Modelling POP long-range transport and contamination levels by MSCE-POP mode

MSC-E Technical Note 1/2007

A. Gusev, O.Rozovskaya, V. Shatalov

ABSTRACT

This technical report is devoted to the description of recent development of MSCE-POP model being used for the evaluation of POP pollution levels under EMEP. The directions of this development are in many respects determined by recommendations of the Task Force on Measurements and Modelling made within the Review of EMEP/MSC-E models. These recommendations include refinement of model description of meteorology-driven processes (re-suspension and re-volatilization from soils), improvement of boundary conditions for regional modeling by the use of hemispheric version of the MSCE-POP model, and application of the model to screening of wide range of POPs including also new substances.

The work on model development in accordance with the TFMM recommendations was begun in 2006. In particular, in order to improve the agreement of between model results and measurements a simple model description of degradation process of particle-bound POPs was worked out and included into the MSCE-POP model. Further, the influence of emission seasonal variations on temporal variations of POP contamination levels was investigated. This work allowed to improve the agreement between calculation results and available measurements of POPs.

This year these investigations were continued. To refine model description of POP seasonal variations the attempt of usage the data on aerosol spatial distribution obtained by CMAQ model for a particular year was done. It was demonstrated on the example of modeling of B[a]P transport over Europe for 2004 that the refinement of the information on spatial distribution of aerosol particles leads to noticeable improvement of agreement between calculated and measured B[a]P air concentrations. The reason is that the refinement of spatial and temporal distribution of aerosol leads to the refinement of model description of gas/particle partitioning. The latter process affects in turn degradation and deposition rates of the pollutant and, as a consequence, leads to essential changes in calculated contamination levels.

Using model description of dust suspension introduced recently in MSCE-HM model, the influence of re-suspension process to contamination levels of POPs was investigated. It was found that under some specific conditions re-suspension could originate re-emission flux comparable with that of emissions. Among these specific conditions can me noted significant fractions of urban lands; low but positive temperature (restricting re-volatilization flux); and high wind speed. In particular, such conditions are characteristic of the UK, where the contribution of re-suspension to re-emission flux is about 50% and amounts to about 30% of national emissions. On the basis of modeling results it was obtained, in general, the process of re-suspension had weak influence to air concentrations as compared to re-volatilization even for the pollutants with large fraction of particulate phase (such as B[a]P).

The work on evaluation of new substances from the viewpoint of their long-range transport potential (LRTP) and overall persistence in the environment (Pover) is continued. In this report the comparison of various methods of evaluation of these two parameters by the use of spatially resolved models is presented. It was found that LRTP and Pover were robust enough characteristics of POPs and different model approaches lead to similar results in evaluation of these parameters. Besides, in the framework of EMEP model intercomparison study it was obtained that POP models of different types agreed well in ranking pollutants with different physical-chemical properties with respect to LRTP and Pover. In the previous years model evaluation of LRTP and Pover was performed for a number of pollutants (PentaBDE, hexachlorobutadiene, pentachlorobenzene, polychlorinated naphthalenes, endasulfans, dicofol) considered as candidates for inclusion into CLRTAP and Stockholm Convention. In present year evaluation of these parameters for three congeners of short chlorinated paraffins is planned. The corresponding information will be included into EMEP Status Report in 2007.

One more important direction of model modification is working out hemispheric/global approaches to POP modelling. Calculation by hemispheric model version is a powerful tool for evaluation of POP intercontinental transport and generating boundary conditions for regional modelling. In the frame of this direction MSC-E participates in hemispheric model intercomparison study organized under TF HTAP. Besides, the work on the development of global version of MSCE-POP model is carried out in co-operation with MSC-W. During the recent technical meeting of the two EMEP Centres (MSC-E and MSC-W) in Moscow coordinated program of work on global modelling was discussed.

This year the list of land cover categories used in MSCE-POP model was extended and harmonized with that used in MSCE-HM model. At present the following land-use types are considered in the model: water bodies, deciduous forest, coniferous forest, grass, scrubs, arable lands bare soil, urban areas and glaciers. The compatibility of the renewed land cover data with the data on Leaf Area Index used in the model was verified with the help of regression analysis. More detailed description of land cover in MSCE-POP model allowed also refining of the model description of dry deposition velocities to different type of underlying surfaces. In particular, values of such meteorological parameters important for the description of dry deposition process as friction velocity and Monin-Obukhov length are now calculated for each land use type separately. Calculations of B[a]P long-range transport and depositions by the modified model for 2004 were compared with available measurements and the reasonable agreement between model results and measurements was obtained.

