MASAGE NH3 inventory
From the abstract to Paulot et al (2014):
We use the adjoint of a global 3-D chemical transport model (GEOS-Chem) to optimize ammonia (NH3) emissions in the U.S., European Union, and China by inversion of 2005–2008 network data for NH4+ wet deposition ﬂuxes. Optimized emissions are derived on a 2° x 2.5° grid for individual months and years. Error characterization in the optimization includes model errors in precipitation. Annual optimized emissions are 2.8 Tg NH3−N a−1 for the contiguous U.S., 3.1 Tg NH3−N a-1 for the European Union, and 8.4 Tg NH3−N a−1 for China. Comparisons to previous inventories for the U.S. and European Union show consistency (∼ +/-15%) in annual totals but some large spatial and seasonal diﬀerences. We develop a new global bottom-up inventory of NH3 emissions (Magnitude And Seasonality of Agricultural Emissions model for NH3 (MASAGE_NH3)) to interpret the results of the adjoint optimization. MASAGE_NH3 provides information on the magnitude and seasonality of NH3 emissions from individual crop and livestock sources on a 0.5° x 0.5° grid. We ﬁnd that U.S. emissions peak in the spring in the Midwest due to corn fertilization and in the summer elsewhere due to manure. The seasonality of European emissions is more homogeneous with a well-deﬁned maximum in spring associated with manure and mineral fertilizer application. There is some evidence for the effect of European regulations of NH3 emissions, notably a large fall decrease in northern Europe. Emissions in China peak in summer because of the summertime application of fertilizer for double cropping.
--Bob Y. 16:33, 5 February 2015 (EST)
Reading MASAGE into GEOS-Chem
Emissions data file
We have implemented the MASAGE NH3 inventory into GEOS-Chem via the HEMCO emissions component. We use the emissions file stored in the HEMCO data path:
This file, which is in COARDS-compliant netCDF format, contains monthly mean NH3 emissions for the various agricultural sectors mentioned in Paulot et al (2014) (see below for totals). The emissions data are placed on the GEOS-Chem 2° x 2.5° horizontal grid, and the base year of emissions is 2006. Units of NH3 emissions are given as kg NH3 m-2 s-1.
The MASAGE NH3 data file is are contained in the HEMCO data directory tree. For detailed instructions on how to download this to your disk server, please see our Downloading the HEMCO data directories wiki post.
--Bob Y. 15:04, 13 February 2015 (EST)
If we convert the units of the data fields in file masage_nh3_2006.2x25.nc from kg NH3 m-2 s-1 to Gg N, we can directly compare with Table A1 of Paulot et al (2014). (The unit conversion involves multiplying each emissions field by the factor 14.0/17.0/1e6, and then summing over all 12 months). We then obtain the following global totals:
EMISSION TOTALS, compare to the "World" column of Table A1 of Paulot et al (2014) COTTON [kg N] 731.352 FRUITS_VEGETABLES [kg N] 820.576 GROUNDNUT [kg N] 80.8127 MAIZE [kg N] 1743.05 OIL_PALM [kg N] 112.432 OTHER_CEREALS [kg N] 122.545 OTHER_CEREALS_WINTER [kg N] 336.14 OTHER_CROPS [kg N] 987.441 PULSES [kg N] 44.6072 RAPESEED [kg N] 155.898 EARLY_RICE [kg N] 554.186 LATE_RICE [kg N] 323.54 SOYBEANS [kg N] 85.5041 SUGAR_CANE [kg N] 428.529 TEMPERATE_ROOTS [kg N] 459.644 TROPICAL_CEREALS [kg N] 92.0221 TROPICAL_ROOTS [kg N] 160.903 WHEAT [kg N] 423.498 WHEAT_WINTER [kg N] 1606.97 BEEF_CROP [kg N] 1998.11 BEEF_HOUSING [kg N] 5900.38 BEEF_PASTURE [kg N] 1672.35 BEEF_STORAGE [kg N] 549.203 BUFFALOES_EMISSION [kg N] 1525.14 DAIRY_CROP [kg N] 685.603 DAIRY_HOUSING [kg N] 1112.69 DAIRY_PASTURE [kg N] 668.778 DAIRY_STORAGE [kg N] 811.267 GOATS_EMISSION [kg N] 690.562 PORK_CROP [kg N] 775.642 PORK_HOUSING [kg N] 2055.58 PORK_PASTURE [kg N] 608.831 PORK_STORAGE [kg N] 680.933 POULTRY_CROP [kg N] 928.715 POULTRY_HOUSING [kg N] 2186.11 POULTRY_PASTURE [kg N] 546.004 POULTRY_STORAGE [kg N] 323.869 SHEEP_EMISSION [kg N] 967.568
--Bob Y. 16:32, 5 February 2015 (EST)
Editing the HEMCO configuration file
The MASAGE_NH3 inventory can be added to any GEOS-Chem simulation by inserting the following lines into the HEMCO configuration file (which is usually named HEMCO_Config.rc). Although the MASAGE_NH3 emissions are agricultural, we have lumped them in with anthropogenic emissions of NH3 (using HEMCO category #1). This follows established practice.
