We introduce an ontology of the domain of risk and catastrophe. This ontology is a conceptual model for describing knowledge from a set of concepts linked by semantic and taxonomic relations. We propose a conceptual framework which can be used to analyze various types of events, concentrated or dispersed, whether natural, industrial / technological or social. Our approach distinguishes a factual part from a representational one. The factual part describes the terminology used to depict what happened (e.g. 30 died on some bridge). The representational part depicts the standpoints and judgments of the various actors of the system (there has been an accident, a serious accident or catastrophe). That's why, the conceptual model is composed of four subsystems: the structure of the system (i.e. the elements), the system dynamics (i.e. the events), the actors who observe the system and their representation (Fig. 1). The connected structure and dynamic systems include the spatial dimension. Thus, it's possible to take into account the interweaving of spatial levels and the diffusion of an event. This model has been represented by UML (Unified Modelling Language) class diagrams.