Representative scenarios with predefined return periods are of importance in many contexts, including risk communication and quantitative risk assessments for which simple Monte-Carlo based analysis is not possible. Determining a hazard scenario associated to a specific return period for designing or assessing an infrastructure system is nontrivial, because the intensity of a natural hazard impact on an infrastructure system is spatially variable. In this paper, we propose a methodology for determining representative spatially-distributed hazard scenarios. It is based on a sampling-based system analysis utilizing a simplified system model based on network efficiency. The methodology is demonstrated by application to an idealized lifeline system.