Methods to address uncertainty in ecological risk assessment of Genetically Modified Organisms

There are three types of uncertainty encountered in ecological risk assessment: linguistic uncertainty, variability and incertitude. This distinction is important because these types arise from very different mechanisms, and risk analysts must use different approaches to represent, propagate and communicate these uncertainties in a risk assessment. Linguistic uncertainty occurs in ecological risk assessment because the language used to describe events and processes suffers from ambiguity, context dependence, underspecificity and vagueness. It is usually dealt with by carefully defining terms used in the assessment but can only be truly minimised by moving to a quantitative risk assessment. Variability is natural variation, or the uncertainty caused by inherent fluctuations or differences in a quantity or process. It cannot be reduced but it can be better understood. Incertitude is uncertainty caused by incomplete descriptions of a mechanism or process and other limitations of scientific knowledge. It can be reduced by gathering additional data. There are a large number of mathematical techniques that can be used to represent and propagate variability and incertitude (individually or jointly) in an ecological ris