EDGAR v4.2 anthropogenic emissions
This page describes the EDGAR v4.2 anthropogenic emissions. This inventory was introduced into GEOS-Chem v10-01 via the HEMCO emissions component.
Contents
Overview
These updates were validated with the 1-month benchmark simulation v10-01h and approved on Approved 01 May 2015.
EDGAR 4.2 emissions of NOx, CO, SO2, and NH3 from 1970-2008 were prepared and implemented into GEOS-Chem. This inventory is the default global anthropogenic emission inventory in GEOS-Chem v10-01 and later versions. Since shipping and aircraft emissions are included in EDGAR 4.2, global emission inventories for these two sources should not be selected to avoid double counting errors.
Please refer to the documentation for more details: http://rain.ucis.dal.ca/upload/mli/Documentation_EDGARinGEOSChem_20140403.pdf.
- NOTE: The above document was prepared for GEOS-Chem v9-02. In GEOS-Chem v10-01 and higher versions, EDGAR v4.2 emissions are now read into GEOS-Chem via the HEMCO emissions component.
--Melissa Sulprizio (talk) 21:50, 16 June 2015 (UTC)
--Bob Yantosca (talk) 14:20, 14 March 2016 (UTC)
Sector information
The EDGAR v4.2 emissions inventory consists of several different files, segregated by species and sector.
Table of sectors
Each EDGAR emissions sector listed below is stored in a separate netCDF data file. This table was taken from the EDGAR v4.2 implementation document: http://rain.ucis.dal.ca/upload/mli/Documentation_EDGARinGEOSChem_20140403.pdf.
| NO sector | Description |
|---|---|
| EDGAR_v42_NOx_IPCC_1A1a | Energy industry (1A1a) |
| EDGAR_v42_NOx_IPCC_1A2 | Combustion in manufacturing industry (1A2) |
| EDGAR_v42_NOx_IPCC_1A3a_c_d_e | Non-road transportation (1A3a+c+d+e) |
| EDGAR_v42_NOx_IPCC_1A3b | Road transportation (1A3b) |
| EDGAR_v42_NOx_IPCC_1A4 | Residential (1A4) |
| EDGAR_v42_NOx_IPCC_1B2a_c_1A1b_c | Transformation, Oil production and refinering (1B1 + 1B2 + c + 1A1b + c) |
| EDGAR_v42_NOx_IPCC_2 | Industrial process and product use (2) |
| EDGAR_v42_NOx_IPCC_4B | Manure management (4B) |
| EDGAR_v42_NOx_IPCC_6A_6C | Solid waste disposal (6A+6C) |
| EDGAR_v42_NOx_IPCC_7A | Fossil Fuel Fires (7A) |
| CO sector | Description |
| EDGAR_v42_CO_IPCC_1A1a_6 | Energy industry and waste incinerator (1A1a + 6C) |
| EDGAR_v42_CO_IPCC_1A2 | Combustion in manufacturing industry (1A2) |
| EDGAR_v42_CO_IPCC_1A3a_c_d_e | Non-road transportation (1A3a+c+d+e) |
| EDGAR_v42_CO_IPCC_1A3b | Road transportation (1A3b) |
| EDGAR_v42_CO_IPCC_1A4 | Residential (1A4) |
| EDGAR_v42_CO_IPCC_1B2a_c_1A1b_c | Transformation, Oil production and refinering (1B1 + 1B2 + c + 1A1b + c) |
| EDGAR_v42_CO_IPCC_2A_2B_2D | Non-metallic paper chemical industry (2A+2B+2D) |
| EDGAR_v42_CO_IPCC_2C | Metal processes (2C) |
| EDGAR_v42_CO_IPCC_7A | Fossil Fuel Fires (7A) |
| SO2 sector | Description |
| EDGAR_v42_SO2_IPCC_1A1a_6C | Energy industry and waste incinerator (1A1a + 6C) |
| EDGAR_v42_SO2_IPCC_1A2 | Combustion in manufacturing industry (1A2) |
| EDGAR_v42_SO2_IPCC_1A3a_c_d_e | Non-road transportation (1A3a+c+d+e) |
| EDGAR_v42_SO2_IPCC_1A3b | Road transportation (1A3b) |
| EDGAR_v42_SO2_IPCC_1A4 | Residential (1A4) |
| EDGAR_v42_SO2_IPCC_1B1_1B2_1A1b_c | Transformation, Oil production and refinering (1B1 + 1B2 + c + 1A1b + c) |
| EDGAR_v42_SO2_IPCC_2B_2D | Paper chemical industry (2B + 2D) |
| EDGAR_v42_SO2_IPCC_2C | Metal processes (2C) |
| EDGAR_v42_SO2_IPCC_7A | Fossil Fuel Fires (7A) |
| NH3 sector | Description |
| EDGAR_v42_NH3_IPCC_1A1a_6 | Energy industry and waste incinerator (1A1a + 6C) |
| EDGAR_v42_NH3_IPCC_1A1b_c | Refineries and transformation (1A1b + 1A1c) |
| EDGAR_v42_NH3_IPCC_1A2 | Combustion in manufacturing industry (1A2) |
| EDGAR_v42_NH3_IPCC_1A3 | Transport non-road and road (1A3) |
| EDGAR_v42_NH3_IPCC_1A4 | Residential (1A4) |
| EDGAR_v42_NH3_IPCC_2A | Non-metallic mineral processes (2A) |
| EDGAR_v42_NH3_IPCC_2B | Chemical industry (2B) |
| EDGAR_v42_NH3_IPCC_4B | Manure management (4B) |
| EDGAR_v42_NH3_IPCC_4C_4D | Agricultural soils (4C + 4D) |
--Bob Yantosca (talk) 14:17, 14 March 2016 (UTC)
Rail transportation and construction
Maria Zatko wrote:
I’m planning to run a simulation using EDGAR v4.2 global anthropogenic emissions. I see that it is recommended that ICOADS/ARCTAS ship and AEIC aircraft emissions be used instead of EDGAR v4.2. non-road transportation emissions.
