GFED4 biomass burning emissions

From Geos-chem
Jump to: navigation, search

GFED4 biomass burning emissions are now included in GEOS-Chem v10-01 and higher versions.

Source code and data

GFED4 was implemented into GEOS-Chem v10-01 as an extension of the HEMCO emissions component. The source code files may be found in the GEOS-Chem source code folders:

File Description
HEMCO/Extensions/hcox_gfed_mod.F90 Module for computing GFED4 biomass emissions
HEMCO/Extensions/hcox_gfed_include_gfed4.H Emission factors for GFED4 species (inlined for computational efficiency)

We have created new GFED4 data files (in COARDS-compliant netCDF format) for use with HEMCO. These new data files are contained in the HEMCO data directory tree. For detailed instructions on how to download these data files to your disk server, please see our Downloading the HEMCO data directories wiki post.

--Bob Y. (talk) 15:53, 17 June 2015 (UTC)

Discussion and updates

Years 2015 and 2016 have been added

This update was included in GEOS-Chem 12.0.0.

As of 28 Jun 2018, we have added years 2015 and 2016 of GFED4.1 biomass burning data (in folder HEMCO/GFED4/v2015-10).

To use the years 2015 and 2016, make sure that your HEMCO_Config.rc file has the text in GREEN. (For clarity, we have omitted file names and other information from to better highlight what has changed.)

111 GFED_TEMP       ... DM_TEMP  1997-2016/1-12/01/0   ...
111 GFED_AGRI       ... DM_AGRI  1997-2016/1-12/01/0   ...
111 GFED_DEFO       ... DM_DEFO  1997-2016/1-12/01/0   ...
111 GFED_BORF       ... DM_BORF  1997-2016/1-12/01/0   ...
111 GFED_PEAT       ... DM_PEAT  1997-2016/1-12/01/0   ...
111 GFED_SAVA       ... DM_SAVA  1997-2016/1-12/01/0   ...

111 GFED_FRAC_DAY   ... GFED_FRACDAY 2003-2016/1-12/1-31/0 ...

111 GFED_FRAC_3HOUR ... GFED_FRAC3HR 2003-2016/1-12/1/0-23  ...

Starting with GEOS-Chem 12.0.0, all GEOS-Chem 1-month and 1-year benchmarks (which now use met field year 2016) will use GFED emissions for 2016, instead of cycling 2014 (which was previously the last year of available data).

--Bob Yantosca (talk) 16:36, 28 June 2018 (UTC)

Update to GFED 4.1

This update was validated with 1-month benchmark simulation v11-01d and 1-year benchmark simulation v11-01d-Run1. This version was approved on 12 Dec 2015.

Prasad Kasibhatla wrote:

Please see attached. This latest version GFED 4.1s is being released July 3, 2015.
The monthly GFED files that go into GEOS-Chem have to be updated (though updates are very minor). In addition, from 2003 onwards there is now daily and diurnal info available in the dataset - this should be added to GEOS-Chem.

--Melissa Sulprizio (talk) 13:25, 13 July 2015 (UTC)

Discussion following 1-month benchmark v10-01h

GFED4 biomass burning emissions were validated in GEOS-Chem v10-01h. The following discussion ensued.

Prasad Kasibhatla wrote:

I have been looking at the CO biomass emission ratio plot in the 10-01h benchmark - there are a couple of regions where things look a bit different than what I expect from the emissions I compute offline at native resolution. Would it be possible for you to send me plots of absolute CO emissions from biomass burning from v10-01h and v10-01f_UCX that went into the ratio plot. And a difference plot?

Melissa Sulprizio wrote:

Please see the attached image for maps of CO biomass emissions from v10-01f_UCX (upper left) and v10-01h (upper right). The absolute difference (lower left) and percent difference (lower right) maps are also included.
V10-01 CO biomass emission diffs.gif

Prasad Kasibhatla wrote:

I computed CO emissions offline for GFED3 for July 2011 and GFED4 for July 2013 (corresponding to what was used in the v10-01f_UCX and v10-01h benchmark runs) and reproduced the emissions plot you sent.
Don't worry about the labels on my plot - my panels correspond to your panels. As you can see your patterns match patterns I compute - but absolute values of emissions that you are getting from the model are higher than what I calculate offline. I have not been able to figure out if I am making a mistake in my offline calculation or if there is a problem in the geoschem implementation.
Obviously, we need to figure this out. Perhaps Christoph and I can focus on one or two pixels and resolve what is going on?
GFED4 CO emissions from prasad.png
For both the GFED3 and GFED4 offline plots, I accumulated dry matter burned in various categories into the generic 1x1 grid, computed emissions in molecules/cm2 on the 1x1 grid using category-specific emission factors, and then regridded to the 4x5 grid using ctm_regridh with the /per_unit_area option.

Christoph Keller wrote:

Thanks, this may explain it. In v10-01e and newer, dry matter burned and vegetation mask are first regridded onto the simulation grid, and emissions are computed on that grid.
I looked into this for version v10-01e, and the changes in GFED-3 emissions (compared to pre-HEMCO versions) are attributable to the regridding order. In v10-01h, we didn’t change anything in GFED-3, and GFED-4 uses the same code as GFED-3 except for the updated dry matter emissions fields.

