Emission Guide
Emission Guide
Emission Guide
1 Emissions Guide
Table of Contents
1.1 Introduction ................................................................................................................... 2
1.2 The WRF-Chem emissions utilties overview ............................................................... 2
1.3 Placement of chemical-emission input data files .......................................................... 2
2.1 Preparation of anthropogenic emissions for use with WRF-Chem .............................. 5
2.2 Using the global-emissions data set .............................................................................. 5
2.2.1 Seting-up prep_chem_sources ............................................................................... 6
2.2.2 Compiling the prep_chem_sources code ............................................................... 7
2.2.3 Running prep_chem_sources utility with WRF-Chem .......................................... 7
2.3 The 4-km resolution NEI data set (2005 NEI emissions data for USA only)............... 9
2.4 Anthropogenic-emissions construction methodology for WRF-Chem ........................ 9
2.4.1 Construction of an anthropogenic-emissions-inventory conversion table ........... 10
2.4.2 Additional details for running emiss_v03.F with the NEI-05 anthropogenic-
emissions data set.......................................................................................................... 14
2.5 The anthro-emiss utiltiy .............................................................................................. 17
3.1 Preparation of biogenic emissions for use with WRF-Chem...................................... 19
3.2 No biogenic emissions ................................................................................................ 19
3.3 Guenther biogenic emissions ...................................................................................... 19
3.4 BEIS 3.14 biogenic emissions .................................................................................... 19
3.5 MEGAN biogenic emissions ...................................................................................... 21
3.5.1 Compiling MEGAN preprocessor ....................................................................... 22
3.5.2 Preprocessing MEGAN data files ........................................................................ 22
3.5.3 Running WRF-Chem with MEGAN ................................................................... 23
4.1 Preparation of biomass burning emissions ................................................................. 25
4.2 Using prep_chem_sources for biomass burning emissions ........................................ 25
4.3 Using FINN biomass burning emissions .................................................................... 27
5.1 Preparation of volcanic emissions for use with WRF-Chem ...................................... 29
5.2 Volcanic ash emissions ............................................................................................... 29
5.3 Volcanic SO2 degassing emissions ............................................................................. 30
6.1 Preparation of tracer emissions for use with WRF-Chem .......................................... 31
6.2 The CO2 tracer option ................................................................................................. 32
6.3 The Greenhouse Gas tracer option .............................................................................. 32
7.1 Boundary conditions with WRF-Chem ...................................................................... 33
7.2 The wrfchembc utility ................................................................................................. 33
7.3 The mozbc utility ........................................................................................................ 34
7.4 Including an upper boundary bounary condition for chemical species ...................... 34
8.1 Converting binary intermediate file to a WRF-Chem data file................................... 36
8.2 Binary data file format ................................................................................................ 37
8.3 Building the WRF-Chemistry emissions conversion code ......................................... 38
8.4 Namelist settings for running convert_emiss .............................................................. 38
9.1 Summary ..................................................................................................................... 41
9.2 WRF-Chem and utility program related publications ................................................. 41
1
WRF-Chem Emissions Guide
WRF-Chem Overview
Table of Contents
1.1 Introduction ................................................................................................................... 2
1.2 The WRF-Chem emissions utilties overview ............................................................... 2
1.1 Introduction
The following figure shows the flowchart for the WRF-Chem modeling system.
In the WRF-Chem model there are several ways in which gridded chemicaldata can be
introduced into the simulation. The static data (e.g., data based upon land useage, or
vegetation) is introduced when running the real data initilization. Data that varies
temporally (e.g., anthropogenic emissions) is read into the simulation when running the
WRF model.
At this time there is no single tool that will construct the chemistry emissions
input data sets for any domain and any chemical mechanism that is selected. This places
the requirement upon you, the WRF-Chem model user, to construct the emissions
data set for your particular domain and desired chemistry. However, several
programs and data sets are provided by the WRF-Chem user community that can help to
create the input data sets for your domain and your choice of chemical mechanism. These
programs an their use are described in the following chapters. Note that it would be best
to know the preferred domain location and chemistry options that will be used in the
simulation before starting. In this way the choices regarding which utilities to be used
will become clear.
