Table 1 Typology of uncertainty

Location: the location at which the uncertainty manifests itself in the assessment

Model structure: Structure and form of the relationships between the variables that describe the system

Parameters: Constants in functions that define the relationships between variables (such as relative risks or severity weights)

Input data: Input data sets (such as concentrations, demographic data, and incidence data)

Nature: the underlying cause of the uncertainty

Epistemic: resulting from incomplete knowledge

Ontic

Process variability: resulting from natural and social variability in the system

Normative uncertainty: resulting from a plurality of socio-ethico-normative considerations within a society

Range: expression of the uncertainty

Statistical (range + chance): specified probabilities and specified outcomes

Scenario (range + "what if"): specified outcomes, but unspecified probabilities

Recognized ignorance: unknown outcomes, unknown probabilities – uncertainties are present, but no useful estimate can be given

Methodological unreliability: Methodological quality of all different elements of the assessment; a qualitative judgment of the assessment process which can based on e.g. its theoretical foundation, empirical basis, reproducibility and acceptance within the peer community

Value diversity among analysts: Potential value-ladenness of assumptions which inevitably involve – to some degree – arbitrary judgments by the analysts.