--Melissa Sulprizio 13:38, 27 March 2015 (EDT)


  1. GFED website:
  2. Akagi, S. K., R. J. Yokelson, C. Wiedinmyer, M. J. Alvarado, J. S. Reid, T. Karl, J. D. Crounse, and P. O. Wennberg, Emission factors for open and domestic biomass burning for use in atmospheric models, Atmos. Chem. Phys., 11, 4039-4072, 2011. Article
  3. Andreae, M. O., P. Merlet, Emission of trace gases and aerosols from biomass burning, Glob. Biogeochem. Cycles, 15, Issue 4, pages 955-966, December 2001. Article
  4. Giglio, L., J. T. Randerson, and G. R. van der Werf, Analysis of daily, monthly, and annual burned area using the fourth-generation global fire emissions database (GFED4)", J. Geophys. Res, Biogeosciences., 118, Issue 1, 317-328, March 2013. Article
  5. Randerson, J. T., Y. Chen1, G. R. van der Werf, B. M. Rogers, and D. C. Morton, Global burned area and biomass burning emissions from small fires, J. Geophys. Res, Biogeosciences, 117, Issue G4, December 2012. Article
  6. van der Werf, G. R, J. T. Randerson, L. Giglio, G. J. Collatz, M. Mu, P. S. Kasibhatla, D. C. Morton, R. S. DeFries, Y. Jin, and T. T. van Leeuwen, Global fire emissions and the contribution of deforestation, savanna, forest, agricultural, and peat fires (1997-2009) Atmos. Chem. Phys., 10, 11707-11735, 2010. Article

--Bob Y. 10:56, 17 March 2015 (EDT)

Previous issues that are now resolved

Update GFED4 emission factors

This update was included in v11-02f (approved 17 May 2018).

Katie Travis wrote:

I was looking through the GFED4 emission factors in v11, and realized they appear out of date from is currently on the website ( This table appears to have been posted last June.
I put the table online into the HEMCO format using Bob’s code. I also updated the burned area categories in the code to match GFED4 to avoid future confusion. Since the GFED3 burned area categories are different as you say, I was not sure how to deal with a few things like Hg0 etc that are needed for the model but not standard for GFED. I cc’d Emily since she also made some changes for her PAN paper that I am not sure are properly represented anymore.

Prasad Kasibhatla wrote:

When I used GFED3 emission factors with GFED4 burned area files, I did something pretty simple. Here is the mapping as I remember:
  GFED4 Burned Area Category                GFED3 Emission Factor Category

  SAVA                                      SAV
  PEAT                                      PET
  BORF                                      FOR
  AGRI                                      AGW
  TEMP                                      FOR
  DEFO                                      DEF
Note in GFED4 implementation of GFED3 emission factors, there is no need of Humid Tropical Forest mask. In GFED3 implementation of GFED3 emission factors, GFED3 burned area category DEF used DEF emission factor for Humid Tropical Forest, and WDL emission factor outside Humid Tropical Forest.
So, now GFED4 emission factors corresponding to GFED4 burned area categories should be used. But the GFED3 code should not be changed. But you seem to gave changed the GFED3 part of the code?
In fact, we should now just get rid of the GFED3 option.

Katie Travis wrote:

I translated Hg0, (and HAC and GLYX, which are not in GFED4), from GFED3 to GFED4 using Prasad’s translation. I stripped out any calls to GFED3 in the code.

--Bob Yantosca (talk) 15:55, 17 May 2018 (UTC)

Fixed unit conversion error on 18 Jun 2015

On 18 Jun 2015, we fixed a unit conversion error in the GFED4 data files that were contained in the HEMCO data directory GFED4/v2015-06/:

Please note that this correction is now obsolete. GFED4/v2015-06 is superceded by the new GFED4 release located at GFED4/v2015-10.

--Lizzie Lundgren (talk) 16:55, 13 December 2017 (UTC)

Zhen Qu wrote:

The biomass burning emission seems to be smaller by about 3 magnitude with the conversion error.

Christoph Keller wrote:

Yes, this sounds about right. The GFED-4 netCDF files that we use in GEOS-Chem are directly generated from the original GFED-4 data. These data changed units between the provisional and the official GFED 4 release (gDM/m2/month to kgDM/m2/month). I didn’t realize that and the GFED-4 files were hence a factor of 1000 too low.

The GFED4 data files with the correct units were uploaded to the HEMCO data directory tree at 4PM EDT on 18 Jun 2015. If you obtained GFED4 data prior to 18 Jun 2015, we ask you to re-download this data.

To download the data, you can use the hemco_data_download installer, as described on our Downloading the HEMCO data directories wiki post. Or if you would like to get these files directly, you can use this FTP command:

ftp "*.nc"

We apologize for the inconvenience.

--Bob Y. (talk) 20:11, 18 June 2015 (UTC)

Currently unresolved issues

None at this time.

--Bob Y. 14:27, 27 March 2015 (EDT)