Scientific uncertainties in atmospheric mercury models I: Model science evaluation

Abstract

Eulerian-based, first-principle atmospheric mercury models are a useful tool to assess the transport and deposition of mercury. However, there exist uncertainty issues caused by model assumptions/simplifications and incomplete understanding of mercury science. In this paper, we evaluate the model science commonly implemented in atmospheric mercury models. The causes of the uncertainties are assessed in terms of gas phase chemistry, aqueous phase chemistry, aqueous phase speciation, aqueous phase sorption, dry deposition, wet deposition, initial and boundary conditions, emission inventory preparation, and domain grid resolution. We also present a new dry deposition scheme for estimating the deposition velocities of GEM and RGM based on RADM formulation. From our evaluation, mercury chemistry introduces the greatest uncertainty to models due to the inconsistent kinetic data and lack of deterministic product identification in the atmosphere. Model treatments of deposition velocities and aqueous Hg(II) sorption can also lead to distinct simulation results in mercury dry and wet depositions. Although model results may agree well with limited field data of GEM concentrations and Hg(II) wet deposition, it should be recognized that model uncertainties may compensate with each other to yield favorable model performance. Future research needs to reduce model uncertainties are projected. © 2006 Elsevier Ltd. All rights reserved.