2
3
they were constructed. It should not make any difference whether you use a link or copy
the files to the WRFV3/test/em_real directory. The main point to take away is that, once
the emission file construction is completed, the WRF-Chem code expects the emissions
data files to be located in the WRFV3/test/em_real directory.
4
Chapter 2: Generation of Anthropogenic Emissions
Data Files
Table of Contents
2.1 Preparation of anthropogenic emissions for use with WRF-Chem .............................. 5
2.2 Using the global-emissions data set .............................................................................. 5
2.2.1 Seting-up prep_chem_sources ............................................................................... 6
2.2.2 Compiling the prep_chem_sources code ............................................................... 7
2.2.3 Running prep_chem_sources utility with WRF-Chem .......................................... 7
2.3 The 4-km resolution NEI data set (2005 NEI emissions data for USA only)............... 9
2.4 Anthropogenic-emissions construction methodology for WRF-Chem ........................ 9
2.4.1 Construction of an anthropogenic-emissions-inventory conversion table ........... 10
2.4.2 Additional details for running emiss_v03.F with the NEI-05 anthropogenic-
emissions data set.......................................................................................................... 14
2.5 The anthro-emiss utiltiy .............................................................................................. 17
The use of a global-emissions data set with WRF-Chem is probably the most
common choice of options. The global emissions data comes from the REanalysis of the
TROpospheric (RETRO) chemical composition over the past 40 years
(http://retro.enes.org) and Emission Database for Global Atmospheric Research
(EDGAR) (http://www.mnp.nl/edgar/introduction). Both RETRO and EDGAR provide
global annual emissions for several greenhouse gases (e.g., CO2, CH4 and N2O) as well as
some precursor gases on a .5x.5 degree (RETRO) or a 1x1 degree (EDGAR) grid.
5
The intermediate binary data file can then be converted into a WRF input data file using
the convert utility program discussed in chapter 9.
In this section the set-up and running of the prep_chem_sources utiltity will be
discussed. The code can be obtained from the WRF-Chem ftp site
(ftp://aftp.fsl.noaa.gov/divisions/taq/global_emissions). After untaring the code it can be
seen that the code exists in several subdirectories. These subdirectories are:
There are several libraries that are not normally needed to compile WRF that are needed
to compile prep_chem_sources. These libraries are:
JPEG distribution release 6b(libjpeg.a). You may download the software from
http://www.hdfgroup.org/release4/obtain.html
ZLIB 1.2.1(libz.a) or later distribution. You may download the software from
the http://www.gzip.org/ site.
HDF5 1.8 (libmfhdf.a, libdf.a) or later distribution. You may download the
software from http://www.hdfgroup.org/downloads/index.html
NetCDF 4.0.1 (libnetcdf.a). You may download the software from
http://www.unidata.ucar.edu/downloads/netcdf/netcdf-4_0_1/index.jsp
When compiling PREP-CHEM-SRC codes on a Linux system make sure the netCDF
and HDF library has been installed using the same FORTRAN and C compilers that are
used for WRF. For example, you will probably need to configure the HDF library build
using command lines:
./configure --prefix=path_hdf5 \
--with-jpeg=path_jpeg \
--with-zlib=path_zlib \
--disable-netcdf
as these are what is often used with WRF. And then to compile the HDF library one uses
the make command:
6
2.2.2 Compiling the prep_chem_sources code
cd bin/build
and verify the compiler and library path settings in the include.opt.<your compiler> make
option file. The user should set similar library paths and compile options (e.g., byte
swapping) as used for the WRF-Chem code compilation otherwise the code might not
compile or the binary intermediate output will be unreadable by the WRF-Chem convert
program. Then, once the make option include file is configured for your computer
system, the make command is issued with the compiler option (e.g., PGI) and the
chemistry option (ie., RADM_WRF_FIM) listed in the command line
And with this command the code should compile and the prep_chem_sources.exe file
being placed in the prep_chem_source/bin directory.
