Uncertainty in the estimation of critical loads: A practical methodology

Critical loads of acid deposition for forest soils, ground and surface water resources are calculated utilising a variety of mathematical models. The estimation of the predictive uncertainty inherent in these models is important since the model predictions constitute the cornerstone of the development of emissions abatement policy decisions in Europe and the United Kingdom. The Generalised Likelihood Uncertainty Estimation (GLUE) approach is presented here as a tool for estimating the predictive uncertainty of PROFILE, a steady-state geochemical model that is widely applied within the critical loads community. GLUE is based on Monte Carlo simulation and explicitly recognises the possible equifinality of parameter sets. With this methodology it is possible to make an assessment of the likelihood of a parameter set being an acceptable simulator of a system when model predictions are compared to observed field data. The methodology is applied to a small catchment at Plynlimon, Mid-Wales. The results highlight that there is a large amount of predictive uncertainty associated with the model at the site: three of the six chosen field characteristics lie within the predicted distribution. The study also demonstrates that a wide range of parameter sets exist that give acceptable simulations of the site characteristics as well as a broad distribution of critical load values that are consistent with the site data. Additionally, a sensitivity analysis of model parameters is presented.