The EDGAR 3.2 Fast Track 2000 dataset (32FT2000)
Download the FT2000 documentation (this page and the caveats) in pdf format
Since making detailed updates of EDGAR is a laborious effort, the EDGAR consortium decided to initiate a new so-called 'Fast Track' action to estimate recent global emissions at country level based on readily available data (1) . It is the aim to produce an annual update of the Fast Track emissions.
The Fast Track 2000 dataset comprises global anthropogenic emissions for the year 2000 of Kyoto Proto-col greenhouse gases CO2, CH4, N2O, and F-gases (HFCs, PFCs and SF6) and of the air pollutants CO, NMVOC, NOx and SO2. The emissions data are presented as country/sector tables and as 1x1 degree grid files at the same level of detail as the EDGAR 3.2 emissions data have been published on the internet.
(1) An earlier exercise to construct global total emission trends from the 1995 data have been published annually as part of the Environmental Data Compendium (RIVM/MNP, 2004) and in Part 3 of the annual IEA publication 'CO2 from fuel combustion' (Olivier, 2004). These estimates do not show any regional or country specific details (for a comparison with the FT approach see Section 4).
2. Data sources and methodology for the year 2000 dataset
In general activity data for the year 2000 has been included following the EDGAR 3.2 method (Olivier and Berdowski, 2001; Olivier et al., 2002). Selection of emission factors is based on the assumption of unchanged control technologies compared to the year 1995, resulting in application of the emission factors as included in version 3.2. However, to take into account emission reduction that has occurred due to control measures implemented since 1995, ('implied') emission factors have been used for those countries for which information on emission reduction was available (mainly OECD'90 countries; here shortly referred to as "OECD"). Implied emission factors are constructed by division of annual emissions by activity selected for the extrapolation. In general these emission factors have been taken from the CRF emission data files which are part of the National Inventory Reports ('NIR') to the UNFCCC (Olivier et al., 2005).
The gridded emissions all use the same grid maps for the within-country distributions as in EDGAR 3.2, except for large-scale biomass burning for which the GFED data were used (see Section 2.5). For the latter also monthly emissions and effective emission heights are provided as auxiliary datasets.
2.1 Fossil fuel production and use (EDGAR categoryF; IPCC sector 1)
Activity data for fossil fuel production and use for 136 countries are taken from IEA statistics for OECD (IEA/OECD, 2003a) and non-OECD (IEA/OECD, 2003b) countries (2). For another 72 countries, the aggregated IEA data for the regions 'Other Latin America', 'Other Africa', and 'Other Asia' have been split into country data using population density figures from FAO (2005a). For yet another 26 small countries, for which no data is presented in the IEA statistics (mostly very small islands), the EDGAR 3.2 1990-1995 trend has been extrapolated to the year 2000. Data on hard coal and brown coal production have been split into surface and underground mining as included in version 3.2. Discontinuities with the EDGAR 3.2 data will be found due to (i) update IEA energy statistics, in particular for former USSR countries and specific developing countries and (ii) distribution of country data included in the "other regions" of IEA using population statistics instead of data from UN statistics applied in EDGAR 3.2.
Emission factors for 2000 have in general been taken from the EDGAR 3.2 data for 1995, except in OECD countries for which control measures have been included using so-called implied emission factors (exclusion: aviation (F57) and international shipping (F58)). This refers in particular to non-CO2 combustion emissions from road transport (F51), industrial combustion (F10) and power generation (F20) (see Table 1).
Exceptions to the above mentioned activity data and emission factors are gas flaring, CH4 emissions from fossil fuel production and international shipping emissions. Gas flaring emissions (F80) have been calculated by combining the EDGAR 3.2 values for 1995 with the 1995-2000 CO2 trend from CDIAC (Marland et al., 2003). For some countries, for which CDIAC did not report CO2 flaring emissions in the year 2000 and for which it seems unrealistic that gas flaring did not occur (e.g. Nigeria, Norway and China), constant 1995 emissions have been applied. For CH4 emissions from fossil fuel production coal mining (F70), oil production (F80), and gas production (F91) and gas transport/distribution (F92) country-specific trends reported to the UNFCCC have been used.
For emissions from international shipping (F58) two methodologies are presented illustrating the uncertainty in activity data in this sector. The first methodology is the EDGAR 3.2 method with updated IEA bunker statistics, the second method is a detailed emission inventory using fuel use calculated from number of ships, engine types and operating conditions and hours (Eyring et al., 2005), which dataset has been labelled "SHIP-EYRING".
Table 1. Sectors for which implied emission factors have been calculated based on reported emissions to the UNFCCC. If marked x, the compound emission has been constructed using implied emission factor for the specific sector. If marked -, the EDGAR 3.2 emission factor for 1995 has been applied. Blanc boxes are not applicable for these compounds. See Table B.1 for OECD'90 average values for CH4 and N2O.