To run the prep_chem_sources executable one must first edit the name listinput
fields. Most of the settings are like the previous versions with the exception of the map
projection settings. But fortunately the map projection settings correspond to the map
projection settings used in the WPS input file namelist.wps. That is:
NINEST=i_parent_start,
NJNEST=j_parent_start,
POLELAT=ref_lat,
POLELON=ref_lon,
STDLAT1=truelat1,
STDLAT2=truelat2.
The prep_chem_sources settings CENTLAT and CENTLON are not used for WRF, but
are most often the latitude and longitude of the central grid point for your domain.
After setting the prep_chem_sources name list one can run the PREP-CHEM-
SRC executable (e.g., prep_chem_sources_RADM_WRF_FIM.exe) for each day to
produce the intermediate data files. These files are the anthropogenic- and biogenic-
emissions file (ending with -ab.bin), the biomass burning-emissions file (ending with -
bb.bin), and the GOCART background data (ending with -gocartBG.bin). While a
typical run configuration will produce all of these emissions files, the focus in this section
will be on just the anthropogenic emissions. Once the prep_chem_sources executable has
been run the binary intermediate files it produces for WRF-Chem will need to be
converted to WRF input data files. The conversion code is discussed in more detail in
7
chapter 8 of this Guide so the running of the code for the anthropogenic emissions will
be outlined next.
For each simulation that you are going to use the global emissions data, you need
to run the convert_emissions program. This program will also need a wrfinput data file
for the desired date as well. So to begin, one can run the real.exe with chem_opt=0 and
start and end time set to your year, month, day, and start hour generate a wrfinput_d01
file. Then the output from prep_chem_sources is linked to the WRF running directory
(e.g., WRFV3/test/em_real) with commands similar to the following:
And then before one can run the convert emissions program convert_emiss
(convert_emiss.exe) the namelist.input settings for the chemistry needs to be turned back
on (e.g., chem_opt=301). Be sure to double check your other namelist settings as well or
you might not get the desired data files.
&time_control
io_form_auxinput5 =2
io_form_auxinput6 =2
io_form_auxinput7 =2
io_form_auxinput8 =2
io_form_auxinput12 = 2,
io_form_auxinput13 = 0,
auxinput6_inname = 'wrfbiochemi_d01',
auxinput7_inname = 'wrffirechemi_d<domain>',
auxinput8_inname = 'wrfchemi_gocart_bg_d<domain>',
auxinput12_inname = 'wrf_chem_input',
auxinput13_inname = 'wrfchemv_d<domain>'
auxinput5_interval_m = 1440,1440
auxinput7_interval_m = 1440,1440
auxinput8_interval_m = 1440,1440
auxinput13_interval_m = 1440,1440
frames_per_auxinput6 = 1,1
frames_per_auxinput7 = 1,1
frames_per_auxinput8 = 1,1
frames_per_auxinput13 = 1,1
/
&chem
kemit = 1,
chem_opt = 301, 2,
io_style_emissions = 2,
emiss_inpt_opt = 1, 1,
8
emiss_opt = 5, 5,
emiss_opt_vol = 0, 1,
biomass_burn_opt = 1, 1,
plumerisefire_frq = 120, 120
And when complete and the data files are verified you can rename output to include the
date if so necessary. Otherwise, the global anthropogenic emissions file is ready to be
used in a WRF-Chem simulation.
