Date: 22 August 2005
Version: 2; no. 2-5 added; no. 6-8 copied from documentation.
Version 3: no. 9 added.
Version 4 (43Kb): no. 9 expanded.
Version 5: no. 10 added.

1. FT2000: Energy Transformation sector (F30):  The CO₂ emissions from the Energy Transformation sector (F30), that comprises coke ovens, oil refineries etc., are calculated from the balance between fuel inputs and outputs. Due to inconsistencies in the energy dataset, the CO₂ emissions from this source category include negative emissions for 12 countries. Both in country/region tables and in the grid file for the F30 category these negative numbers should be replaced by 0 (zero). See this file  (15Kb) for details.
2. FT2000: Middle and high latitude grassland fires (L47): For large-scale biomass burning a new source category “L47” has been defined. Based on the ecosystem database of Olson et al. (1983), some fires detected by satellite in agricultural regions were attributed to grassland fires (L42 and L47). This new source category has an insignificant overlap with the EDGAR category agricultural waste burning (L43) (for more details see the  FT2000 documentation online or as pdf-document (183Kb)).
3. FT2000: Gridded aircraft emissions below 1 km (F57): The first FT2000 aircraft emissions datasets do not include *only* aircraft emissions near airports “F57”, the so-called Landing and Take-Off emissions below 1 km altitude, but accidentally are the sum of *all* aviation emissions at all altitudes, mapped on the LTO map for surface emissions (i.e. in the lowest altitudes band 0-1 km). Since globally LTO emissions are about 12.5%, 16.9% and 13.3% of total aviation emissions of NO_x, CO and NMVOC, respectively, and 11.3% of CO₂, the correct F57 LTO emissions can be calculated by multiplying all FT2000 F57 datasets by this fraction. Also the global total anthropogenic emissions are now slightly overestimated and should be adjusted by the same amount.
4. FT2000: Gridded aircraft emissions – altitudes > 1 km (F57): The FT2000 aircraft emissions dataset provided are only the emissions below 1 km (but see caveat above). For other altitude bands, the emissions can be calculated from the EDGAR 3.2 aircraft emissions for 1995, by multiplying these datasets by the global total aircraft fuel consumption 1995-2000 growth factor of 1.161. Please note that for some compounds no gridded aircraft emissions at altitudes above 1 km have been provided in EDGAR 3.2, as they are not occurring (CH₄) or negligible (N₂O and SO₂).
5. FT2000: Format and definition of gridded datasets: The format of the 1x1 degree gridded datasets is identical to that of the EDGAR 3.2 gridded data. See this file (34Kb) for details. Also the source definitions are the same, except for large-scale biomass burning for which a) a new category “and high latitude grassland fires” has been added (see caveat no, 2); and b) multiple (and monthly) datasets have been provided, using either multi-year (1997-2002) averaged activity data or actual year 2000 activity data and/or updated emission factors (for more details see the FT2000 documentation online or as pdf-document (183Kb)).
6. FT2000: Fossil fuel combustion (Fxx) – Activity data: 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 likely 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 developing countries.
7. FT2000: Fossil fuel combustion (Fxx) – ‘Other IEA regions’: The aggregated IEA data for the regions ‘Other Latin America’, ‘Other Africa’, and ‘Other Asia’, i.e. for 72 countries, have been split into country data using population density figures from FAO. Note that this country allocation scheme differs from the one used for the EDGAR 3.2 dataset, and may introduce discontinuities for these countries between the 2000 FT emissions and the 1995 emissions in EDGAR 3.2, in particular for former USSR countries and for developing countries.
8. FT2000: Biofuel combustion in industry (B10) and power generation (B20): Note that for these sources 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. 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. 
9. FT2000: Spreadsheet files corrected on 19 August 2005:    

• FT2000: Corrected PFC emissions in 5 former USSR countries (H2x+I24): In compiling the spreadsheet with emissions per gas/country the emissions of five countries was mixed up. This refers to Azerbaijan, Estonia, Georgia, Tajikistan and Ukraine. Both PFC and F-gas totals were corrected. See the xls-file for details.  
• FT2000: CO₂ emissions from large-scale biomass burning: CO₂ emissions from L42 (savannas), L44 (Temperate Vegetation Fires) and L47 (Temperate Grassland Files) were accidentally included in the xls-files; these have been removed 18-8-05 (see notes added to these files).   
• FT2000: CH₄ emissions from oil production (F80): CH₄  emissions from total oil production/processing sources (F80) were accidentally double included in the xls file with country-specific emissions, i.e. in addition to the subcategories F81, F83 and F84, also in country totals; this column has been removed and national totals corrected 19-8-05 (see note added to this file).   
• FT2000: N₂O emissions from oil production (F80): The N₂O emissions from total oil production/processing sources (F80) were accidentally in error  the xls-files with regional and country-specific emissions; this row/column has been corrected 19-5-05 (see note added to these files).   
• FT2000: N₂O emissions from post-burn effects of deforestation (L45): This source category was accidentally missing the xls files with regional and country-specific emissions; this row/column has been added 19-5-05 (see note added to these files).  

