EDGAR anthropogenic emissions

Emissions history

The construction of the historical anthropogenic emissions of NO$_{x}$, NH$_{3}$, SO$_{2}$, CO, and NMV is described in [van Aardenne et al., 2001]. (The CH$_{4}$ emissions in this database are not used.) The emissions are based on EDGAR (Emissions Database for Global Atmospheric Research, webpage
http://arch.rivm.nl/env/int/coredata/edgar/index.html, [Olivier et al., 1999]). http://arch.rivm.nl/env/int/coredata/edgar/index.html, [Olivier et al., 1999]).
[van Aardenne et al., 2001] have estimated trends in these emissions on the basis of demographic, economic, agricultural, and technological developments during the past century. For example, the abatement strategies have caused the SO$_{2}$ emissions to strongly decrease in Western Europe and the United States, while economic growth has caused an increase in emissions of NO$_{x}$, CO, and of SO$_{2}$ in parts of the world. The time resolution of the database is 10 years from 1890 until 1970 and 5 years thereafter, up to 1990. After 1990 the emission data are extrapolated from the 1990 values using CO2 emissions statistics from [Marland et al., 2000]. These data are also available on-line at
http://cdiac.esd.ornl.gov/trends/emis/em_cont.htm To obtain emission data for specific years a linear interpolation between the base years is applied.

One should be careful for years later than 1995.

EDGAR2 Emission categories

EDGAR2 distinguishes the following categories:

Abbreviation	Category
AGL	Agricultural soil emissions
NAT	Natural soil emissions (microbes)
AGR	Agricultural waste burning
ANM	Animal manure emissions
CROP	Crop emissions
FER	Fertilizer emissions
DEF	Deforestation (once) emissions
HIS	Emissions from industrial processes
BF3	Biofuel combustion emissions
E_1	Fossil fuel consumption related emissions
LAN	Land use emissions
SAV	Savanna burning related emissions
OTS	Other transformation sector emissions
NRT	Ships (non road transport, but without aircraft)
FP2	Fossil fuel production emissions
SOL	Solvent emissions

Only AGL has monthly resolution. E_1 (surface fossil fuel combustion) includes categories industrial sector (F10, not be confused with HIS), power generation (about 20 %, F20), road transport (about 31 %, F51), and RES+COM+OTH(F40).
We used the following emissions in TM3:

Abbreviation	NO$_{x}$	NH3	SO2	CO	CH4	NMV
AGL	x
NAT		x
AGR	x	x	x	x		x
ANM		x
CROP		x
FER		x
DEF	x	x	x	x		x
HIS	x	x	x	x		x
BF3	x	x	x	x		x
E_1	x	x	x	x		x
LAN		x
SAV	x	x	x	x		x
OTS			>=1970	>=1970		>=1970
NRT			>=1970	>=1970		>=1970
FP2						x
SOL						x

The input filenames are CategorySpecies.d, e.g. AGRCO.d. The EDGAR emissions are preprocessed by program edgar. Since AGLNOX and NATNH3 are natural emissions they are not discussed not but in the chapter on natural emissions. I further noted that ammonia is not included in the EDGAR2 database
http://arch.rivm.nl/env/int/coredata/edgar/v2/ and that the categories of the EDGAR-GEIA NH3 emissions
http://arch.rivm.nl/env/int/coredata/geia/data_NH3.html differ from ours described above. For EDGAR-GEIA NH3 emissions the categories are:

• L20 Domestic Animals
• L30 Wild Animals
• L11 Synthetic Fertilizers
• L30 Biomass Burning (incl. L41)
• B00 Biofuel Combustion
• L13 Crops
• L44 Human Population
• F00 Fossil fuel use
• I30 Industry
• N10 Soils Under Natural Vegetation (Natural emission)
• N40 Oceans (Natural emission)

EDGAR3.2 Categories

EDGAR3.2, used in TM5, follows the IPCC guidelines more in detail. It distinguishes the following categories:

IPCC	EDGAR3.2	Description
	F	Fossil fuel use
		Fossil fuel combustion
1A2	F10	Industry (excluding coke ovens, refineries, etc.)
1A1a	F20	Power generation (public and auto; including cogeneration)
1A1b,c	F30	Other transformation sector (refineries, coke ovens, gas works, etc.)
1A4	F40	Residentials, Commercials and Other Sector (RCO)
1A3b	F51	Road transport
1A3c,d-ii,e	F54	Non-road land transport (Rail+Inland water+Pipeline+Non-specified)
1A3a	F57	Air transport (Domestic+International)
1A3d-i	F58	International Shipping Transport (marine bunkers)
1A2 or 2B,C	F60	Non-energy use and chemical feedstocks (CO2 only)
		Fossil fuel production and transmission
1B1a	F70	Coal production+(including CH4 recovery)
1B2a,c	F80	Oil production, transmission and handling
1B2b	F90	Gas production and transmission
-	F95	Fossil fuel fires
	B	Biofuel combustion
1A2	B10	Industry (excluding coke ovens, refineries, etc.)
1A1a	B20	Power generation (public and auto; including cogeneration
1B1b	B30	Charcoal production
1A4	B40	Residentials (excl. commercial and other sector (0 emission))
1A3b	B51	Road transport
	I	Industrial processes
2C1	I10	Iron and Steel
2C3,5	I20	Non-ferro metals
2B	I30	Chemicals
2A1	I40	Building materials
	I41	????
2D1	I50	Pulp and Paper
2D2	I60	Food
3	I70	Solvent use/Miscellaneous
1A3b	I80	Transport evaporation (NMVOC only) already in F51
2-other	I90	Miscellaneous industry
	L	Agricultur and biomass burning (land use)
4D	L10	Arable land (fertilizer use)
4C	L15	Rice cultivation
4A	L20	Animals (enteric fermentation)
4B	L30	Animal waste management (confined N2O; all CH4)
5A1	L41	BB-Deforestation
4E	L42	BB-Savanah burning
4F	L43	BB-Agricultural waste burning
5A2,3	L44	BB-Vegetation fires (temperate)
5B1	L45	BB-Deforestation post-burn effects (CO2, N2O)
4D	L50	Crop production (N)
4B	L60	Animal waste (deposited on soil - N2O)
4D	L71	Atmospheric Deposition
4D	L75	Leaching and Run-off
	W	Waste handling
6A1,2	W10	Landfills+(incl. CH4 recovery)
6D	W15	Humans/pets (p.m)
6B1,2	W20	Waste water treatment+(incl. CH4 recovery)
6B2	W30	Human waste water disposal
6C	W40	Waste incineration (non-energy)
6A3	W50	Miscellaneous waste handling (hazardous waste)

The following table indicates what is used in TM5:

EDGAR3.2	NO$_{x}$	NH3	SO2	CO	CH4	NMV
F10	IF		IF	IF	IF	IF
F20	IF		IF	IF	IF	IF
F30	IF		IF	IF	IF	IF
F40	IF		IF	IF	IF	IF
F51	IF		IF	IF	IF	IF
F54	IF		IF	IF	IF	IF
F57
F58	IF			IF	IF
F60
F70					IF
F80	IF		IF	IF	IF	IF
F90					IF	IF
F95
B10	B		B	B	B	B
B20	B		B	B	B	B
B30	B		B	B	B	B
B40	B		B	B	B	B
B51	B			B	B	B
I10	IF		IF	IF	IF	IF
I20			IF	IF
I30	IF		IF		IF	IF
I40
I41?	IF		IF
I50	IF		IF	IF		IF
I60						IF
I70						IF
I80
I90						IF
L10
L15					BA
L20					BA
L30					BA
L41				BB	BB
L42				BB	BB
L43				BB	BB
L44			T	T	T	T
L45
L50
L60
L71
L75
W10					BA
W15
W20					BA
W30					BA
W40	B		B	B	BB	B
W50

Here IF=Industry and Fossil fuel, T = Temperate fires, B=biofuel, BB=Biomass Burning and BA=biogenic-anthropogenic (CH4-only) are categories whose total is printed by TM5. What is I41 ?

SO4

Of the SO$_2$ anthropogenic emissions 2.5 % is emitted as sulfate, SO$_4$, see parameter frac_so4 in sources_sinks of TM3.

Attribution of anthropogenic VOC-emissions to CBM-4 compound types

The anthropogenic hydrocarbon emission database used a compound classification into 25 types of compounds. Each type has been translated by [Houweling et al., 1998] into the corresponding CBM-4 compound representation according to [Gery et al., 1989]. As a result, 78 percent of the anthropogenic emissions are represented as PAR, 13 percent as OLE and 9 percent as ETH, ALD2, MGLY and CH2O. In total, 87 percent of the EDGAR NMHC-carbon is represented by CBM-4 compounds.
Evaporation, biomass burning and other VOC emissions are treated separately: They are read in from feva.d, fbmb.d and foth.d. They have been constructed from GEIA by [Houweling et al., 1998] and are valid for 1990. They are calculated from resp. by program geia2tm3 called by script non_annual. Is this all? The following total emissions in 10$^{6}$ kg(C)/yr are imposed in TM3 in subroutine sources_sinks:

ipar,ieth,iole,iald2,imgly,ich2o Category	Paraffinic	Ethene	Olefinic	aldehydes	Methylglyoxal	Formaldehyde
Biomass burning	9865	4666	1801	758	149	682.5*0.4
Evaporation	54640	2546	1444	184	1842	507.5*0.4
Other	20730	5288	1931	703	295	750*0.4

0.4 is the (approximate) ratio of molecular masses of carbon (amu${_C}$=12) and formaldehyde (amu${_CH{_2}O}$=30.03).