2.3 The 4-km resolution NEI data set (2005 NEI emissions data for USA only)
The methodology for constructing an emissions data set using NEI data is the same as
that when constructing an emissions data set using your own anthropogenic-emissions
data set. For simplicity, this section will walk through the steps assuming one has the
NEI emissions. The steps are as follows:
Obtain the raw anthropogenic-emissions data. This data could come from a
variety of data sources and be on multiple map projections and/or domains. The
main point is that the data is already mapped to a cartesian grid.
o A 4-km emissions data set (area) and point source is available for the U.S.
from 2005 (ftp://aftp.fsl.noaa.gov/divisions/taq/emissions_data_2005) and
2011 (ftp://aftp.fsl.noaa.gov/divisions/taq/emissions_data_2011). Use of
this data is recommended when the simulation domain has a horizontal
grid spacing of 12 km or greater.
Specify, or make a table listing that relates raw emissions to the speciation of
the desired chemical mechanism and PM mechanism (see following section)
9
o The provided routines (emiss_v03.F) assume that the RADM2 chemical
mechanism and MADE/SORGAM modal aerosol models are being used
in the simulation.
In this section some of the listed steps are expanded in greater detail. To generate
the anthropogenic emissions data file, begin with a list of known chemical species that
are emitted in the domain of interest. These species may need to be translated into a list
of chemical species that are used by your particular photochemical and aerosol
mechanisms within the WRF-Chemistry model. If you are uncertain about the names and
units of the emissions data, the registry.chem file in the WRFV3/Registry subdirectory
contains the names and dimensions of the chemical species used within the WRF-Chem
model.
The translation from raw to WRF-Chem species emissions will often result in
either lumping several emitted chemical species into one simulated species, or the
partitioning of one emitted species into fractions of several simulated species. As an
example, the following emission assignment table (Table 2.1) translates the raw NEI05
based emission species into the WRF-Chem RADM2 species. The columns contain the
following information:
10
names of the emitted species in the raw data derived from the EPA NEI05
inventory, VOC speciation is that used in the SAPRC-99 chemical mechanism.
names of the emitted species used in the WRF-Chemistry model, Variable names
(e.g. e_co) must match the WRF-Chem Registry names of the emission variables.
the fractional amount of the raw emitted species assigned to the model emission
name,
the molecular weight (used as a switch in emiss_v03.F applies only to primary
NOx, SO2, CO and NH3 emissions),
the technical name of the raw emitted species
Table 2.1. Conversion table within emiss_v03.F used to produce input-emissions data
for a WRF-chemistry simulation. This table lists the raw emission name, the emissions
field name used in the WRF model, the weight factor applied to the chemical field, the
molecular weight of the species (NOx, SO2, CO and NH3 only) and the full species name.
The fields are then converted to an emissions speciation suitable for use with the RADM2
chemical mechanism (+MADE/SORGAM aerosol module).
11
HC22 e_ald 1.00 00 Glyoxal
HC23 e_ald 1.00 00 Methylglyoxal
HC24 e_ald 1.00 00 Biacetyl
HC25 e_csl 1.00 00 Phenols
HC26 e_csl 1.00 00 Cresols
HC27 e_ald 0.50 00 Methacrolein
HC27 e_olt 0.50 00 Methacrolein
HC28 e_ket 0.50 00 Methylvinyl ketone
HC28 e_olt 0.50 00 Methylvinyl ketone
HC29 e_ket 1.00 00 IPRD SAPRC-99 species (other ketones)
HC31 e_olt 1.00 00 Propylene
HC32 e_hc3 0.40 00 Acetylene
HC33 e_tol 0.29 00 Benzene
HC34 e_hc3 1.11 00 Butanes
HC35 e_hc5 0.97 00 Pentanes
HC36 e_tol 1.00 00 Toluene
HC37 e_xyl 1.00 00 Xylenes
HC38 e_hc3 0.57 00 Propane
HC39 e_oli 1.00 00 Dienes
HC40 e_olt 1.00 00 Styrenes
HC41 e_ora2 1.00 00 Organic Acids
