parasite aggregation. Finally, I develop a mathematical framework using Integral Projection Models (IPMs) to model parasite load dynamics when parasite load is a continuous variable. In combination with laboratory and mesocosm experiments, I apply this approach to an amphibian species infected with a fungal pathogen and show that disease-induced host extinction is far more sensitive to the load dynamics of the parasite than to the transmission dynamics in the system. This work highlights the importance of considering parasite load dynamics when developing strategies to mitigate disease-induced host declines. Broadly, this dissertation illustrates how both bottom-up, mechanistic approaches and top-down, statistical approaches can be used to provide unique insights into the mechanisms structuring consumer-resource interactions.