Intercomparison Study of Numerical Models for Long-Range Atmospheric Transport of Mercury

MSC-E Technical Report 1/2003

A.Ryaboshapko, R.Artz, R.Bullock, J.Christensen, M.Cohen, A.Dastoor D.Davignon, R.Draxler, R.Ebinghaus, I.Ilyin, J.Munthe, G.Petersen, D.Syrakov

ABSTRACT

In accordance with the recommendation of the Executive Body for the Convention on Long-Range Transboundary Air Pollution (CLRTAP) an intercomparison study of atmospheric long-range transport models for heavy metals was initiated by (Cooperative Programme for Monitoring and Evaluation of the LRTAP) EMEP in 1994. The intercomparison study is considered to be one of essential prerequisites for the development and application of operational heavy metal models. This report presents data on the model comparison carried out as the 2nd stage of the multi-stage program of the intercomparison of numerical models for the long-range atmospheric transport of mercury. The stage is focussed on the comparison of modelling results with observations obtained during short-term measuring campaigns. Seven regional and global models developed in Canada, Bulgaria, Denmark, Germany, Russia and the USA participated in this stage of the study. The measurements were carried out during two field campaigns each of two-week duration. Three main atmospheric mercury forms were measured - Total Gaseous Mercury (TGM), Reactive Gaseous Mercury (RGM) and Total Particulate Mercury (TPM). The measurements were done at the five monitoring stations in Germany, Sweden and Ireland in 1995 and 1999.

The comparison demonstrates that the models currently used for the assessment of regional and global pollution by mercury can provide satisfactory agreement between the modelling and observational data on mean TGM concentrations. The models can catch short-term concentration peaks but often underestimate the peak amplitudes. The comparison of RGM data reveals large differences between the modelling and observational results: for individual samples the difference can exceed an order of magnitude. On the average, the models can generally predict RGM values within a factor of 4. For TPM all models have achieved rather good agreement with observations. The models can well reproduce elevated TPM concentrations. Generally, the agreement between models and measurements for TPM results is on the level of the factor of 2. On the base of the results of this intercomparison, some recommendations for future work have been formulated.