Seasonal variation of NO$_{x}$ and SO$_{2}$ fossil fuel surface emissions

The seasonal variation of the above described NO$_{x}$ and SO$_{2}$ surface emissions from industrial processes (HIS) and Fossil fuel consumption (E_1) is imposed using the seasonal variation in the SO$_{2}$ and NO$_{x}$ emissions from the Global Emission Inventory Activity (GEIA, http://www.geiacenter.org), valid for 1985 ([Benkovitz et al., 1996]).
These data are most likely exactly identical to the files that can be downloaded from http://www.ortech.ca/cgeic/ (click on products). The following 4 files can there be found: High-nit.zip, High-sul.zip, Low-nitr.zip, and Low-sul.zip. After unzipping you will have: HIGH-NIT.PRN, HIGH-SUL.PRN, LOW-NITR.PRN, and LOW-SUL.PRN, which should correspond to our original files: G_NO_hi.1x1, G_SO2_hi.1x1, G_NOX_lo.1x1, G_SO2_lo.1x1. This data set is known as GEIA Version 1B, sulphur and nitrogen global inventory. The annual cycle is represented by four periods of 3 months.

In addition these NO$_{x}$ and SO$_{2}$ emissions are distributed over two atmospheric layers, below and above 100 m height, as in GEIA.

These distributions are read in from GSO2s.d and GNOXs.d which are calculated by readgeia. The redistribution for TM has been set up by [Jeuken, 2000].

There is also an annual mean inventory of NO$_{x}$ and SO$_{2}$ known as GEIA Version 1A, sulphur and nitrogen global inventory, which we don't use.

Of the SO$_2$ anthropogenic emissions 2.5 % is emitted as sulfate, SO$_4$, see parameter frac_so4 in sources_sinks of TM3.

Fossil fuel CO emissions

For CO emissions from fossil fuel combustion, industry, biofuel and ships E_1CO, HISCO, BF3CO and NRTCO there is also a high (northern) latitude seasonal CO distribution function
coseas = (/1.146,1.134,1.081,0.995,0.916,0.920,0.910,0.907,0.934,0.962,1.019,1.076/) It is applied from about 45 N northward:
if(j.ge.jm*3/4) cofac=coseas(mo)

This cycle has been added by Frank Dentener (what is it based on?). We are hoping something better will become available. Should not be something done for southern latitudes as well? (This will of course have a much smaller impact).

Biomass burning cycle

bmb_cycle.d contains monthly mean seasonal cycle of biomass burning emissions from [Hao and M.H.Liu, 1994] (or [Hao et al., 1991]) for 198 that is TM3 is applied to the EDGAR AGR, DEF and SAV categories. At locations where no data were available this distribution is set to a uniform value of 1/12 for every month. bmb_cycle.d is calculated with biomseas_tm3 from the orginal Hao files aftotal.txt, amtotal.txt, and astotal.txt. I do not think these are the original Hao data, but rather already the seasonal cycle (PvV). The prefix am stands for Central and South America (40 W - 110 W, 25 N - 55 S), af for Africa (20 W - 50 E, 35 N - 35 S) and as for Asia (60 E- 160 E, 35 N - 10 S).
The original Hao data can be ordered from NASA Langley
http://eosweb.larc.nasa.gov/HORDERBIN/HTML_Start.cgi (select biomass burning as Project and Parameter). Some information is availbale also there:
http://eosweb.larc.nasa.gov/HORDERBIN/HTML_Start.cgi (select biomass burning as Project and Parameter). http://eosweb.larc.nasa.gov/GUIDE/dataset_documents/bio_burn_5x5_hao_dataset.html The Hao data have only 5 x5 degree resolution. Here the units are 1000 tons/year or 1000 tons/month of dry biomass (CO2 or C ?) burned per 5 x 5 degree longitude x latitude box. It includes forests (deforestation) and savannah burning. The data are based on FAO statistics.