Air Quality Uncertainty

One of my main Defra projects was looking at uncertainty in air quality and emissions modelling. The policy questions often sounded quite simple, for example, how likely are we to meet a particular emissions target? but the models we used were complicated, and uncertainty was not often quantified, let alone communicated consistently. I investigated methods including Monte Carlo simulation, correlated error propagation, and Gaussian Process emulation, using R to build test workflows and interactive outputs. I also spent a lot of time thinking about how to explain uncertainty clearly, rather than leaving it buried in technical documentation. The work fed into dashboards, Official Statistics and policy analysis used for target setting and international negotiations.