ABSTRACT
The range of hosts a parasite can successfully occupy is partially determined by the niche breadth, that is, the set of environmental conditions necessary to maintain a stable population. Niche breadth is often quantified using host specificity, which encompasses the number of host species a parasite can exploit and the parasite’s distribution among its hosts. Parasites with a wider niche breadth can potentially occupy more host species and are often more evenly distributed among hosts than parasites with a narrower niche breadth. However, parasites interact with potential hosts within the context of a geographic locality and the set of environmental characteristics associated with it. The extent to which environmental filters associated with host individuals and the geographic context explain variation in occupancy of parasites, and the extent to which variation in occupancy is associated with host range and specificity, is poorly understood. Using data from small mammals and ectoparasites in Vermont, I used a multiscale, multispecies occupancy model (MSOM) to (1) estimate ectoparasite occupancy at 10 geographic sites and on individual hosts within each site, (2) quantify the variation in occupancy explained by the site and host levels of the model using Bayesian R2, and (3) evaluate associations between explained variation and host range of ectoparasites. For ectoparasites collected from at least 4 different host species, I calculated structural specificity to determine the distribution of these parasites across the hosts, and β-specificity to evaluate changes in host use across habitats. Host range was significantly associated with host-level Bayesian R2: generalist parasites had more variation in occupancy explained by host-level covariates than specialist parasites. This result may be explained by differences in structural specificity: many generalists disproportionally occurred on a single-host species, suggesting that host characteristics act as habitat filters for these parasites. There were no significant associations between site-level Bayesian R2 and host specificity. However, some generalists demonstrated high β-specificity, suggesting these parasites may “switch” hosts, depending on host availability. These results highlight that the terms specialist and generalist are context dependent and may not accurately describe the niche breadth of parasite taxa. Understanding variation in host specificity as it pertains to potential habitat filters may be important for predicting which parasites can bypass host filters and “jump” to a novel host, which has implications for the surveillance and management of vector-borne diseases.