1. Add a switch for MASAGE to the EXTENSION SWITCHES section of the HEMCO configuration file:
--> MASAGE : true
2. Add the MASAGE emission information to the BASE EMISSION SECTION of the HEMCO configuration file. These lines can be placed anywhere between 'BEGIN SECTION BASE EMISSIONS' and 'END SECTION BASE EMISSIONS':
#============================================================================== # --- MASAGE_NH3 agricultural NH3 emissions --- #============================================================================== (((MASAGE 0 MASAGE_COTTON $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc COTTON 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_FRUITS_VEGETABLES $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc FRUITS_VEGETABLES 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_GROUNDNUT $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc GROUNDNUT 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_MAIZE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc MAIZE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_OIL_PALM $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc OIL_PALM 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_OTHER_CEREALS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc OTHER_CEREALS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_OTHER_CEREALS_WINTER $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc OTHER_CEREALS_WINTER 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_OTHER_CROPS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc OTHER_CROPS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_PULSES $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc PULSES 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_RAPESEED $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc RAPESEED 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_EARLY_RICE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc EARLY_RICE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_LATE_RICE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc LATE_RICE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_SOYBEANS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc SOYBEANS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_SUGAR_CANE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc SUGAR_CANE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_TEMPERATE_ROOTS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc TEMPERATE_ROOTS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_TROPICAL_CEREALS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc TROPICAL_CEREALS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_TROPICAL_ROOTS $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc TROPICAL_ROOTS 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_WHEAT $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc WHEAT 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_WHEAT_WINTER $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc WHEAT_WINTER 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_BEEF_CROP $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc BEEF_CROP 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_BEEF_HOUSING $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc BEEF_HOUSING 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_BEEF_PASTURE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc BEEF_PASTURE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_BEEF_STORAGE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc BEEF_STORAGE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_BUFFALOES_EMISSION $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc BUFFALOES_EMISSION 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_DAIRY_CROP $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc DAIRY_CROP 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_DAIRY_HOUSING $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc DAIRY_HOUSING 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_DAIRY_PASTURE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc DAIRY_PASTURE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_DAIRY_STORAGE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc DAIRY_STORAGE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_GOATS_EMISSION $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc GOATS_EMISSION 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_PORK_CROP $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc PORK_CROP 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_PORK_HOUSING $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc PORK_HOUSING 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_PORK_PASTURE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc PORK_PASTURE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_PORK_STORAGE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc PORK_STORAGE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_POULTRY_CROP $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc POULTRY_CROP 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_POULTRY_HOUSING $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc POULTRY_HOUSING 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_POULTRY_PASTURE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc POULTRY_PASTURE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_POULTRY_STORAGE $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc POULTRY_STORAGE 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 0 MASAGE_SHEEP_EMISSION $ROOT/MASAGE_NH3/v2015-02/masage_nh3_2006.2x25.nc SHEEP_EMISSION 2006/$MM/1/0 C xy kg/m2/s NH3 - 1 1 )))MASAGE
--Bob Y. 16:40, 5 February 2015 (EST)
Non-agricultural NH3 for use with MASAGE
This update is slated for inclusion in GEOS-Chem v11-02.
Amos Tai wrote:
Following Christoph's advice I ended up implementing a new category "6", changing all of MASAGE and agricultural emissions from EDGAR, NEI2011 and MIX to that category but placing MASAGE at a higher hierarchy than all of them. It still left me with the problem of some regional inventories (CAC, EMEP, STREETS) only have total NH3, so not to double-count them, I added "6" to their "1/2" categories, and placed these inventories at a higher hierarchy than (thus overwriting) MASAGE.
But that's not a good solution because ultimately we want MASAGE NH3 replacing the agricultural part in CAC, EMEP and STREETS. So what I [did] is to derive non-agricultural NH3 from these three inventories by using MASAGE to estimate their agricultural fractions in 2006 (then scaling up and down with annual factors with respect to each of these inventories). Then, if we turn on MASAGE, we can simply oblige HEMCO to read only the non-agricultural files when they read NH3 for CAC, EMEP and STREETS.After confirming with Katie [Travis], I realized NEI2011 NH3_SURFACE indeed contains agricultural emission, and thus I derived non-agricultural emissions for NEI2011 to be used with MASAGE NH3. Previous attempts to use MASAGE would have double-counted agricultural NH3 in the US.
- Paulot F., Jacob, D.J., Pinder R.W., Bash J.O., Travis, K., Henze D.K., Ammonia emissions in the United States, Europe, and China derived by high-resolution inversion of ammonium wet deposition data: Interpretation with a new agricultural emissions inventory (MASAGE_NH3), J. Geophys. Res., 119, 4,343-4,364, 2014. PDF