Table 2.

A list of variables considered for inclusion in generalized linear mixed models predicting pathogen and parasite exposure. Variable descriptions and rationales or predictions are provided; a * indicates the variable was included in the final complete model, a ⁺ indicates the variable was included in the geographic model.

Variable name	Description	Rationale for inclusion/prediction
Latitude+	Latitude at study area centroid	Latitude may capture geographic variation in pathogen infections; we predicted that seroprevalence decreases as latitude increases.
Longitude+	Longitude at study area centroid	Longitude may capture geographic variation in pathogen infections.
Age class*+	Estimate of wolf age class: pup (< 1), subadult (1–2), and adult (≥ 3)	As individuals age, they have more time to be exposed to pathogens, thus older wolves will have higher seroprevalence. Age category is less error-prone than numerical age estimates.
Year*	Biological year, birth month = first month	Pathogen exposure may be predictable by year (i.e., endemics), or unpredictable (i.e., epidemics).
Study area*	Study area abbreviation	Study area may describe variation in pathogen exposure, not accounted for by other variables.
Habitat quality*	Index for habitat quality based on land cover type and topography	A continuous estimate of the habitat quality of the study area, this covariate considers habitat characteristics that carnivores, especially wolves, positively select. This is a proxy for the presence of sympatric carnivore hosts. Prediction: seroprevalence increases with habitat quality.
Human density*	Number of people/1000-km2	Provides information about how urban the area is, and thus the potential for contact between unvaccinated dogs or synanthropic species (e.g., rodents, coyotes, raccoons, skunks, cats) and wolves. Prediction: seroprevalence increases with human density.
Wolf density*	Number of wolves/1000-km2; mean annual density results in one estimate per study area	Population density is related to direct transmission rates and environmental contamination. Prediction: seroprevalence increases with wolf density.
Pack size*	Mean annual pack size; one estimate per study area	This tells us about the daily contacts of a wolf, which differs from contact rate at the population-level. Prediction: seroprevalence increases with pack size.
Sex*	Male or Female	There is evidence that males have higher pathogen prevalence than females across many taxa and pathogens—we predict males have higher seroprevalence.
Coat color*	Gray or Black	The locus that confers black coat color in wolves is linked to beta-defensin genes, which increases the responsiveness of the innate immune system. We assume gray = missing k-locus, black = presence of k-locus. Prediction: black wolves have higher seroprevalence.
Age	Estimate of wolf age; integer to two decimal places	As individuals age, they have more time to be exposed to pathogens, thus we predicted older wolves have higher seroprevalence.
Social status	Breeder or non-breeder	Breeders typically have higher stress levels and energetic demands than non-breeders, which we predict increases seroprevalence.
Prey species	Top two primary prey species	N. caninum or T. gondii may be more prevalent in different intermediate hosts. Prediction: seroprevalence is higher where white-tailed deer are a primary prey species.
Pack membership	Name of the pack the wolf was a member of when sampled	There may be heterogeneities in pathogen exposure based on pack membership.
Pack density	Number of packs/1000-km2; mean annual density results in one estimate per study area	Contact among wolves from different packs is likely influenced by the number of packs in the population. Prediction: seroprevalence increases with pack density.