10. FT2000: HFC-227ea error identified on 22 August 2005: In the calculation of the HC-227ea emissions an error has been made. Instead of a trend factor of 1 (i.e. emissions equal to 1995 values) for most countries a much larger factor of 13.9 for OECD countries and 4.3 for non-OECD countries was used. This affects all countries with non-zero HFC-227ea emissions (except AUT, BEL, FRA, DEU, IRL, ITA, NOR, ESP, SWE, NZL and JPN, which which reported country-specific trends were used).

1. Former USSR: Activity data for this regions is weak, in particular for energy consumption and production, since many of the time series for the 15 new countries does often not go back to 1970, or even 1990 in some cases. Time series were completed using old time series for the total USSR, however these did not always match in the 1990-1992 period.
2. Temperate forest fires: forest biomass densities were applied to the areas burned; no correction was made  (yet) for the fraction of the area burned that is not forested. For NMVOC we used a higher emission factor than for tropical forest fires based on a reference available at that time. Data presented in a recent review paper by Andreae and Merlet (2001) suggests that the emission factor would be approximately similar to tropical forest fires.
3. Road transport: we assumed no significant 1990-1995 trend in the CO emission factors from road transport. Recent data suggests that the CO emission factors have decreased in most countries. Therefore the CO emissions from road transport in EDGAR 3.2 in 1995 tend to be somewhat biased to the high side of the uncertainty range. For example, CO emissions from road transport in the USA and OECD Europe may have decreased by about 2 to 5% per year instead of overall increases reported in EDGAR 3.2. Similar differences in trends may have occurred in other regions.
4. Other transformation sector: this sector is a notoriously difficult sector, where energy statistics are often incomplete or otherwise inconsistent, also for larger countries.
5. Agricultural waste burning: for a number of countries and for some years, the amount of agricultural residue calculated to be produced multiplied by the assumed fraction being burned on the field or as vegetal biofuel was less than the amount assumed to be consumed as vegetal biofuel, which was based on independent assumptions. In those cases there are no agricultural waste burning emissions in the dataset for those years and countries.
6. Deforestation emissions: the area annually deforested was calculated from 10- and 5-year averages, which were then smoothed in order to remove discontinuities in the time series. The procedure applies, however, resulted in some cases (some countries and for some years) in negative removal figures. In these cases, the amount of deforestation was assumed to be zero. Thus, in those cases there are tropical forest fire emissions in the dataset.
7. Solvent use: in the present dataset it is assumed that solvent use has remained constant in the 1990-1995 period. The results from a study by TNO on trends in consumption for specific uses could not yet be incorporated in the EDGAR 3.2 dataset.
8. Gridded emissions: Initially, the surface emissions datasets will not include the aircraft emissions near airports, the so-called Landing and Take-Off emissions below 1 km altitude.
9. CO2 emissions 1970: The 1970 CO2 emissions for the non-OECD regions 06 to 13 and for the fossil fuel sources (source codes F10-F90) were estimated from the 1972-1971 trend, since IEA energy data for non-OECD countries start in 1971.
10. CO2 emissions from biofuel use: The CO2 emissions from biofuel combustion (source codes B10-B50) have been reported assuming a 'unsustainable' fraction of 10%; i.e. assuming that 10% was produced in a non-sustainable way: actual combustion emissions of CO2 are 10 times as high (when neglecting incomplete oxidisation products). CO2 emissions as accounted for in the UN Climate Convention would be 0, since these are assumed to be included and reported in the Land-Use Change and Forestry (LUCF) category (L40).
11. N2O from adipic acid production: The present dataset assumes adipic acid manufacture in the Russian Federation. However, according to official reports to the UN Climate Secretariat, there is no adipic acid production in the Russian Federation.
12. PFC emissions from aluminium production: CF4 and C2F6 emissions from aluminium production (I24) 1970-1985 have been corrected (also region totals) in file pfc-trend-v33.xls on 27-6-2002.
13. CO2 extended table for 1990 and 1995: Rows with sectoral emission data for Fossil fuels (F) and Biofuels (B) were accidentally mixed; corrected in file co2-v32-90-95.xls on 10-10-2002. 
14. Road transport: CO emission factor for Germany modified in Version 3.3 dataset.