West-Nile virus (WNV) is an increasing public health risk in Canada, especially in the context of climate change and increased urbanization. Many uncertainties remain regarding its periodic re-emergence, but some commonly accepted etiologies are associated with infected migratory birds, survival of infected mosquito vectors during winter and a possible change in feeding preference of the vector during summer. The main objective of this work was to explore the relative importance of these hypotheses on local WNV transmission. We developed a spatialized and heterogeneous compartmental SEIR (Susceptible, Exposed, Infected, Recovered) WNV transmission model, using a Cellular Automata (CA) approach and calibrated it with the eBird project data and mosquito data from Public Health Ontario. In MATLAB software, we simulated three sets of scenarios in which we modified the number of infected American robins arriving in spring, the number of infected mosquitoes surviving the winter and the progressive shift in mosquito feeding preferences from birds to mammals during the summer season. We then characterized the resulting epidemics in bird, mosquito, and human populations in time and space and compared the relative importance of the re-emergence scenarios. Preliminary results showed that increasing the number of infected migratory American robins cause an increased risk of WNV transmission in mosquitoes and humans, exacerbated by the shift in feeding behavior from birds to mammals. Simulations related to the survival of infected overwintering mosquitoes are in progress. This research provides decision-makers with key information to better understand and prevent risk associated with WNV transmission in eastern Canada.