Addressing uncertainty in environmental modelling: A case study of integrated assessment of strategies to combat long-range transboundary air pollution

Model development and testing tend to concentrate on how well models represent “reality” or reproduce measurements. However, there are many sources of uncertainty in modelling atmospheric pollution, and those responsible for decisions on abatement strategies need to use modelled scenarios without fear that inaccuracies and assumptions in the modelling may mislead them. This paper explores how techniques from risk assessment may be used to examine a modelling study systematically. Those assumptions and uncertainties which could have significant consequences, whether arising from data used, the modelling itself, or factors omitted and incompleteness, may be identified using hazard and operability studies. This helps to target supporting studies—possibly using more complex models, or Monte Carlo uncertainty analysis; and to indicate potential implications to the decision makers. As a case study we have used work undertaken on uncertainties with the Abatement Strategies Assessment Model for the task force on integrated assessment modelling under the convention on long-range transboundary air pollution of the UN Economic Commission for Europe.