2) Please note that instead of using aggregated IEA sectoral trend data for extrapolation calibrated to the EDGAR 3.2 data for 1995, we used the full IEA 2004 dataset for 2000. This is like to have introduced discontinuities between the 2000 FT emissions and the 1995 emissions in EDGAR 3.2, in particular for former USSR countries and specific de-veloping countries.
2.2 Biofuel production and use (EDGAR category B; IPCC sector 1)
To maintain consistency with the 1995 emissions data, for biofuel consumption in the residential/commercial sector (B40), the same trend estimation procedure was used as for EDGAR 3.2: for indus-trialised countries the total population trend was used; for developing countries the weighted trends of rural and urban population (see Olivier et al., 2001). However, for biofuel use in industry (B10) and power generation (B20) for the year 2000 data from IEA statistics for OECD non-OECD countries were used (IEA/OECD, 2003b) countries (4).
Due to lack of data, for charcoal production (5) and biofuel use in road transport (B51) constant 1995 values have been applied. Under the assumption of unchanged control technologies in the production and use of biofuels, emission factors have been assumed to remain constant from 1995 to 2000.
(4) Please note that the IEA data for these sources have been updated substantially compared to the datasets released and applied in EDGAR 3.2, which may have introduced discontinuities between the 2000 FT emissions and the 1995 emissions in EDGAR 3.2.
2.3 Industrial processes (excl. F-gases) and solvent use (EDGAR category I; IPCC sectors 2 and 3)
Production data on iron and steel (I10; split into different technologies) have been taken from (IISI, 2004). Production data of the non-ferrous industry (I20; copper, lead, zinc, aluminium) are based on USGS (2004), while for PFCs from primary aluminium production the fractional contribution of different processes from EDGAR 3.2 has been applied. Industrial production data for the chemical industry (I30) are from the UN commodity statistics (UN, 2004). For those countries were no UN data was available, constant 1995 values are assumed. An exception was made for N2O emissions from adipic acid (AA) and nitric acid (NA) manufacture from OECD countries. These were extrapolated from 1995 using the country-specific 1995-2000 trends reported to the UNFCCC; for other OECD countries the reported OECD total trend of total AA and NA emissions was used.
For the other industrial source categories the following data sources have been used: cement (I40; USGS, 2004), paper and pulp (I50; FAO, 2005b), food (I60; FAO, 2005b) or constant values for countries with no data in FAO). For NMVOC from solvents (I70), the trend in total population was used (FAO, 2005b). Emission factors have been assumed to remain constant from 1995 to 2000 except for country-specific trends of N2O emissions from adipic acid and nitric acid manufacture in OECD countries which showed an average emission decrease of ~50% (see Table B.1).
2.4 Agriculture (EDGAR category L; IPCC sector 4)
Biomass burning emissions have been taken from the Global Fire Emissions Database (GFED; Van der Werf et al., 2003). The provided CO emissions were scaled with the ecosystem dependent emission factors provided by Andreae and Merlet (2001). For NOx, the emission factors were taken from a recent update by Andreae (pers.comm., 2004) resulting in an 40% lower emission factor for savannah fires. To this pur-pose, the ecosystem database of Olson et al. (1983) was aggregated into five classes: shrub/bush, forest, agriculture and other (e.g. urban regions/deserts). Biomass burning emissions are sub-divided into the following source categories:
GFED 1.0 data in agricultural regions were attributed to savannah and grassland fires (L42 and L47). There is an insignificant overlap with EDGAR category agricultural waste burning (L43) which is presented separately with constant 1995 emissions (6). In addition, for N2O the indirect post-burn emissions from tropical forest fires (L45) have been extrapolated using the calculated 1995-2000 trend in direct N2O emissions from that source category (L41), which bas been estimated in EDGAR 3.2 as a fixed ratio be-tween direct and post-burn emissions.
Table 2 presents the biomass burning emission factors as taken from Andreae and Merlet (2001), Andreae (2004, pers. comm.) and EDGAR 3.2. For easy reference in Table B.2 the multiplication factors are provided to convert one set into the other (for regional and per country emissions).
Table 2. Large-scale biomass burning emission factors 32 and AM2
EF_3.2 = EDGAR 3.2 (Table 4.6.b in Olivier et al., 2001).
Emission heights for biomass burning
Table 3. Fractional distribution of effective emission heights (%) in prescribed height intervals of biomass burning emissions.
(6) In future FT updates, emissions will be scaled to trends in selected FAO crops, using the same ratios as weight factors for the crop residue to net crop production as used in the last full EDGAR dataset.