PM01 e_pm25i 0.20 01 Unspeciated primary PM2.5 - nuclei mode
PM01 e_pm25j 0.80 01 Unspeciated primary PM2.5 - accumulation
mode
PM02 e_so4i 0.20 01 Sulfate PM2.5 - nuclei mode
PM02 e_so4j 0.80 01 Sulfate PM2.5 - accumulation mode
PM03 e_no3i 0.20 01 Nitrate PM2.5 - nuclei mode
PM03 e_noj 0.80 01 Nitrate PM2.5 - accumulation mode
PM04 e_orgi 0.20 01 Organic PM2.5 - nuclei mode
PM04 e_orgj 0.80 01 Organic PM2.5 - accumulation mode
PM05 e_eci 0.20 01 Elemental Carbon PM2.5 - nuclei mode
PM05 e_ecj 0.80 01 Elemental Carbon PM2.5 - accumulation mode
PM10- e_pm10 1.00 01 Unspeciated Primary PM10
PRI
The next step is to construct a program that reads the raw anthropogenic-
emissions data, converts each chemical species to those used by the WRF-Chem model
following the information from your particular conversion table and finally maps it onto
the 3-dimensional simulation domain. Therefore, within this step any plume rise, or
above surface anthropogenic emission placement needs to be specified. Particular
attention to geospatial details, such as the WRF-Chem domain grid locations, and the
elevation of model vertical levels relative to the raw emissions data set must be
considered.
12
05 and NEI-2011 U.S. anthropogenic-emissions inventories - emiss_v03.F. While your
application may not use the raw emissions data for your simulation, it is provided as an
example of the adopted methodology. The product of the program is a binary data file
containing three-dimensional emissions data, output at each hour that is mapped to a
specified simulation domain. The data format in the provided program is provided in
Table 2.2.
For spatial partitioning, the emiss_v03.F program implicitly assumes the WRF-
Chem grid has a horizontal grid spacing larger than 4 km, and simple grid dumping from
the raw 4-km domain into the specified WRF-Chem domain is appropriate. Currently
no plume rise calculations directly couple WRF dynamics to anthropogenic point
emissions. The emiss_v03.F routine includes some plume rise from the Briggs
formulation due to momentum lift from direct injection, and a specified horizontal wind
climatology. Emissions within nested domains are also handled within emiss_v03.F by
specifying mother domain map parameters, the nested domain grid resolution, and
beginning x and y locations of the nested domain within the mother domain. These
variable names are listed, and described further in the following section.
Table 2.2. Converted or intermediate binary emission data used to produce input
emissions data for a WRF-chemistry simulation. This table lists each output variable, its
variable declaration, dimensions, and any additional information. The output-data fields
are specific to the RADM2-chemical mechanism (+MADE/SORGAM aerosol module).
13
ket real (nx,nz,ny)
csl real (nx,nz,ny)
iso real (nx,nz,ny)
pm2.5i real (nx,nz,ny)
pm2.5j real (nx,nz,ny)
so4i real (nx,nz,ny)
so4j real (nx,nz,ny)
no3i real (nx,nz,ny)
no3j real (nx,nz,ny)
orgi real (nx,nz,ny)
orgj real (nx,nz,ny)
eci real (nx,nz,ny)
ecj real (nx,nz,ny)
pm10 real (nx,nz,ny) (end loop)
The emissions units for both surface and elevated point sources are in mole per
square kilometer per hour for gas phase species and in units of microgram per square
meter per second (microgram m-2 s-1) for the aerosol species. These are the units assumed
within the WRF-Chem input processor for the emissions files, and the convert_emiss.F
processing step that generates the netcdf emission file(s) described further below.
(Conversion of gas-phase emissions into the mixing ratio increases at each grid is
handled within module_emissions_anthropogenic.F. Aerosol increases due to emissions
are handled in individual aerosol mechanism modules.)
It is entirely incumbent upon the user to specify location and time of emissions
from the raw emissions for their own applications within this intermediate step of the
emissions processing.
