A majority of infectious diseases in humans are zoonotic. The incidence of infectious diseases in humans and wildlife is increasing worldwide and the main drivers for diseases outbreaks are land use and climate change. At the same time, human activities are also reducing biodiversity, and thus the relation between biodiversity loss and the rise of infectious zoonotic diseases has been extensively debated. We evaluated the strength of this link and the mechanisms driving Lyme disease prevalence in Southern Quebec, a rapid zone of emergence for this disease. Lyme disease is the most common tick-borne disease worldwide and has reached epidemic proportions. In Southern Quebec, the incidence of the disease has been rising rapidly, since the first endemic case was detected in 2008. We developed models validated with empirical field data to estimate the risk for Lyme disease in wildlife and the pattern of spread of the disease under climate change. We show that the relation between diversity and disease is highly context dependent. The magnitude and direction of this relation varies across spatial (regional vs. local) and temporal (established vs. emerging) scales.