Using this approach, are rail emissions excluded from the simulation? If so, is this because rail emissions are much smaller compared to ship and aircraft emissions?
Also, are emissions from construction activity provided in EDGAR? I’m wondering if these emissions are lumped into another category, such as residential emissions. If construction emissions are not available in EDGAR, do you know if there are any other HEMCO emission inventories that contain this information?
Melissa Sulprizio replied:
The non-road emission category (1A3a_c_d_e) in EDGAR v4.2 lumps rail, ship, and aircraft emissions. The PARANOX ship plume model in GEOS-Chem explicitly requires separate ship NOx emissions and we didn’t see a way to extract the individual source types from EDGAR’s lumped non-road emissions. In addition, the EDGAR emissions have no vertical information, so it wasn’t clear how we should vertically distribute aircraft emissions. For those reasons, we excluded this category from GEOS-Chem and obtain ship emissions from ICOADS/ARCTAS and aircraft emissions from AEIC. That decision was also consistent with our typical approach of relying on specialized and regional inventories when possible. By excluding EDGAR’s non-road emissions, the rail emissions were excluded but, as I mentioned, that was only because we didn’t see a way to obtain just the rail emissions from EDGAR v4.2.
According to the EDGAR methodology website the 1A2 category includes “manufacturing industries and construction.” This emission category is included in GEOS-Chem.
--Bob Yantosca (talk) 22:42, 9 March 2016 (UTC)
Ship emissions
Ship emissions of NO, CO, and SO2 are not read in the HEMCO implementation of EDGAR v4.2. The EDGAR v4.2 ship emissions are lumped into a non-road transportation source category (1A3a_c_d_e) along with domestic aviation and rail transportation emissions. Therefore, it is not possible to isolate ship NOx emission for PARANOX.
Randall Martin wrote:
- I'm in favor of excluding the aircraft and ship emissions completely from the EDGAR implementation,and relying on ICOADS & AEIC for that information. That would be consistent with our prior approach of relying on specialized & regional inventories where available, and using EDGAR only where other information is unavailable.
--Melissa Sulprizio 16:25, 17 February 2015 (EST)
Applying scale factors for EDGAR emissions in HEMCO
Jaegun Jung wrote:
For the use of following annual scaling factor file,
###############################################################################
### BEGIN SECTION SCALE FACTORS
###############################################################################
# ScalID Name sourceFile sourceVar sourceTime C/R/E SrcDim SrcUnit Oper
#==============================================================================
# --- annual scale factors ---
#==============================================================================
1 TOTFUEL_THISYR $ROOT/AnnualScalar/v2014-07/AnnualScalar.geos.1x1.ncNOxscalar 1985-2010/1/1/0 C xy 1 1
This annual scaling file had [previously] been call ed with two years in the emission related subroutine in ./GeosCore before HEMCO was introduced. One argument was the base year and another was the target year. Instead of using direct value, a ratio between target year and base year was used. Following is an example of use of the file (in GEOS-Chem v9-02 and prior versions) before HEMCO was implemented.
