1. From experience of emission inventory preparation in Belarus JOINT ACCENT/GEIA Workshop on Anthropogenic emissions for non-OECD countries in global inventories 8-10 February 2006, IIASA, Laxenburg, Austria Sergey Kakareka Institute for Problems of Natural Resources Use & Ecology Minsk, Belarus

2. Issues considered 1. State emission inventory system in Belarus: main features 2. Emission inventory for EMEP in Belarus 3. Regional inventory: experts estimates 4. Connections between inventories: national and RAINS 5. Sources of uncertainty and prioritizing

3. Institutional arrangements of state emission inventory in Belarus Emission inventory system in Belarus is based on annual statistical reporting of enterprises. Primary reporting forms are summarized by regional offices of the Ministry on Statistics and Analysis. Main office of the Ministry on Statistics and Analysis generalize reports of the regional offices and produce annual report on air protection in the split of regions, branches of economy, cities and ministries. The data on emissions in annual reports includes data on emission of main pollutants (SO 2, CO, NO x, hydrocarbons and VOC) and specific pollutants (more than 70).

4. Data on emission of the main pollutants is given divided into emission from fuel combustion and emission from technological and other processes. Additional information in the annual report: number of reported enterprises, number of sources of emission, level of abatement etc. Annual emission reporting system summarizes data from more than 2000 enterprises; it is assumed that they represent about 95% of total emission.

5. Specific features of state emission inventory system in Belarus - Emissions in national statistics are summarized according to branches classification scheme (so-called OKONH) which did not coincide with SNAP and NFR classification schemes. Additional information is necessary for distribution of emission. - Mobile sources emission are not reported. They are estimated by the consumption of fuel on the national and region levels. - Domestic sources (for instance, heating) are not taken into account.

6. - Agriculture sources (collective farms, agricultural activities on the whole) are not taken into account in a regular way. - Waste management and disposal are not accounted regularly (except CH4 in GHG inventory and waste incineration). - Solvents and paints application generally are not considered except industrial activities. - There are no information on emission of some pollutants (all POPs, HM – mercury, Zn, Se, PM10 and PM2.5)

7. EMEP emission inventory EMEP inventory report is prepared using the following methodology: 1. Emission data on pollutants which sources are rather completely covered by statistics This data was distributed by SNAP and NFR classification schemes and reported as-is (main pollutants – SOx, NOx, CO). VOC emission data was also prepared by this approach. 2. Emission data on pollutants some sources of which are not covered by national statistics (NH3, TSP, heavy metals)

8. For these sources emissions were calculated by the simplest approach (using emission factors) and these values were incorporated in common reporting table together with statistical information. 3. Emission data for pollutants for which there are no any information in statistics (all POPs) For these pollutants emissions were calculated by simplest approach using emission factors and included into the report.

9. Initial statistical data for emission calculation Data of the Ministry on Statistics and Analysis, data of the Ministry on Natural Resources and Environmental Protection, Ministry on Emergency and some others are used for emission assessment. Emission factors are taken from the Atmospheric Emission Inventory Guidebook (2002) and by results of own emission sources testing as a contribution to EMEP (Belarusian contribution to EMEP 1996-2004) – for heavy metals and POPs.

10. Parallels between national and SNAP (NFR) sources classifications Emission report on the Ministry on Statistics and Analysis gives main pollutants distributed by fuel combustion emission and technological process emissions. This makes possible to reclassify emissions into SNAP and NFR formats using the following aggregation; - category “combustion in energy production and transformation” was considered as equivalent of the SNAP sector 1; - category “sold to population” – as category SNAP 0202; - category “combustion in residential and communal sector” – as category SNAP 0201; - category “used for transport” – as analogous of the sector 07+08; - difference between “used directly as fuel” and sum of “combustion in residential-communal sector” and “used for transport” – as SNAP 0301.

11. Emission report of Belarus: 2002 sample

12. An assessment of completeness of EMEP inventory According to guidelines as missing the sources reported as NE are considered. 1 A 4 b Residential - SO2, Nox, CO, NMVOC PM, POPs and HM emissions were calculated. 3 A PAINT APPLICATION - Important for NMVOC 3 B DEGREASING AND DRY CLEANING - Important for NMVOC 4 B MANURE MANAGEMENT (c) - All Except NH3 4 D AGRICULTURAL SOILS - most 4 F FIELD BURNING OF AGRICULTURAL WASTES – most 5 B FOREST AND GRASSLAND CONVERSION – All Important for GHG only

13. 6 A SOLID WASTE DISPOSAL ON LAND – All 6 B WASTE-WATER HANDLING - Maybe can be shown as IE: this sector emission can’t be extracted from tota Greatest missing sources in emission inventory are in sectors 1A4b (Residential – some subsectors), 3A (Paint application), 3B (Degreasing and dry cleaning), 4D (Agricultural soils). Main contribution of missed emission sources for main pollutants are expected into NMVOC. Some missing are also into PM, SO2, NOx and CO.

