EDGAR v3 anthropogenic emissions
This page describes the EDGAR v3 anthropogenic emissions, which were used in GEOS-Chem versions prior to v10-01. Much of the information on this page is now obsolete, but we will leave it here for future reference.
EDGAR v3 anthropogenic emissions were used in GEOS-Chem versions through GEOS-Chem v9-02. EDGAR v4.2 emissions (read via the HEMCO emissions component)are now the default in GEOS-Chem v10-01 and newer versions.
The EDGAR emissions of NOx, CO, and SOx are the standard emissions. However, the EDGAR emissions are typically overwritten by data from various regional emissions inventories listed on our Anthropogenic emissions wiki page.
The EDGAR inventory also includes emissions from ship exhaust.
--Bob Y. 12:27, 20 December 2012 (EST)
For more information about the EDGAR ship emissions data files, please see the following README document: GEOS_1x1/EDGAR_200607/README
The source code routines that read the EDGAR ship emissions data from disk are:
- SEASCL_EDGAR_SHIP_SO2 (GeosCore/edgar_mod.F)
- GET_EDGAR_SHIP_SO2 (GeosCore/edgar_mod.f)
The EDGAR ship SO2 emissions are reported as per year. Routine SEASCL_EDGAR_SHIP_SO2 applies monthly scale factors (which are computed as the ratio of monthly/total ship SO2 emissions from the Corbett et al inventory to the annual EDGAR ship SO2 emissions. This is required to impose aseasonality onto the EDGAR ship SO2 emissions.
--Bob Y. 12:27, 20 December 2012 (EST)
- Olivier, J.G.J. and J.J.M. Berdowski, 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., 2001
- A related document can be found here: EDGAR description (pdf)
- Aaron van Donkelaar has created a PDF document that shows some of the effects of using the EDGAR emissions.
--Bob Y. 12:31, 20 December 2012 (EST)
Previous issues that are now resolved
Potential bug in EDGAR ship emissions code
This update was tested in the 1-month benchmark simulation v9-01-02s and approved on 5 Nov 2011.
Michael Long discovered this potential bug in the EDGAR emissions code.
The current GEOS-Chem default configuration uses the ICOADS ship emissions instead of the EDGAR ship emissions. In routine READ_EMISSIONS_MENU in input_mod.f, we have a safety check...we turn off the logical switch LEDGARSHIP if switch LICOADS is turned on:
!================================================================= ! Check global ship emissions options !================================================================= ! Add an ship emissions options (cklee, 6/30/09) ! Replace with ICOADS ship emissions ! User can still use regional EMEP, and ARCTAS-SO2 if he wants (phs) IF ( LICOADSSHIP ) THEN LEDGARSHIP = .FALSE. ! ICOADS precedes EDGAR LSHIPSO2 = .FALSE. ! SO2-ICOADS precedes SO2-Corbett ENDIF
In edgar_mod.F, the EDGAR ship emissions arrays are not allocated if the switch LEDGARSHIP == .FALSE.:
IF ( LEDGARSHIP ) THEN ALLOCATE( EDGAR_SO2_SHIP( IIPAR, JJPAR ), STAT=AS ) IF ( AS /= 0 ) CALL ALLOC_ERR( 'EDGAR_SO2_SHIP' ) EDGAR_SO2_SHIP = 0d0 ALLOCATE( EDGAR_NOx_SHIP( IIPAR, JJPAR ), STAT=AS ) IF ( AS /= 0 ) CALL ALLOC_ERR( 'EDGAR_NOx_SHIP' ) EDGAR_NOx_SHIP = 0d0 ALLOCATE( EDGAR_CO_SHIP( IIPAR, JJPAR ), STAT=AS ) IF ( AS /= 0 ) CALL ALLOC_ERR( 'EDGAR_CO_SHIP' ) EDGAR_CO_SHIP = 0d0 ENDIF
However, we still pass EDGAR_CO_SHIP to routine COMPUTE_EDGAR_CO even if it is not allocated:
IF ( LEDGARCO ) THEN CALL COMPUTE_EDGAR_CO( YEAR, EDGAR_CO, EDGAR_CO_SHIP ) ENDIF
Here, EDGAR_CO_SHIP is an output variable, which is supposed to return the EDGAR ship emissions at the current model resolution. This may produce undefined behavior. This may trigger an error when compiling with the check-array-bounds option (BOUNDS=yes).
Some of the more forgiving compilers may "fix" this on-the-fly and thus, you might not see any errors. However, for the sake of good programming practice, this should be fixed.
--Bob Y. 12:30, 20 December 2012 (EST)
Typos in edgar_mod.f
NOTE: This issue was resolved in GEOS-Chem v9-01-01.
In routines READ_C2H2 and READ_C2H4 of edgar_mod.f, the data directory was wrongly listed as EDGAR200607 instead of EDGAR_200607. We have now fixed these typos in v9-01-01.
--Bob Y. 12:35, 20 December 2012 (EST)