Detailed assessment of heavy metal and POP pollution in the EMEP countries, PART II: Poland
Joint MSC-E & IER-NRI & EGAR & CHMI Technical Report 5/2018

Assessment of cadmium atmospheric pollution levels in Poland in 2014 has been completed. The work has been carried out in close cooperation with national experts. The assessment includes analysis of national emission and monitoring data, model-based source apportionment in particular Polish voivodships and pollution in the selected cities. Modelling was carried out with spatial resolution 0.1°x0.1° using the latest verion of GLEMOS model.
Modeling results based on official and national emission data were compared with concentrations in air observed at EMEP and Polish stations. In was found that the model tended to underestimate the observed levels in cold period.  Analysis of factors responsible for seasonal changes of pollution levels led to conclusion that the identified discrepancies were linked with possible underestimation of cadmium emission from sector ‘Residential Combustion’.
In order to improve the model performance and to provide the national experts with the results which better agree with measurement values emission scenario has been developed. The suggested emission scenario assumes higher emissions in the southern and south-western part of Poland in cold period.

Original (left) and scenario (right) emissions of cadmium in 2014 in the modelling domain

Compared to national total value of the emission provided by national experts, annual scenario emission in Poland is 26% higher. This value is smaller than the uncertainty of annual national emissions estimated by national experts.  


Monthly sums of national emissions in Poland (left) and changes in average emission fluxes in voivodships (right) prescribed by emission scenario

The usage of higher emissions from residential combustion sector resulted in the improvement of agreement between modelled and observed concentrations in air in cold period. For the warm period the differences are negligible because the scenario assumes the increase of emissions only in cold period.

Examples of monthly mean modelled (original and corrected) and observed concentrations in air at national station PL0505 (left) and EMEP station CZ1 (right)

Scenario emission values were used for calculation of country-specific information on cadmium pollution in Poland. Deposition from anthropogenic sources are caused by national (Polish) emission sources and by transboundary transport of cadmium emitted to the atmosphere by foreign sources. For the country as a whole around 21% of anthropogenic deposition comes from foreign sources while national sources are responsible for other 79%. Contribution of foreign sources to anthropogenic deposition in the south-western part of the country is the lowest varying from 10 to 20%, and in some parts of these voivodships is even below 10%. In the central part of Poland the contribution of foreign sources ranges from 10% to 30%. In the regions near the state borders the contribution exceeds 50%.

Contribution of national and foreign sources to deposition from anthropogenic sources in Poland (left) and spatial distributions of contribution of foreign sources (right) to anthropogenic deposition in Poland in 2014 with resolution 0.1°x0.1°  

The main emission source categories in Poland include ‘A_PublicPower’, ‘B_Industry’, ’C_OtherStationaryComb’ (residential Combustion), ’D_Fugitive’ and ‘F_RoadTransport’. Model calculations were done for these five sectors, and emissions from the remaining categories were grouped into ‘Remaining sectors’. Contributions of these sectors as well as foreign sources to anthropogenic deposition in Poland were calculated for a country as a whole and for its particular voivodships. The main contributors to cadmium deposition are sectors Industry and Residential Combustion.

Spatial distribution of cadmium deposition from anthropogenic sources and contributions of main national emission sectors and foreign sources to deposition in Polish voivodships in 2014

Special attention was paid to evaluation of pollution in cities of Poland. Although the model has not been designed to simulate urban-scale pollution levels, the calculated annual mean concentrations agree with the observed values for most of the selected Polish cities. Spatial correlation coefficient is 0.71 and relative bias is -8%.

Annual mean modelled and observed concentrations of cadmium at urban background stations in selected Polish cities depicted as bar charts for particular stations

Seasonal variability of the observed concentrations was also reproduced by the model. Correlation coefficient between averaged over stations values is 0.95. For most of stations the correlation between modelled and observed monthly mean values is higher than 0.6.

Averaged over stations monthly mean modelled and observed concentrations. Range of concentration values within limits 10th – 90th percentiles is denoted by blue area for observed and by whiskers for modelled values.

Concentrations of cadmium in air in Polish cities are presented as a sum of two components. The first one comes from anthropogenic sources outside city area (external) and the second is produced by city sources (urban increment). In large cities such as Warsaw, Krakow, Gdansk, Katowice, Poznan city sources contribute significant (20 – 70%) faction to anthropogenic concentrations. In other considered cities the fraction of city sources is relatively small making up from 1-2% to about 10-15%.  

Calculated concentrations of cadmium in air caused by anthropogenic sources outside city (external) and by city sources (urban increment).  

Calculated seasonal changes of urban increment were compared with those based on observations in cities located in the north-western part of Poland. In a number of Polish cities seasonal changes were captured by the model. Both modelled and observed values of the urban increment exhibited higher levels in cold season and lower values in warm season.  At the same time, the model failed to reproduce the measurement-based increment at some other cities, which was connected with underestimation of Cd emissions from the ‘Residential combustion’ sector.

Seasonal changes of calculated normalized cadmium urban increments reproduced (left) and missed (right) by the model in comparison with the increments derived from measurements


The study demonstrated that the model can be used as a tool for evaluation of national emissions. Using modelling and observed information it is possible to identify regions where emission may need further improvement. However, this activity requires close cooperation with national experts and relevant EMEP centres and Task Forces.

Available official or national emission data contain uncertainties which affect results of assessment of pollution levels. The usage of emission scenarios or expert (non-official) emission estimates can produce more realistic, from viewpoint of comparison with the observed values, alternative results of pollution levels assessment. These alternative results could be presented along with the results based on official emission data.