Data sources and methodology

Introduction

The Emission Database for Global Atmospheric Research, in short the EDGAR 3 system, has been developed jointly by the Netherlands Organisation for Applied Scientific Research (TNO) and the National Institute of Public Health and the Environment (RIVM) with financial support from the Dutch Ministry of the Housing, Spatial Planning and Environment (VROM) and the Dutch National Research Programme on Global Air Pollution and Climate Change (NRP). The aim of the EDGAR system, which started in 1992, is to provide global anthropogenic emissions of greenhouse gases CO2, CH4, N2O, HFCs, PFCs and SF6 and of precursor gases CO, NOx, NMVOC and SO2, per source category, both at country/region levels as well as on a 1x1 degree grid. It is meant to serve as reference database for policy applications, e.g. to provide RIVM's integrated climate change model IMAGE 2 with emissions data and for assessments of potentials for emission reductions, as well as for scientific studies by providing gridded emissions as input for atmospheric models. The latter function is part of the Global Emissions Inventory Activity (GEIA), that combines efforts to produce gridded inventories for all compounds relevant for the modelling activities within the International Atmospheric Chemistry Programme (IGAC) of the International Geosphere-Biosphere Program (IGBP). 

Activity data were mostly taken from international statistical data sources and emission factors were selected mostly from international publications to ensure a consistent approach across countries. RIVM and TNO have made all reasonable efforts to ensure that the information was generated correctly, but it is the responsibility of RIVM and TNO to modify activity data when required to arrive at complete time series and for selecting the emission factors. It is stressed that the uncertainty in the resulting dataset at national level may be substantial, especially for methane and nitrous oxide, and even more so for the F-gases. The uncertainty is caused by the limited accuracy of international activity data used and in particular of emission factors selected for calculating emissions on a country level (Olivier et al., 1999, 2001). However, since methods used are comparable with IPCC methodologies (see below) and global totals comply with budgets used in atmospheric studies and the data were based on international information sources, this dataset provides a sound basis for comparability.
Although this dataset has been constructed with great care, RIVM and TNO do not accept any liability from using the data provided in this report including any inaccuracies or omissions in the data provided. For details on  uncertainty and  caveats identified in the dataset, as well as more detailed source category estimates, we refer to this website.

Source definitions

For  carbon dioxide : 

• Fuel combustion refers to fossil fuel combustion and non-energy/feedstock use (IPCC category 1A) estimated using the IPCC Sectoral Approach;
• Fugitive refers to flaring of associated gas in oil and gas production (IPCC category 1B)
• Industry refers to cement production (IPCC category 2); and 
• Other refers direct emissions from tropical forest fires plus 10% of biofuel combustion emissions, which is the fraction assumed to be produced unsustainably (IPCC category 5).

For  methane : 

• Energy comprises production, handling, transmission and combustion of fossil fuels and biofuels (IPCC category 1A and 1B); 
• Agriculture comprises animals, animal waste, rice production, agricultural waste burning (non-energy, on-site) and savannah burning (IPCC category 4); 
• Waste comprises landfills, wastewater treatment, human wastewater disposal and waste incineration (non-energy) (IPCC category 6); 
• Others include industrial process emissions and tropical and temperate forest fires (IPCC categories 2 and 5).

For  nitrous oxide : 

• Energy comprises combustion of fossil fuels and biofuels (IPCC category 1A and 1B); 
• Agriculture comprises fertiliser use (synthetic and animal manure), animal waste management, agricultural waste burning (non-energy, on-site) and savannah burning (IPCC category 4); 
• Industrial Processes comprises non-combustion emissions from manufacturing of adipic acid and nitric acid (IPCC Source Category 2); and
• Others include N2O usage, tropical and temperate forest fires, and human sewage discharge and waste incineration (non-energy) (IPCC Source Categories 3, 5 and 6).

