A Formal Framework for Disaster Risk Properties

Disaster risk properties (or disaster variables) such as intensity, exposure, severity, vulnerability, resilience, and capacity are significant because they provide essential information for understanding and managing disaster risk and cascading effects. While there are an increasing number of datasets that record these properties based on different criteria, such as regional levels (e.g., community resilience at counties vs. census tracts), thematic levels (e.g., social vulnerability based on race vs. socioeconomic status), or even for different hazard types (e.g., disaster risk for earthquakes vs. hurricanes), we lack a formal model that captures the semantics of these properties, i.e., their interactions with one another, and their context. Context is described through relations that constrain each property specific to an entity or as a property of an entity with respect to another entity. For example, intensity is exclusively the property of a disaster event, whereas vulnerability is a property of an element-at-risk concerning a specific hazard type. Here, we propose the Disaster Properties Ontology (DPO) that formalizes seven core properties in the disaster domain. It is built by re-using existing standard ontologies such as OWL-Time, GeoSPARQL, SOSA, and PROV-O. Additionally, DPO is developed as a sub-module of a more comprehensive Disaster Management reference Domain Ontology (DMDO).