Making Sense of Air Pollution Modelling: Framed Uncertainty

Scientific results on air pollution are often conveyed to publics in absolute numbers implying accuracy and that researchers know specific matters related to residential wood stove emissions, premature mortality or economic costs with precision. The movers and shakers of society, in other words, appear to know such air pollution issues with accuracy communicating unambiguous numbers in public policy. But wood stove emissions are particularly uncertain and health experts suggest that the adverse health effects that can be quantified likely represent just the visible tip of an iceberg. Numerous impact dimensions on human health, biodiversity and climate elude quantification and remain highly uncertain, which begs the question of how researchers are communicating about such uncertainties to policymakers. To answer this question, I first explore how key actors in Denmark make sense of air pollution. I find that key actors make sense of air pollution through processes that involve multiple forms of data and different political purposes. Having established the foundation for studying uncertainty in this field, I proceed to analyse how uncertainty is being treated in three case studies related to measuring ultrafine particles, calculating residential wood stove emissions, and estimating the adverse health costs of air pollution. In each of these cases, I find that unmeasurable uncertainty is conflated with measurable uncertainty, leading to the marginalisation of unmeasurable uncertainty in science conducted for policy. The implication of my analysis is that the incumbent public policy tradition which favours quantitative values as public policy input must be reconsidered. Its limitations lie in preventing public officials from acting on unmeasurable uncertainty and exploring innovative new courses of action. Because uncertainty is particularly prominent in air pollution modelling and therefore consequential at the policy level, I propose that uncertainty is foregrounded in a manner that is actionable, concentrating on harm-reduction. By foregrounding critical uncertainties in air pollution modelling, I argue that researchers can provide policymakers with a more credible and helpful understanding of the profoundly uncertain context from which they must make difficult choices.