For  HFCs ,  PFCs and  SF6 :

• HFC emissions comprise by-product emissions of HFC-23 from HCFC-22 manufacture and the use of HFCs (IPCC Source Categories 2E and 2F);
• PFC emissions comprise by-product emissions of CF4 and C2F6 from primary aluminium production and the use of PFCs, in particular for semiconductor manufacture (IPCC Source Categories 2C, 2E and 2F); and
• SF6 emissions stem from various sources of SF6 use, of which the largest is the use and manufacture of Gas Insulated Switchgear (GIS) used in the electricity distribution networks (IPCC Source Categories 2C and 2F).

Data sources and methodology

Energy / Fugitive / Biofuel

Data for fossil fuel production and use for 112 countries are taken from the IEA energy statistics for OECD and non-OECD countries 1970-1995 (extended energy balances, in ktoe units) (IEA, 1997). For the countries of the former Soviet Union (SU) a modified dataset was used to achieve a complete time series for the new countries for 1970-1995 of which the sum converges to the older dataset for the total former SU. For another 71 countries, the aggregated IEA data for the regions 'Other America', 'Other Africa' and 'Other Asia' have been split using the sectoral IEA data per region and total production and consumption figures per country of hard coal, brown coal, gas and oil from UN energy statistics (UN, 1998). Note that the EDGAR 3.0 data are based on IEA statistics published in 1997 and thus may differ somewhat from more recent IEA datasets; in particular for countries of the former Soviet Union the IEA data have been updated considerably. Moreover, for estimating CH4 emissions, hard coal and brown coal production data have been split into surface and underground mining based on various national reports.
Biofuel data for developing countries in 1990 have been based on Hall et al. (1994), with biofuel type splits from EDGAR 2.0 (Olivier et al., 1996, 1999), which includes vegetal waste used as fuel. Data for the time series 1970-1995 were based on the trend per country in urban and rural population. However, for Latin American countries biofuel statistics from OLADE were used (OLADE, 1999, pers. comm). Fuelwood and charcoal consumption (also production) in OECD'90 and EIT countries  (Economies-In-Transition, i.e. Eastern Europe and former Soviet Union countries) were based on FAO (1998b), thereby replacing any IEA data for biofuel combustion in the 'Other sector' in these countries. For biofuel combustion in industry and power generation in OECD'90 countries, the data come from the IEA dataset (IEA, 1997). However these data were often not provided for all years and all countries.

Emission factors for CO2, CH4 and N2O from fossil fuel production and use are described in Olivier et al. (1999), except for CO2 from gas flaring/venting, which were based on data compiled by CDIAC (Marland, 1998, pers. comm.) from data collected by the U.S. Geological Survey. For N2O from gasoline cars in road transport, the fraction of cars equipped with a catalytic converter was taken into account (based on data from Eurostat and others). The factors for biofuel combustion and charcoal production are based on a review made for the Revised 1996 IPCC Guidelines (IPCC, 1997). For CO2 from biofuels, it was tentatively assumed that 10% was produced unsustainably. For methane emissions from coal mining the methane recovery for ten countries amounted to about 1 Tg in 1990 (of which about half was allocated to the USA and Germany). Recovery in 1995 was estimated at 2 Tg (Thakur et al., 1996; Bibler et al., 1998; and national reports to Climate Convention).

Industrial Processes

Production data of cement, nitric acid, iron and steel, and various chemicals were based on UN Industrial Commodity Statistics (UN, 1998). However, for many countries interpolations and extrapolations were necessary to arrive at complete time series per country for 1970-1995. Special attention had to be given to new EIT countries, in particular to former USSR countries, to match the older totals for the former countries. Cement production data were supplemented with data from the USGS. For adipic acid production data were taken from SRI (1998) (smoothed and averaged); steel production was split into different technologies using data from IISI (1997), supplemented with UN data. For nitric acid (NA) production data are primary based on UN statistics. However, since industry estimates of global total production are substantially higher, the data set has been expanded, first by adding countries not included in the UN NA statistics, for which the amount of N in the production of nitrogen fertilisers according to FAO statistics was used as an estimate for NA production, secondly by increasing the official UN production statistics of nitric acid by 40% to arrive at the estimated global industry total of about 55 Mton HNO3.