14. Expert emission estimates: regional emission inventory Here is an example of expert estimates of emission of certain pollutants for improvement of data completeness (INTAS project ). Pollutants: key heavy metals (mercury, lead, cadmium). Base years: 1990, 1995, 1997. Region: 12 Former Soviet Union countries

15. Emission estimation domain

16. Methodology and procedure Methodology of emission estimation was based on emission factors and production statistics application. Heavy metals emission statistics for NIS especially for Central Asia is very scarce. Only lead data can be found. Emission from vehicles and sometimes from non-ferrous industry are estimated. Obtaining of full information on production and usage statistics is troubled by statistics imperfection in the NIS countries. For instance, there is a scarce information on leaded gasoline usage.

17. Spatial types of emitters Area sources Stationary fuel combustion, mobile sources, some ferrous industry processes (gray iron foundries, electric steel plants), mineral products industry, most of chemical products industries, some others have been considered as area sources. Administrative regions (provinces or the whole small countries, capitals and some large cities were considered as elemental units of area sources. Area sources emissions were distributed according to density of population (rural and urban).

18. Large point sources Ferrous and non-ferrous works, and some others (SNAP 010406, 030203, 030301, 030304, 030305, 030306, 030307, 030308, 040205, 040206, 040301, 040413) are considered as point sources of HM emission. Additional information for this category was gathered in order to obtain more precise estimates of emissions and their spatial distribution

19. Some results Lead emission Annual anthropogenic lead emissions were decreased in the countries of the FSU from 24903.0 tonnes in 1990 up to 9652.5 tonnes in 1997). Share of European NIS in total lead emission of the NIS decreased from 19% in 1990 to 12% in 1997. Share of Russia was rather stable within 53-55%, share of Central Asia NIS increased from 24% to 32%. Greater contributions among the source categories were fulfilled by road traffic due to use of leaded gasoline (48% in 1990 and 26% in 1997) and production of non-ferrous metals (30% in 1990 and 50% in 1997).

20. Lead emission in the NIS by sectors 1990 1997

21. Lead emission in 1995 by 1x1 degree grid, tonnes

22. Trends in spatial structure of lead emission in European NIS

23. Differences between current and previous estimates of anthropogenic lead emissions are mainly within 30-70% for European NIS although presented data for the Ukraine are 2.6 times lower for 1995 and 1997 than ones mined from other sources. The results of presented estimates for Central Asia NIS (Kazakhstan and Uzbekistan) are significantly higher than data prepared by national experts (World Bank, 1998). This differences are appeared due to the fact, that non-ferrous metal industry rather developed in these countries was not almost taken into consideration in that document because road transport was ultimately considered as the most important lead emission source with correspondent inputs of 62-72% in Kazakhstan and 86-95% in Uzbekistan. Comparison of lead emission estimates for 1997

25. RAINS analysis and application RAINS analysis was performed to test its applicability for PM emission inventory for Belarus. Were checked: - fuel totals and distribution of fuels by sectors; - distribution of fuels by type of installations; - activity projection; - control options and types of abatement used; - PM emission factors.

26. 1. Fuel balance Share of solid and liquid fuels seems overestimated 2. Shares of fuels consumption by sectors Generally share of small combustion is underestimated and share of power generation is overestimated. 3. Shares of fuel consumption by type of installation. Some approaches and routines were elaborated for estimation of the share of stoves, fireplaces, small and medium boilers etc. in residential sector. Generally share of grate firing and stoves is underestimated.

27. 3.Level and types of abatement installation used for fuel combustion Generally RAINS is based on analysis of West Europe situation with abatement and control options and levels of abatements used for Belarus are too high. 4. Emission factors The analysis of emission factors is in progress because no PM speciated emission factors available for Belarus.

28. Comparison of TSP estimates for Belarus, ktonnes Source category 199019952000 RAINS National RAINS National RAINS National Stationary226.9132.1122.450.9111.138.0 Mobile17.252.613.041.114.226.5 Total244.1184.7136.692.0125.364.5

29. SNAP source category RAINS 2000 Statistics (2002 ) Combustion in energy and transformation industries20 0.16 Non industrial combustion plants23.63 15.0+(13.5) Combustion in manufacturing industry10.66 10.63 Production processes20.05 14.92 Extraction and distribution of fossil fuels3.66 3.033.03 Road transport8.55 25.50 Other mobile sources and machinery5.63 Waste treatment and disposal9.55 Agriculture23.62 (8.0) Total125.34 69.3+(21.5) Comparison of TSP estimates for Belarus, ktonnes

30. For future discussion: Sources of uncertainty in international emission inventories International inventories generally contains not emissions but statistical data (projections) and EF (technologies parametrization). So uncertainties result from quality of last datasets. Statistical datasets - quality of international statistics is unequal for different years; - generally out of date in comparison with national. Emission factors and their parametrization - type of abatement and its efficiency, - distribution of installations by type etc.

31. Prioritizing in emission inventories - high accuracy of estimates (to estimate only reliable sources with good emission ) or completeness (to estimate all sources)? - good national totals (main attention to priority sectors) or good sector estimates? - accurate estimates for certain years or reliable estimates (same quality) for a raw of years (to detect trends)? - hard common format or listing of distinct sources? - common format for all pollutants or distinct source; - requirements to spatial distribution? In general how to measure quality of inventories (intercomparison?)