Integrated natural hazard risk assessment aims at capturing the impacts
and diverse consequences of natural hazards on different types of elements at risk
(i.e. evaluation criteria). Typically, a risk assessor selects such evaluation criteria
relying on expert knowledge. Moreover, legal frameworks, best-practice guidelines
as well as manifold requirements by stakeholders should be considered in the
assessment process. This is also true for risk mapping. In order to provide useful
deliverables for stakeholders, their preferences in regard to content and
visualization have to be borne in mind. Developing tools to assist risk assessment
may focus on some or all of these aspects. Necessary in this context are solutions
which are capable of capturing relevant expert knowledge, but which are also
easily adaptable to changing requirements. Ontologies, and knowledge bases built
upon these, seem to pose ideal concepts to tackle this objective. Ontologies are
formal, sharable and machine-interpretable knowledge representations. They thus
seem suitable to capture explicit and formal expert knowledge on natural hazard
risk assessment. Ontologies may also serve as knowledge stores for tacit
information. This way, stakeholder preferences as well as local knowledge, a
knowledge body often untapped, can be captured and made accessible. Ontology
building methodologies provide the required means to structure and formalize tacit
knowledge. They also allow for an iterative extension and adaption of an ontology
as required. We present an ontology which captures core concepts of risk
assessment and their relations as well as local knowledge and stakeholder
preferences. The ontology is used to enhance a software tool which has previously
been employed in flood risk assessment case studies. In doing so, we show how
semantics can assist an integrated natural hazard risk assessment, and further
evaluate the pros and cons of the proposed approach.