2.6 Waste (EDGAR W; IPCC sector 6)
Landfill emissions (W10) (net CH4) from OECD countries and a few EIT countries were extrapolated from 1995 onwards using the country-specific 1995-2000 trends reported to the UNFCCC; for other OECD countries the reported OECD total trend of 9% decrease was used. For other countries (developing and other EIT), where CH4 recovery is assumed to be insignificant, the 1990-1995 emission trend was extrapolated since annual landfill emissions are less sensitive to recent changes in activity data as they are the sum of emissions from waste which was deposited several decades ago. Waste water treatment and disposal emissions (W20 and W30) of net CH4 have been extrapolated using the 1995-2000 trend of total national population, except for wastewater treatment by OECD countries for which country-specific 1995-2000 trends reported to the UNFCCC were used or the reported OECD total trend of 2% increase. N2O from waste water treatment (W20) from OECD countries and a few EIT countries was extrapolated using the country-specific 1995-2000 trend reported to the UNFCCC; for other OECD countries the reported OECD total trend of 6% increase was used. For other countries (developing and other EIT the 1995 emis-sions were extrapolated using the 1995-2000 trend of total national population. N2O from waste water disposal (W30) was extrapolated using the 1995-2000 trend of total national population. Finally, emissions from uncontrolled waste incineration (W40) (miscellaneous compounds) were kept constant since these are generally very small and no good indicator is available for this source.
2.7 Fluorinated gases: HFC's, PFC's, SF6 (EDGAR category H; IPCC sector 2F-G)
Generally, for the largest sources of HFC, PFC and SF6 emissions, country-specific or OECD-average trends reported to the UNFCCC were used for OECD countries, while using reported global total emissions, production or consumption trends as a proxy for non-OECD countries (Table 4). Below, details of the extrapolation of these key F-gas sources are described. The OECD-average and global trend figures mentioned are listed in a separate table (Table B.3).
Table 4. Sectors for which country-specific and OECD'90 average emission factors have been calculated based on reported emissions to the UNFCCC. If marked + the global total emission trend have been estimated for that source category, if marked (1), the emissions have been assumed to be constant. See Table B.3 for actual OECD'90 average and global trend values used.
HFC-23 byproduct emissions from HCFC-22 manufacturing (I91) from OECD countries were extrapo-lated from 1995 using the country-specific 1995-2000 trend reported to the UNFCCC; for other OECD countries the reported OECD total trend was used. For all other countries the global total HCFC-22 production trend reported by RAND of 0% was used. Emissions from HFC-134a use (H14) were dealt with in the same way, using a 1995-2000 trend factor of 2.7 for non-reporting OECD countries (including the UK); however, for non-OECD countries the global total HFC-134a emissions trend reported by RAND of a trend factor of 4.3 was used. For emissions from other HFC use (HFC-125, 143a, 152a and 227ae) from OECD countries the same procedure was followed. For non-OECD countries, for HFC-152a and 227ae emissions the same factor of 4.3 as for HFC-134a was used, while the relatively small emissions of HFC-125 and 143a emissions from non-OECD countries were assumed to remain constant; the same was done for the very small HFC-23 use emissions. These emissions from the use of HFC-23, 125 and 143a account for 4% of global total HFC emissions in 1995.
PFC byproduct emissions from aluminium production (I24) (CF4 and C2F6) from OECD countries were extrapolated from 1995 using the country-specific 1995-2000 trend for CF4 reported to the UNFCCC; for other OECD countries the reported OECD total trend was used. For all other countries the 1995 emissions were extrapolated using the 1995-2000 trend of country-specific primary aluminium production reported by USGS. PFC emissions from semiconductor manufacture (H21) and from PFC use as solvent (H26) from OECD countries were extrapolated from 1995 using the country-specific 1995-2000 trend reported to the UNFCCC; for all other countries the reported OECD total trend was used. PFC emissions from all other sources, which only account for 5% of global PFC emissions, were assumed to remain constant.
SF6 emissions from semiconductor manufacture (H36) and from use in magnesium production (H45 and H46) from OECD countries were extrapolated from 1995 using the country-specific 1995-2000 trend reported to the UNFCCC; for other OECD countries the reported OECD total trend of these sources was used. The other SF6 sources were dealt with in the same way (country-specific or the OECD average trend), but for OECD countries the trend for total national SF6 emissions from these countries was used as a proxy. For all other countries the global total consumption trend reported by RAND was used, except for magnesium production where the RAND trend was much lower than the UNFCCC trend for OECD countries and therefore the latter trend as used as a proxy.
The format of standard reporting of EDGAR 32FT2000 emissions at source, region, country and grid level is in the same format as was done for the provision of the EDGAR 3.2 datasets at the website. Table 5 shows the number of standard reporting source categories for each compound. Please note that for some sources multiple datasets have been provided.
