Landscape Structures Affect Risk of Canine Distemper in Urban Wildlife

Urbanization rapidly changes landscape structure worldwide, thereby enlarging the human-wildlife interface. The emerging urban structures should have a key influence on the spread and distribution of wildlife diseases such as canine distemper, by shaping density, distribution and movements of wildlife. However, little is known about the role of urban structures as proxies for disease prevalence. To guide management, especially in densely populated cities, assessing the role of landscape structures in hampering or promoting disease prevalence is thus of paramount importance. Between 2008 and 2013, two epidemic waves of canine distemper hit the urban red fox (Vulpes vulpes) population of Berlin, Germany. The directly transmitted canine distemper virus (CDV) causes a virulent disease infecting a range of mammals with high host mortality, particularly in juveniles. We extracted information about CDV serological state (seropositive or seronegative), sex and age for 778 urban fox carcasses collected by the state laboratory Berlin Brandenburg. To assess the impact of urban landscape structure heterogeneity (e.g., richness) and shares of green and gray infrastructures at different spatial resolutions (areal of 28 ha, 78 ha, 314 ha) on seroprevalence we used Generalized Linear Mixed-Effects Models with binomial distributions. Our results indicated that predictors derived at a 28 ha resolution were most informative for describing landscape structure effects (AUC = 0.92). The probability to be seropositive decreased by 66% (0.6 to 0.2) with an increasing share of gray infrastructure (40 to 80%), suggesting that urbanization might hamper CDV spread in urban areas, owing to a decrease in host density (e.g., less foxes or raccoons) or an absence of wildlife movement corridors in strongly urbanized areas. However, less strongly transformed patches such as close-to-nature areas in direct proximity to water bodies were identified as high risk areas for CDV transmission. Therefore, surveillance and disease control actions targeting urban wildlife or human-wildlife interactions should focus on such areas. The possible underlying mechanisms explaining the prevalence distribution may be increased isolation, the absence of alternative hosts or an abiotic environment, all impairing the ability of CDV to persist without a host