2.4.2 Additional details for running emiss_v03.F with the NEI-05 anthropogenic-
emissions data set
14
Before generation of the anthropogenic-emissions data file can begin, the real.exe
program should be used to create the wrfinput_d01 and wrfbdy_d01 data files for
your desired domain. There are two reasons for doing this. The first is so that you
know exactly where the simulation domain is located. The second is because the
emissions conversion program (convert_emiss.exe) will read the netCDF header
information from the wrfinput_d01 data file and write this information into the WRF-
chemical-emissions data files. If the wrfinput_d01 data file does not exist, the
program will abort with an error.
Download the raw-emissions data tar file from the anonymous ftp server
(ftp://aftp.fsl.noaa.gov/divisions/taq/emissions_data_2005/em05v2_file*.tar) and
extract the data into its own directory (e.g., $home/emissions_data).
Modify emiss_v03.F to set map and grid parameters for your particular domain as
well as the directory that contains the raw emissions data. For the provided test
domain you should have the following settings:
15
DATA ZFA/ 0., 255., 515., 781., 1054., 1335., &
1527., 1824., 2130., 2553., 2994., 3454., &
4059., 4967., 6741., 8723.,10992.,14084., &
16461.,20346./
Note: the model ZFA height levels have been determined separately by processing the
wrfinput_d01 file generated by real.exe to obtain the average height of each vertical
domain levels in meters.
Now that the program is built and runs, one should have produced one or more
binary intermediate data files containing the anthropogenic emissions data for your WRF
domain. The next step in the process is to convert the binary intermediate files to a
WRF-Chem input file. The conversion code is discussed in chapter 8.
As the NEI emissions data is hourly and typically contains 12 hours of data, you
need to run the convert_emissions program twice to generate a full days worth of
emission. This program will also need a wrfinput data file for the desired date and
domain. This input file can be generated from a non-chemistry run of real.exe (e.g.,
chem_opt=0) as long as the start and end times match with the emissions start hour. That
is, the 00 UTC emissions file needs a wrfinput file that begins at 00 UTC and likewise for
12 UTC. The emissions input file needs to be copied, or linked to the
WRFV3/test_em_real directory. And it should be noted that the conversion program is
capable of converting the other files from prep_chem_sources (less the anthropogenic
emissions) during the same run. In this case, you might have linked the files with names
as:
And then before one can run the convert emissions program convert_emiss
(convert_emiss.exe) the namelist.input settings for the chemistry needs to be turned back
on (use chem_opt=2 and emiss_opt=3 for NEI emissions). Be sure to double check your
other namelist settings as well or you might not get the desired data files.
&time_control
io_form_auxinput5 =2
io_form_auxinput6 =2
io_form_auxinput7 =2
io_form_auxinput8 =2
io_form_auxinput12 = 2,
io_form_auxinput13 = 0,
auxinput6_inname = 'wrfbiochemi_d01',
auxinput7_inname = 'wrffirechemi_d<domain>',
auxinput8_inname = 'wrfchemi_gocart_bg_d<domain>',
16
auxinput12_inname = 'wrf_chem_input',
auxinput13_inname = 'wrfchemv_d<domain>'
auxinput5_interval_m = 1440,1440
auxinput7_interval_m = 1440,1440
auxinput8_interval_m = 1440,1440
auxinput13_interval_m = 1440,1440
frames_per_auxinput6 = 1,1
frames_per_auxinput7 = 1,1
frames_per_auxinput8 = 1,1
frames_per_auxinput13 = 1,1
/
&chem
kemit = 1,
chem_opt = 2, 2,
io_style_emissions = 2,
emiss_inpt_opt = 1, 1,
emiss_opt = 2, 2,
emiss_opt_vol = 0, 1,
biomass_burn_opt = 1, 1,
plumerisefire_frq = 120, 120
When complete and the data files are verified you can rename output to include the date
if so necessary. Otherwise, the global anthropogenic emissions file is ready to be used in
a WRF-Chem simulation.
17
o A time variable representing the time since a given date. The base date
must be contained in the time variable units attribute and be of the form
yyyy-mm-dd.