Global annual total production of HCFC-22 and consumption of HFC-134a are based on AFEAS (1997). Primary aluminium production statistics per country from UN (1998) were combined with smelters types characterised by one of five process types according to Aluminium Verlag (1998). Global consumption data of PFCs for semiconductors are taken from Mocella (1993) and for SF6 per application from S&PS (1997) and Smythe (2000). These global totals were distribution over individual countries using related variables and statistics such as CFC consumption per country, per country semiconductor production and electricity use.

Emission factors for CO2, CH4 and N2O are described in Olivier et al. (1999). Note that emissions of CO2 from cement production are only a proxy for cement clinker production. The emission factors for NA production are based on IPCC (2000), assuming that in 1990 20% of global total production is equipped with Non-Selective Catalytic Reduction (NSCR) technology, all in OECD'90 countries, and that for other plants the emission factor in 1990 is the average of the IPC default for non-NSCR plants, whereas the emission factors for 1975 and before have been assumed to be equal to the IPCC defaults for  'old plants'. The emission factors for the F-gases were taken from various sources (Olivier and Bakker, 2000). We note that both the variables for distributing global total consumption and the emission factor may vary widely between different plants. This means the emissions at country level of the F-gases should more or less be considered as an order of magnitude estimate.

Solvent and other product use

For N2O from the use of anaesthesia in hospitals, a fixed amount of N2O per capita in OECD'90 countries was used, tentatively set at 25 g/cap/year, based on Kroeze (1994).

Agriculture

Activity data for livestock number were taken from FAO (2000), which were combined with information on animal waste generated per head in IPCC (1997) to estimate the total amount of animal waste. Net crop production was also taken from FAO (2000), with harvested areas of rice production split over different ecology types (rainfed, irrigated, deep water and upland) using the draft version of March 1977 the RICE-ECO database of FAO (Van Gnuu, 1997, pers. comm.). In addition, the total harvested area of rice production in China was increased by 40%, due to recognition that official harvested rice area statistics of China are largely underestimating the actual area (Denier van der Gon, pers. comm., 2000).
The fraction of agricultural waste associated with net crop production was based on a recent study be Smill (1999), whereas the fraction of agricultural residues burned on-site have been based on an analysis made by Bouwman (1997) and data reported in the Second National Communications. For OECD'90 countries 5% was assumed, for EIT countries 20% and for developing regions 30% - including amounts used as biofuel in developing countries, except for OECD'90 Europe, where a decreasing trend from 40% in 1970 to 5% in 1995 was assumed.
Emission factors for CH4 and N2O for enteric fermentation, animal waste (confined and outside), N-fixing crops were taken from the Revised 1996 IPCC Guidelines (IPCC, 1997), where a 1x1 degree grid map for non-dairy cattle from Lerner et al. (1988) and the annual average temperature per grid cell from New et al. (1999) was used to calculate the fraction of the countries in the three climate zones (cold, temperate, warm). Other additional information, such as factors for indirect emissions of N2O from agriculture, were taken from IPCC (1997) but were replacing values updated in IPCC (2000). However, the emission factors for CH4 from rice production in 1990 were taken from a review by Neue (1997); for the period 1970-1990 an emission factor improvement based on data of Van der Gon (1999, 2000) was assumed. For agricultural biomass burning the emission factors for CH4 and N2O were based on IPCC (1997).

Large-scale biomass burning

Biomass burning data (large-scale vegetation fires) were based on FAO reports providing ten year or five year averaged estimates per country of the change in forested areas for the 1970s, 1980s and the first half of the 1990s (FAO, 1993, 1995, 1998). Following the methodology described in the Revised 1996 IPCC Guidelines, these data were used as a proxy for estimating the amount of biomass being burned in tropical countries. Since there is no time-series data per country on this subject readily available, a smoothing function to construct a continuous time series per country for the 1970-1995 period was used. Tentatively, it was assumed that 50% of the biomass removed is burned. Given the uncertainty in this figure, the fraction oxidised is assumed to be 1. For OECD'90 and EIT countries, forest fire statistics for 1986-1997 have been included based on UN/ECE statistics of annual area burned (UN-ECE/FAO, 1996) combined with forest biomass densities per hectare from FAO (1995). There is a large uncertainty in the assumption for the carbon density of 0.5 and the fraction of carbon that is actually being burned of 0.5, and thus in the amount of burned carbon. The data selected, although often criticised for their limited accuracy are, however, well known and relatively well-documented.
 Emissions CO2 from deforestation and temperate vegetation fires are calculated according to IPCC (1997) and include only direct burning effects (thus no emissions due to decay of biomass). For large-scale biomass burning the emission factors for CH4 and N2O were also based on IPCC (1997), except for CH4 from deforestation fires, where the GEIA  value proposed by Veldt and Berdowski (1995) was used, and N2O where post-burn emissions (Bouwman et al., 1997) were used. The emission factors of CH4 and N2O used for temperate vegetation fires are the same that are used for other large-scale biomass burning activities. For accounting purposes, net CO2 emissions from temperate vegetation fires and savannah fires have been assumed to be zero (organic carbon in a short cycle).