Table 5. Number of source categories used for global total trend estimates cf. FT method.
4. Comparison with aggregated global total emission trends published earlier
Until recently, annually estimates for global total emissions of greenhouse gases were made by RIVM/MNP for years after 1995 by extrapolation per major source category (see Table 6). Global activity trends were used to estimate the source's emissions in more recent years, but were corrected if the 'im-plied emission factor' - i.e. the division of annual sectoral emissions by the activity data selected as vol-ume indicator and (used for the extrapolation) - of the global 1990-1995 emissions show a significant trend or if substantial changes are known from national submissions to the UN Climate Secretariat (UNFCCC). These more aggregate global total source trends are published annually on the RIVM/MNP website as part of the Environmental Data Compendium, in Dutch abbreviated as 'MNC' (RIVM/MNP, 2004) and in Part 3 of the annual IEA publication 'CO2 from fuel combustion' (Olivier, 2004).
Table 6. Number of source categories used for global total trend estimates cf. MNC method
Comparison of Tables 5 and 6 shows that the MNC method, using only half the number of sources com-pared to the FT method, analyses global trends at a much more aggregated source level than the Fast Track method. More importantly, the MNC method does not show any regional or country-specific details, so it does not capture the more delicate differences in volume trends by countries with higher and with lower emission factors. For compounds such as methane and nitrous oxide these differences result in strikingly different 1995-2000 trends, for both 5%-points different as shown in Table 7: for CH4 now 6% vs. 1% and for N2O now also 6% vs. 1%. This illustrates that apparently the much more aggregated MNC method sometimes does not capture key determining trends. Although the FT method uses another method for estimate large-scale biomass burning emissions, an analysis of total emissions excluding these sources does not alter the main conclusions (see Appendix A).
Table 7. Comparison global total 2000 emissions of CH4 and N2O of FT** and MNC methods.
EDGAR 32 data; ** FT dataset: BB-AVG-EF32.
Future updates of the aggregated global emission trend estimates using the MNC method, which is still useful for the most recent years for which detailed statistics are not yet complete and thus the much more refined FT method is not yet applicable, will therefore be calibrated to the more detailed global total estimate per source category as determined for particular years by the Fast Track method.
It is concluded that the Fast Track method used to estimate recent 5-year emissions trends is a major improvement compared with the MNC method used for greenhouse gases. It uses twice the number of sources of the MNC method and uses actual country-specific trends rather than global trends in the year's prior to the year of extrapolation. The FT method provides more accurate trend estimates than the global 'MNC' method:
Further analysis, e.g. comparison with the EDGAR 4 data for 1995 and 2000 that are currently under development, may provide quantitative indications of the trend accuracy of the FT method. It should, however, also be concluded that the accuracy of the MNC method for emission trend estimation of periods of five years or more may be limited, in particular when it comprises major sources where the emission in-tensities are highly regionally stratified.
Not discussed above, but observed when compiling the FT2000 datasets, are the following:
6. Citation and other key references for the EDGAR 32FT2000 dataset
Please use the following references to cite the EDGAR 32FT2000 datasets or to read details on the construction of the dataset:
General documentation on EDGAR 3.2 can be found in: Olivier, J.G.J. and J.J.M. Berdowski (2001). Global emissions sources and sinks. In: Berdowski, J., Guicherit, R. and B.J. Heij (eds.) The Climate System, pp. 33-78. A.A. Balkema Publishers/Swets & Zeitlinger Publishers, Lisse, The Netherlands. ISBN 905809 255 0.
Appendix A: Comparison of CH4 and N2O emissions in 2000: the FT and MNC method
Although part of the differences stem from different data sources and assumptions made for large-scale biomass burning, the main conclusion is that a more detailed calculation for a few key sources is largely responsible for the observed differences in the global total trend estimate: coal and gas production for CH4 and Nitric Acid and Adipic Acid production and animal waste applied to soils for N2O.
Table A.1. Methane emissions per source category in 2000 (FT dataset: BB-AVG-EF32)
Table A.2. Nitrous oxide emissions per source category in 2000 (FT dataset: BB-AVG-EF32)
Appendix B: Miscellaneous data tables
Table B.1. OECD'90 average trends 1995-2000 for CH4 and N2O from selected fossil fuel and industrial processes as reported to UNFCCC.
Table B.2. Ratios of large-scale biomass burning emission factors EF_32 and
Table B.3. OECD average trends and global trends 1995-2000 of F-gas sources used for extrapolation when country-specific reported trends were not available (sources in red and BOLD are the largest sources, accounting for 95% of the global total) (trends between brackets are values copied from values for other sources).