IF (YEAR >= 1970 .AND. YEAR <= 2005) THEN
ELSEIF (YEAR < 1970) THEN
BASE_YEAR = 2000
CALL GET_ANNUAL_SCALAR( 72, BASE_YEAR, YEAR, SC_CO )
E_CO = E_CO * SC_CO(:,:)
ELSEIF (YEAR > 2005) THEN
BASE_YEAR = 2005
CALL GET_ANNUAL_SCALAR( 72, BASE_YEAR, YEAR, SC_CO )
E_CO = E_CO * SC_CO(:,:)
ENDIF
This part of code was from the old GeosCore/edgar_mod.F modified for EDGAR v4.1. At that time, EDGAR v4.1 covered from 1970 to 2005, if the model year is less than 1970, EDGAR v4.1 year 2000 data is used as a base year for extrapolation using the above annual scaling file. If it is newer than 2005, the year 2005 was used as base year, and the ratio is copied to SC_CO array. I cannot figure out how HEMCO works with the annual scaling factor. Does it simply take target year’s scaling factor values following the model year? In other words, is it absolute number instead of the ratio between base target year and base year?
Christoph Keller wrote:
The annual scale factors are still applied as previously: field TOTFUEL_THISYEAR are the scale factors relative to year 2000. To scale an emission inventory that has reference year 2000, you only need to apply this scale factor. To scale an inventory that has a reference year of 1985, you need to apply scale factor TOTFUEL_THISYEAR (which gives you the scaling relative to year 2000) and a normalization factor to take into account the different reference year (i.e. 1985 instead of 2000). This is what scale factors TOTFUEL_1985, TOTFUEL_2005, etc. do. Thus, applying scale factors TOTFUEL_THISYEAR and TOTFUEL_1985 to an emission inventory will give you the EDGAR annual scale factors relative to year 1985 (base year = 1985, scale year 1985-2010). What base year you use depends on the reference year of your emission inventory.
Jaegun Jung wrote:
I am still not clear how this HEMCO is working. For example, we have following HEMCO config file,
###############################################################################
### BEGIN SECTION BASE EMISSIONS
###############################################################################
# ExtNr Name sourceFile sourceVar sourceTime C/R/E SrcDim SrcUnit Species ScalIDs Cat Hier
(((EDGAR
0 EDGAR_NO_1A1a $ROOT/EDGARv42/v2015-02/NO/EDGAR_v42_NO_IPCC_1A1a.generic.01x01.nc emi_no 1970-2008/1/1/0 C xy kg/m2/s NO 1/40/25/30 1 2
)))EDGAR
#==============================================================================
# --- annual scale factors ---
#==============================================================================
1 TOTFUEL_THISYR $ROOT/AnnualScalar/v2014-07/AnnualScalar.geos.1x1.nc NOxscalar 1985-2010/1/1/0 C xy 1 1
40 TOTFUEL_1985_2008 $ROOT/AnnualScalar/v2014-07/AnnualScalar.geos.1x1.nc NOxscalar 1985-2008/1/1/0 C xy 1 -1
Let’s ignore ScalIDs 25 and 30 as they are not annual scale factors. The EDGAR_NO_1A1a has use two annual scaling factor files with different operator. Say the model year is 2010. The EDGAR_NO_1A1a covers up to 2008, but the TOTFUEL_THISYR’s base year is 2000. So, instead of using the latest 2008 as a base year, it will read 2000 annual emission year from the file as a base year. After that, it uses 2010 scaling values from the AnnualScalar.geos.1x1.nc to scale to 2010. Is it correct? And what happens to the ScalID 40? As 2010 is out of the range of 1985-2008, does it ignore the ScalIDs 40?
How about I run 2012. 2012 is not covered by either EDGAR_NO_1A1a or AnnualScalar.geos.1x1.nc. However, Cycling (C) is turned on. Do they use 2008 EDGAR_NO_1A1a as it is closest year? Or use 2000 values multiplied by 2010 scaling factor values in the annual scaling file?
This is another question related to base year. According to HEMCO user guide, I can set the different base year from the model year using,
Emission year
This is available from version v1.1.005. How can I check the version of HEMCO? My HEMCO configuration file does not have this.
When I use the scaling factor files, AnnualScalar.geos.1x1.nc, is it possible to set own base year?
Christoph Keller wrote:
This is indeed a little bit complicated! File AnnualScalar.geos.1x1.nc contains annual scale factors for all years between 1985 and 2010 relative to year 2000. So the scale factor for year 2006 in this file represents the scale factors of year 2006 relative to year 2000. If the reference year of your emission inventory is 2000, you only need to use scale factor TOTFUEL_THISYEAR since the reference years of your emission inventory and of field TOTFUEL_THISYEAR are the same. This is what was done for the older EDGAR emission inventories.