Waste handling

For solid waste generation, the 1970-1995 trend in activity data per country has been based on a fit with international waste generation figures per capita for 1990 - as recently published by IPCC and EPA and references mentioned therein - with per capita income per country. This fit was also used to estimate the activity data for 1990, for countries not mentioned in IPCC (1997) and in an EPA report by Adler (1994). Country-specific fractions of total MSW generated that is disposed of in landfills were based on IPCC (1997). For most countries it was assumed that this fraction has remained constant over time. Many other parameters, such as the fraction of Degradable Organic Carbon (DOC) were also based on the Revised 1996 IPCC Guidelines; in addition, many others were estimated through consultation of experts (Olivier et al., 2001). The methodology used for the calculation of CH4 emissions from landfills in EDGAR 3.0 is a first order decay model resembling the description in the Revised 1996 IPCC Guidelines of the more complex Tier 2 method, taking into account that the generation of methane from landfills is not an instantaneous process. Thus, the methodology calculates emissions in a specific year as the sum of delayed emissions from all MSW deposited in past years. A 40-year integration period was used, assuming emissions from MSW deposited more than 40 years ago are negligible. Based on national reports submitted to the Climate Convention methane recovery amounts for eight OECD countries were included, amounting to about 2 Tg in 1990 and 4 Tg in 1995, about half of which was allocated to the United States.

For domestic and industrial wastewater discharged in city sewers and subsequently treated by municipal Waste Water Treatment Plants (WWTP), the approach based on per capita organics loading and industrial waste water generation selected by Doorn et al. (1997) was used as information on domestic wastewater generation rates is very sparse and because it is essentially the same as the default IPCC methodology (IPCC, 1997). Estimates are based on population data from the UN (1999), whereas waste water generation was based on industrial production statistics of the UN (1998) combined with waste water generation rates of Doorn et al. (1997). It is well known that in OECD countries, which cover about 60% of this source, a large fraction of the methane generated in municipal WWTPs is generally recovered. Therefore methane recovery for municipal WWTPs in OECD'90 countries was tentatively assumed to be 75%, effectively reducing the total emissions of OECD countries in 1990 by 0.6 Tg.

For untreated domestic waste water handling, treatment and disposal emission factors and other factors were based on Doorn et al. (1999), who distinguished disposal in septic tanks, latrines and sewers. The later was divided into sewage with municipal wastewater treatment and open sewers. Emission factors for CH4 from domestic wastewater in latrines or open pits and septic tanks and from stagnant open sewers (untreated wastewater) were based on Doorn et al. (1999). Here the same approach as for domestic WWTPs was followed, but distinguishing national population into three population groups: rural and urban, with urban population further split into high and low income groups. For the each of four municipal wastewater disposal types, region-specific and country-specific utilisation fractions were estimated for each of these three population categories. Emissions from open sewers were increased by 25% to account for the global amount of industrial wastewater annually discharged in municipal sewers. Globally, according to the assumptions of Doorn et al. (1999), this source of CH4 appears to be as large as emissions from landfills. 
For N2O from human sewage the default IPCC methodology was used, with protein intake per country for various years from FAO (2000); for the small emissions of N2O from DOC in wastewater from the meat processing industry the emission factor provided by Doorn et al. (1997) was applied.

