Abstract

Vaccination of free-roaming dogs (Canis lupus familiaris) and cats (Felis catus) remains a major public health challenge. Rapid urbanization forces these species into complex contact zones, where structural failure of pulsed vaccination under high demographic turnover undermines standard One Health interventions. In such cases species-specific intervention cycles are needed to reduce zoonotic disease risk. We integrated field-data within a simulated vaccination campaign to determine how species-specific turnover rates drive the erosion of herd immunity at an Amazonian urban sentinel site (university campus). We monitored free-roaming populations (72 dogs, 75 cats) using a non-invasive photographic mark-resight protocol from 2023 to 2025. We modelled time-to-disappearance using Cox Proportional Hazards and simulated the trajectory of effective vaccination coverage against a 40% herd immunity threshold, distinguishing between loss of vaccinated individuals and recruitment of susceptible individuals. The campus functioned as a high-turnover system, with 72% of dogs and 48% of cats classified as transients. Species significantly predicted persistence (Hazard Ratio = 0.56; 95% CI: 0.33-0.94; p=0.029), with cats exhibiting double the median residency of dogs (432 vs. 193 days). Consequently, the species experienced divergent epidemiological failure modes. For dogs, simulated vaccination coverage collapsed below a 40% herd immunity threshold in 160 days, driven by rapid immunity attrition (the loss of vaccinated individuals). Although cats persisted longer, their effective coverage was eroded by immunity dilution due to recruitment of naive juveniles, creating a 33% gap between cohort survival and population-level immunity by day 365. Annual vaccination campaigns are likely insufficient in this high-turnover urban dog population. Effective One Health zoonotic control strategies must transition from static abundance-based targets to dynamic, species-specific and turnover-adjusted intervention schedules.