The new EDGAR emissions (EDGAR v42) contains emissions for every year between 1970 and 2008. For these years, we don’t want to scale emissions to another year because EDGAR already contains the wanted annual emissions. But for any year after 2008 we like to scale the emissions of 2008 (the last available year) using AnnualScalar.geos.1x1.nc relative to year 2008. And this is what scale factors 1 and 40 do: for example, for simulation year 2006 HEMCO will read EDGAR v42 emissions of year 2006, then scale them with AnnualScalar.geos.1x1.nc of year 2006 (-> ScalID 1), but also normalize it with AnnualScalar.geos.1x1.nc<tt> of year 2006 (<tt>-> ScalID 40). So for any year up to 2008, scale factors 1 and 40 cancel each other out. However, if you get to year 2009, ScalID 1 will use the scale factor of 2009 while ScalID 40 stays ‘frozen’ at year 2008. So for this year the EDGAR emissions will actually be scaled relative to year 2008 - exactly what we want.
I hope this explanation makes sense.
Regarding the version of HEMCO you are using, you can see this at the beginning of the HEMCO log file (it’s also written into the GEOS-Chem log-file, I think). My guess is that you are using HEMCO v1.1.005.
However, the "emission year" setting only works correctly in v1.1.006 and higher (I just updated the HEMCO wiki accordingly). Sorry about that. But if you are interested in using this feature I can provide you with the updated file. It’s only one HEMCO file that needs to be replaced with a newer one.
Note that this feature does not change the base year, though. Rather, it specifies the actual emission year used by HEMCO - independent of the simulation year. For example, you can make HEMCO use emissions of year 2000 for your entire GEOS-Chem simulation (no matter what GEOS-Chem simulation time period you use).
--Bob Yantosca (talk) 15:33, 4 March 2016 (UTC)
Previous issues that are now resolved
Correct bug in units of EDGAR v4.2 SO2 emissions
This update was validated with the 1-month benchmark simulation v11-01j and approved on 03 Dec 2016
Jaegun Jung wrote:
- EDGAR v4.2 SO2 – A README file and HEMCO configuration file say that the file has a unit of kgSO2 while in the emission netcdf file header shows kg(S). I calculated annual emission rate by creating HEMCO diagnostic output file. The value that I have is about 40 – 50 % from what is reported in,
- But, when I multiply by Mw(SO2)/Mw(S). It gets much closer within 10%. So I am pretty sure that the README and HEMCO configuration file are wrong.
To fix this bug, the SULFUR2SOX scale factor (scalID 62) should be added to the EDGAR v4.2 SO2 and SO4 entries in HEMCO_Config.rc. For example:
0 EDGAR_SO2_1A1a_6C $ROOT/EDGARv42/v2015-02/SO2/EDGAR_v42_SO2_IPCC_1A1a_6C.generic.01x01.nc emi_so2 1970-2008/1/1/0 C xy kg/m2/s SO2 11/42/31/62 1 2 0 EDGAR_SO4_1A1a_6C - - - - - - SO4 11/42/31/62/63 1 2 0 EDGAR_SO2_1A2 $ROOT/EDGARv42/v2015-02/SO2/EDGAR_v42_SO2_IPCC_1A2.generic.01x01.nc emi_so2 1970-2008/1/1/0 C xy kg/m2/s SO2 11/42/31/62 1 2 0 EDGAR_SO4_1A2 - - - - - - SO4 11/42/31/62/63 1 2 etc.
--Melissa Sulprizio (talk) 15:14, 5 April 2016 (UTC)
Prevent biofuel emissions from being double counted
This update was validated with the 1-month benchmark simulation v11-01j and approved on 03 Dec 2016
EDGAR v4.2 includes biofuel emissions in their 1A4 category. From the EDGAR website:
- EDGAR Factsheet - Energy: Combustion in 1A4
- Stationary combustion in residential, commercial and public services, agriculture, forestry, fisheries and other sectors.
However, in the default HEMCO_Config.rc that source is only being added to HEMCO Category 1 (i.e. anthropogenic emissions). Instead the EDGAR_*_1A4 entries should be changed from:
0 EDGAR_NO_1A4 $ROOT/EDGARv42/v2015-02/NO/EDGAR_v42_NO_IPCC_1A4.generic.01x01.nc emi_no 1970-2008/1/1/0 C xy kg/m2/s NO 1/40/25/30 1 2
to:
0 EDGAR_NO_1A4 $ROOT/EDGARv42/v2015-02/NO/EDGAR_v42_NO_IPCC_1A4.generic.01x01.nc emi_no 1970-2008/1/1/0 C xy kg/m2/s NO 1/40/25/30 1/2 2
to reflect that this source includes both anthropogenic and biofuel emissions. Otherwise, GEOS-Chem double counts biofuel emissions by including both EDGAR's 1A4 source and the Yevich and Logan biofuel emissions. This should only impact South America, Africa, Russia, and Australia because biofuel emissions over the other regions are zeroed to account for the fact that the existing regional inventories (BRAVO, CAC, NEI2011, EMEP, MIX) already include biofuel emissions.
--Melissa Sulprizio (talk) 20:32, 19 May 2016 (UTC)
References
--Melissa Sulprizio 10:11, 18 February 2015 (EST)