In addition, for domestic waste burning (i.e. by households for non-energetic purposes) a fixed amount per urban capita burned per year by urban households in less developed countries was used. In rural areas of LDC it was assumed that there is no uncontrolled burning in addition to the agricultural residue burning and biofuel use that has already been accounted for in other source categories. In contrast, for industrialised countries, it was assumed that domestic waste burning only occurs in rural areas, where waste incineration regulation is less well controlled.

References

Adler, M.J. (1994) International anthropogenic methane emissions: estimates for 1990. EPA, Washington, DC (USA), EPA report no. 230-R-93-010, January 1994.

AFEAS (1997) Production, sales and atmospheric release of fluorocarbons through 1995. Alternative Fluorcoarbons Environmental Acceptability Study (AFEAS) Program Office, Washington DC, USA. Also see: http://www.afeas.org/production_and_sales.html

Aluminium Verlag (1998) Primary Aluminium Smelters and Producers of the World, Vol. 2, Compiled by R. Pawlek. Aluminium Verlag, Düsseldorf, Germany.

Bibler, C.J., Marshall, J.S. and R.C. Pilcher (1998) Status of worldwide coal mine methane emissions and use. Int. J. of Coal Geology, 35, 283-310.

Bouwman, A.F., Lee, D.S., Asman, W.A.H., Dentener, F.J., Van Der Hoek, K.W. and J.G.J. Olivier (1997) A Global High-Resolution Emission Inventory for Ammonia, Global Biogeochemical Cycles, 11:4, 561-587.

Denier van der Gon, H. (1999) Changes in CH4 emission from rice fields from 1960 to 1990s 2. The declining use of organic inputs in rice farming. Global Biogeochemical Cycles, 13, 1053-1062.

Denier van der Gon, H. (2000) Changes in CH4 emission from rice fields from 1960 to 1990s 1. Impacts of modern rice technology. Global Biogeochemical Cycles, 14, 61-72.

Doorn, M.R.J., Strait, R.P., Barnard, W.R. and B. Eklund (1997). Estimates of global greenhouse gas emissions from industrial and domestic waste water treatment. Report no. NRMRL-RTP-086. R 8/18/97. Pechan & Ass., Durham.

Doorn, M.J. and D.S. Liles (1999) Quantification of methane emissions and discussion of nitrous oxide, and ammonia emissions from septic tanks, latrines, and stagnant open sewers in the world. EPA, Washington DC, USA. EPA report EPA-600/R-99-089, October 1999.

FAO (1993) Forest resources assessment 1990; Tropical countries. FAO, Rome. Report FP-112.

FAO (1995) Forest resources assessment 1990; Global analysis. FAO, Rome. Report FP 124.

FAO (1998) State of the Worlds Forests 1997. FAO, Rome.

FAO (1998b) FAOSTAT Forestry Data. Data downloaded from www.apps.fao.org on 12-3-1998.

FAO (2000) FAOSTAT Agricultural Data.

Hall D.O., Rosillo-Calle F. and Woods J. (1994) Biomass utilization in households and industry: energy use and development, Chemosphere 29, 1099 1119.

IEA/OECD  (1997) Beyond 20/20, Release 4.1, Energy Balances of OECD Countries, Energy Balances of Non-OECD Countries. Ivation DatasystemsInc. 1997.

IEA/OECD (2002) Beyond 20/20, Release 5.2, Energy Balances of OECD Countries, Energy Balances of Non-OECD Countries. Ivation DatasystemsInc. 2002.

IISI (1997) 1996 Steel Statistical Yearbook, International Iron and Steel Institute, Brussels.

IPCC (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories. IPCC/OECD/IEA, Paris.

IPCC (2000) Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories, IPCC-TSU NGGIP, Japan.

Kroeze, C. (1994) Nitrous oxide (N2O). Emission inventory and options for control in the Netherlands. RIVM, Bilthoven. Report no. 773001 004.

Lerner, J., Matthews, E. and Fung,I. (1988) Methane emission from animals: a global high resolution database, Global Biogeochemical Cycles 2, 139-156.

