Sergey Kakareka Anna Malchykhina

1
Sergey KakarekaAnna Malchykhina
PM emission data for the NIS current state

• Institute for Problems of Natural Resources Use
  Ecology
• Minsk, Belarus

7th JOINT UNECE TFEIP EIONET Workshop on
Emission Inventories and Projections 31 October -
2 November 2006, Thessaloniki, Greece
2

Presentation contains draft results of analysis
of PM emission data available now for some NIS
countries made as a task of national contribution
in-kind of the Republic of Belarus into EMEP
for 2006. Purpose - assessment of the quality
(completeness, consistency, transparency etc.) of
PM emission data available for the NIS,
prioritization of sectors and determination of
ways for its improvement.
3

• Data sets on emission analized
• 1. Emission statistics (from National Reports on
  the State of Environment, Statistical Yearbooks
  and other official edition)
• 2. Official PM data reported to EMEP from EMEP
  database (Webdap)
• 3. Expert estimates of EMEP (Webdap)
• 4. RAINS model estimates
• 5. CEPMEIP dataset
• 6. MSC-W by country reports.

4
Countries analyzed Belarus, Moldova, the
Ukraine, Russian Federation (the European
part). Years 2000, 2001, 2002, 2003,
2004 Methodology of analysis ?) Comparison of
emission estimates by totals by key sectors
(SNAP 1) by years (trends)

•  
•  

5
B) Comparison with activity data Data sources on
activity national statistics. Compilation of
analized data for PM national totals by countries
is shown on the charts below. Explanation of
notations on the charts Statistics statistical
data on emissions (generally for stationary
sources only domestic sector and some others not
accounted) Official data reported to EMEP
(Webdap) Expert emission estimates by EMEP
(Webdap) RAINS estimation by RAINS model (for
2000 and 2005).

•  
•  

6
By the data EMEP database contains scarce
official information on PM10 or PM2.5. Only for
Moldova there is a whole set of PM10 data. For
other NIS it available for 1-3 last years.
Statistics do not contain it at all. So mainly
TSP emission data can be used for in-deep
analysis. TSP official data available for most
of years 2000-2004 for Moldova, 2002-2004 for
other countries. Generally it is rather smooth
though some jumps present. Statistics available
for all years except the European part of the
Russian Federation for which mainly emission data
for total Russia available. Emission on the
European part can be estimated using known share
of the European part of Russia in total emissions
for 2003 (about 31).

7

•  
•  

PM emission data for Belarus
8

•  
•  

PM emission data for the Ukraine
9

• To compare expert and official estimates we can
  accept that roughly PM10 comprises 50-60 of TSP.
  
• Expert estimates and RAINS data are generally
  higher then official and statistical. Thus for
  Belarus expert estimates are higher than official
  (about 20). For Moldova expert emissions for
  2000-2001 are higher than for 2002-2003 7-8
  times. For 2000-2001 they are closer to official
  and close to RAINS, for 2002-2003 to statistics
  and differ from RAINS.

10

• For the European part of Russia official emission
  values in 2004 1030 thous. tones (by statistics
  830.8 thous. tones). Expert estimates for 2003
  1351.9 thous. tones PM10, close to RAINS values
  of emission.
• For testing of probable reasons of detected
  differences emission data by key sectors was
  analized and available production statistical
  data was used.

11

• Differences of classificators should be taken
  into account. Thus, in the NIS statistics Power
  Industry emission includes not only emission from
  fuel combustion, but also technological
  emissions. Technological and combustion emissions
  can hardly be separated for other key sectors
  (Ferrous, Non-Ferrous, Building Materials,
  Communal etc.).
• Analysis of SNAP sectors emission data have
  revealed significant differences between
  estimates.

12

•  
•  

PM emission data for SNAP1 sector of Belarus
13

• Thus expert estimates of SNAP1 emissions for
  Belarus are higher than statistical about 100
  times while taking into account that Power
  Industry in Belarus practically do not utilize
  solid fuels.
• For the Ukraine SNAP1 emissions by expert
  estimates compared to statistics looks rather low
  taking into account that most of coal in the
  Ukraine is combusted in Power Industry.
• For the whole territory of the Russian Federation
  about 35 of total dust is emitted by power
  plants (965.6 thous. tones in 2004). It should be
  expected that in the European part of Russia
  150-250 thous. tones dust maybe emitted by Energy
  sector according to expert estimates for 2003
  50.6 thous. tones PM10 was emitted in this sector
  in EMEP part of Russia.

14

•  
•  

PM emission data for SNAP1 sector of the Ukraine
15

• Similar analysis was made for other key sectors.
• Conclusions
• - progress was made last years in PM emission
  inventory improvement for NIS region a few
  datasets on PM emission for the NIS are available
  now (statistics, official, expert), some of them
  include data on PM10 and PM2.5
• - all datasets are obtained by different
  methodologies
• - determination of the quality of dataset needs a
  clarification of requirements to emission data

16

• application of traditional EMEP criteria to all
  PM emission datasets (completeness, accuracy,
  consistency, transparency et al.) is useful no
  one now can be considered as fully compliant with
  all these requirements
• even not fully compliant emission estimates are
  useful if based on a known methodology
• - emission estimates should be validated by
  independent methods
• - by the date for the NIS TSP emission data is of
  the best quality, thus it is important to
  collect TSP emission data in EMEP database in
  view of verification purposes of PM10 and PM2.5
  data.

17

• prioritization between totals and sectors
  estimates is necessary estimates can coincide
  in total emission values but differ significantly
  in sector estimates sometime only because put the
  same activity into different classes
• estimates should be made for natural classes -
  for which a statistics can be collected
  application of detailed artificial classification
  schemes do not led to the growth of the quality
  of estimates
• should be expected that uncertainty of estimates
  is growing from top to down this is a specific
  feature of top-down approach while for bottom-up
  should be expected backward tendency

18

• key problem in inventory improvement in the NIS
  lack of necessary statistical information
  especially for mobile sources
• - emission factors are results of a certain
  emission inventory (emissions divided
  consumption), after that they are used in another
  inventory. So default emission factors which are
  planned for deriving national totals can be
  attributed from national emission inventory in
  other country.

19

• Proposals for PM emission inventory improvement
• - guidelines or courses on the sources of
  statistical data and its transformation before
  application in emission inventory will be useful
  
• emission factors can be derived from best
  quality emission inventory data and included in
  the Guidebook as default for certain cases
• sectors of priority improvement by regions
  taking into consideration input in total
  emissions and level of ucertainty
• regular intercomparison of emission estimates
• training courses in inventory tools like COPERT,
  RAINS, CEPMEIP etc.

20

• Thank you for your attention!

21

•  
•  

PM emission data for Moldova
22

•  
•  

PM emission data for SNAP1 sector of Moldova
23

•  
•  

PM emission data for the Russian Federation
(European part)
24

•  
•  

PM emission data for the Russian Federation
(European part)