Model evaluation of trends of contamination in Europe for a long time period is demonstrated by calculations of PCDD/F transport and accumulation for 1990-2004. Emission data for these calculations were compiled on the basis of official emission data complemented by unofficial data when necessary. Taking into account high persistence of PCDD/Fs in the environment (particularly in soil) pre-calculations for the period 1970 - 1989 were done to generate initial concentrations in main environmental media (the atmosphere, soil, vegetation and seawater). The analysis of the obtained trends of contamination shows that under strong emission reduction in Europe (about 2 times for the period from 1990 to 2004) the influence of re-emission process increases. This influence strongly depends on local conditions, including emission reduction, temperature regime, and so on. For some European countries annual average of re-emission flux can amount up to 30% of annual emissions. In warm season re-emission flux can reach as much as 60% of that of emissions. Usage of hemispheric modelling together with regional one allow including into consideration EECCA countries. Under the absence of the data on national emissions for these countries at present only contamination in these countries caused by other European countries is evaluated.

The application of the model to the evaluation of media response to future emission scenarios is tested with two emission scenarios compiled on the basis of projections of PCDD/Fs emissions worked out by TNO and officially reported data. To do this, calculations of PCDD/F transport in Europe and Northern Hemisphere for the period 2005-2020 was performed. Initial data for these calculations were obtained during the evaluation of trends from 1990 to 2004 described above. The results obtained were analyzed from the point of view of trends of contamination in each European country and in Europe as a whole, changes in spatial distribution of contamination under the two above-mentioned scenarios, and the influence of re-emission that can be expected in future.

The outline of the report is as follows.

Chapter 1 is devoted to the description of emission data on B[a]P and PCDD/Fs. Available official data on emissions and unofficial data used for compilation of emission data for modelling are described. Spatial distribution of emissions for the considered pollutants is presented. For PCDD/Fs the emission data used for evaluation of trends in contamination in period 1990-2004 and in future up to 2020 based on two emission scenarios are given.

Chapter 2 presents the progress in refinement of model input data. In particular the work on refining of meteorological data and land cover information is described. More detailed description is presented for the data on specific aerosol surface and concentrations of OH radical prepared for usage in MSCE-POP model. These data are generated as three dimensional fields by the CMAQ model for the EMEP grid with 6-hour temporal resolution.

The description of model modifications already included into MSCE-POP model is presented in Chapter 3. The results of test calculations made for B[a]P for 2004 with modified model are presented including examples of spatial distribution of depositions to various types of underlying surface. In the end of the Chapter the analysis of the agreement between calculations and available measurements is presented.

Chapter 4 describes MSC-E activities in the field of hemispheric modelling. MSC-E takes part in the TF HTAP multi-model intercomparison study being carried out for different pollutants including POPs. First results for POPs obtained in accordance with the HTAP model intercomparison procedure using MSCE-POP model are presented in this Chapter.

Chapter 5 is aimed at the description of calculated trends of PCDD/Fs for the period from 1990 to 2004 and of the analysis of media response to possible emission scenarios. Here trends of contamination in the whole Northern Hemisphere and in particular European countries originated by sources of Europe, USA and Canada are presented. The analysis of the obtained results is carried out from the viewpoint of the influence of re-emission against a background of considerable emission reduction. The possibility of usage of MCSE-POP model for evaluating future emission scenarios is demonstrated by calculations of PCDD/F long-range transport for the period from 2005 to 2020 with two emission scenarios compiled on the basis of TNO estimates and available official data.

In Chapter 6 various approaches to the evaluation of long-range transport potential (LRTP) and overall persistence (Pover) for screening new substances are compared. It is shown that the two above characteristics are described similarly by application of different approaches. Particular attention is paid to methods of LRTP and Pover by means of spatially resolved models and particularly by MSCE-POP. Some results on the comparison of evaluation of new substances by models of different types are also discussed.

Chapter 7 is devoted to the description of model modifications that are now in progress. These are the evaluation of the influence of re-suspension process to the contamination by POPs and usage of data on aerosol particles obtained by CMAQ model for model description of gas/particle partitioning. The last modification showed promising results on agreement of calculated and measured air concentrations of B[a]P on the level of monthly averages (seasonal variations). These investigations will be continued in the following year to be used in operational version of MSCE-POP model.