Mocella, M.T. (1993) Production and Uses of C2F6. Proceedings of the Workshop on Atmospheric effects, origins, and options for control of two potent greenhouse gases: CF4 and C2F6. US EPA, Global Change Division, April 21-22, Washington, D.C.

Neue, H.U. (1997) Fluxes of methane from rice fields and potential for mitigation. Soil Use and Management, 13, 258-267.

New, M.G., Hulme, M. and P.D. Jones (1999) Representing 20th century space-time climate variability. I: Development of a 1961-1990 mean monthly terrestrial climatology. J. Climate, 12, 829-856.

Olivier, J.G.J. and J. Bakker (2000) Historical global emission trends of the Kyoto gases HFCs, PFCs and SF6. Proceedings of "Conference on SF6 and the Environment: Emission Reduction Strategies", November 2-3, San Diego. EPA, Washington DC, USA.  Conference Proceedings published at http:/www.epa.gov/highgwp1/sf6/partner_resources/proceedings.html

Olivier, J.G.J., Berdowski, J.J.M., Peters, J.A.H.W., Bakker, J., Visschedijk, A.J.H. and J.P.J. Bloos (2001) Applications of EDGAR. Including a description of EDGAR V3.0: reference database with trend data for 1970-1995. RIVM, Bilthoven. RIVM report 773301 001/NRP report 410200 051. Available online at www.rivm.nl.

Olivier, J.G.J., Bouwman, A.F., Berdowski, J.J.M., Veldt, C., Bloos, J.P.J., Visschedijk, A.J.H., Van der Maas, C.W.M. and P.Y.J. Zandveld (1999) Sectoral emission inventories of greenhouse gases for 1990 on a per country basis as well as on 1o x 1o. Environmental Science & Policy, 2, 241-264.

Olivier, J.G.J., Berdowski, J.J.M., Peters, J.A.H.W., Bakker, J., Visschedijk, A.J.H. and J.P.J. Bloos (2001) Applications of EDGAR. Including a description of EDGAR 3.2: reference database with trend data for 1970-1995. RIVM, Bilthoven. RIVM report 773301 001/NRP report 410200 051. Available online at  www.rivm.nl/bibliotheek/rapporten/410200051.htmlRIVM (2002) Environmental Compendium 2002. At www.rivm.nl/environmentaldata/index.html

Smill, V. (1999) Nitrogen in crop production: an account of global flows. Global Biochemical Cycles, 13, 647-662.

Smythe, K. D.  (2000) Production and Distribution of SF6 by End-Use Application. Proceedings of "Conference on SF6 and the Environment: Emission Reduction Strategies", November 2-3, San Diego. EPA, Washington DC, USA. Conference Proceedings published at:http:/www.epa.gov/highgwp1/sf6/ partner_resources/proceedings.html

S&PS (1997) Sales of sulphur hexafluoride (SF6) by end-use applications. Annual sales for 1961 through 1996. Sales projections for 1997 through 2000. Washington, USA, March 1997. Science & Policy Services (S&PS), now RAND.

SRI (1998) Adipic acid production data 1974-1997. SRI Consulting, Menlo Park, USA. Table dated January-1998, received 30-1-1998.

Thakur, P. C., H. G. Little and W. G. Karis (1996) Global Coalbed Methane Recovery and Use, in: Riemer, P. and A. Smith (eds.) (1996) Proceedings of the International Energy Agency Greenhouse Gases Mitigation Options Conference, Pergamon-Elsevier, 789-794.

UN (1998) Industrial commodity production statistics 1970-1995. UN Statistical Division, New York. Data file received 30-3-1998.

UN (1999) UN World Population Prospects, 1996 Revision. UN Population Division, New York.

UN/ECE-FAO [United Nations Economic Commission for Europe & Food and Agriculture Organisation of the United Nations] (1996) Forest Fire Statistics 1993-1995, Timber Bulletin, Vol. XLIX, No. 4, ECE/TIM/BULL/49/4, United Nations, 1996.

Veldt C. and Berdowski J.J.M. (1995) GEIA - Note on the combustion of biomass fuels (Emission factors for CO, CH4  and NMVOC). TNO-MW, Delft. TNO Techn. Report R 94/218