A refinement of the emission data for Kola Peninsula based on inverse dispersion modelling
1Finnish Meteorological Institute, P.O. Box 503, 00101 Helsinki, Finland
2TNO, Princetonlaan 6, 3584 CB Utrecht, The Netherlands
3University of Tartu, Ülikooli 18, 50090 Tartu, Estonia
4University of Helsinki, Department of Physics, P.O. Box 68, 00014 Helsinki, Finland
Abstract. The study reviews the emission estimates of sulphur oxides (SOx) and primary particulate matter (PM) from the major industrial sources of Kola Peninsula. Analysis of the disagreements between the existing emission inventories for the Kola region combined with forward and inverse ensemble dispersion modelling, analysis of observation time-series and model-measurement comparison showed that the emission of the Nikel non-ferrous metallurgy plant was missing from the EMEP inventory, as well as from some others, being in some cases misplaced or mis-attributed to other sources of the region. A more consistent inventory of the anthropogenic emissions of SOx and PM has been compiled for the peninsula, compared with the existing estimates and verified by means of dispersion modelling. In particular, the SILAM model simulations for 2003 and 2006 with the revised emission data showed much lower bias – up to 6 times for the most-affected sites – for SO2 with regard to the measured concentrations of 8 Finnish and Norwegian observational stations in the region. Temporal correlation improved moderately (10–20%) but homogeneously over Lapland. The study demonstrates the value of a combined usage of forward and inverse ensemble modelling for source apportionment in case of limited observational data.