One or more three-dimensional variables containing the actual anthropogenic
emissions, designated as sub categories. The emissions data must be
dimensioned as (longitude, latitude, time).
Additionally the input anthropogenic dataset may have the molecular weight of the
input species in g per mole as either:
a variable with the name molecular_weight,
a global attribute with the name molecular_weigh,
a sub category variable attribute with the name molecular_weight.
If the input dataset does not specify the species molecular weight then the user may
specify the input species molecular weight in the src_names namelist variable. If the
molecular weight is specified via the src_names namelist variable then any molecular
weight information in the input dataset is ignored.
Compilation of the anthro_emiss program is accomplished using the command
make_anthro which will produce the anthro_emiss executable file. Before running
the anthro-emiss program the input file(s) for the WRF domain(s) needs to have been
constructed and the location known. In addition, there needs to be an input file (e.g.,
anthro_emiss.inp) constructed for the anthro_emiss program containing information
such as the executable directory and the wrfinput file locations. With the anthro_emiss
input file built, the code can be run using the redirecting command
anthro_emis < anthro_emis.inp.
If the user desires to capture the output going to screen to a data file (e.g.,
anthro_emiss.output) the command is instead:
18
Chapter 3: Generation of Biogenic Emissions Data
Files
Table of Contents
3.1 Preparation of biogenic emissions for use with WRF-Chem...................................... 19
3.2 No biogenic emissions ................................................................................................ 19
3.3 Guenther biogenic emissions ...................................................................................... 19
3.4 BEIS 3.14 biogenic emissions .................................................................................... 19
3.5 MEGAN biogenic emissions ...................................................................................... 21
3.5.1 Compiling MEGAN preprocessor ....................................................................... 22
3.5.2 Preprocessing MEGAN data files ........................................................................ 22
3.5.3 Running WRF-Chem with MEGAN ................................................................... 23
The WRF-chemistry model does have four options to compute the biogenic-
emissions data. These options are: None, Guenther, BEIS and MEGAN biogenic
emissions. Each of these options will be discussed in the following sections.
The first option is not to use an additional biogenic-emissions input data file
(bio_emi_opt= 0). The user could add the biogenic-emission to the anthropogenic-
emissions data if it is desired. For example, the prep_chem_sources program has the
option to include estimated biogenic emissions with the anthropogenic emissions. Be sure
to include biogenic emissions for every update time in the anthropogenic emissions input
data. The biogenic emissions information is not static and it will be lost whenever there
is an update to the anthropogenic emissions data.
For the second option (bio_emi_opt=1), the model calculates the biogenic
emissions online using the USGS land-use classification, which is generated by WRF
WPS and available for the meteorological and chemical model. The user does not prepare
any biogenic-emissions input data set. Therefore, this option is the easier to use.
However, the overall simulation might not compare well with observations due to the
limited vegetation types in the simulation resulting in several key chemical species (e.g.,
isoprene) having relatively low emisison rates.
For the third option, the user specifies reference fields for the biogenic emissions,
which are then modified online by a subroutine from the Biogenic-Emissions Inventory
System (BEIS) version 3.14. The land use for this emissions inventory is obtained from
19
the Biogenic-Emissions Land-use Database version 3 (BELD3). The reference fields
need to be provided as an additional input data file (wrfbiochemi_d01) for the real.exe
program. At this time the data files containing the reference fields have not been made
available to the general public.
Table 3.1. List of biogenic-emission variable names, and their RACM mechanism
meaning.
20
Ald Acetaldehyde (and higher aldehydes)
Hcho Formaldehyde
Eth Ethane
ora2 Acetic and higher organic acids
Co Carbon Monoxide
Nr Nonreactive VOC
noag_grow Agricultural NO fertilized growing
noag_nongrow Agricultural NO nonfertilized growing
Nononag Nonagricultural NO
Slai LAI for isoprene emissions
When using this option the WRF-Chem code will expect a data file called
wrfbiochemi_d01 (for domain 1) and a setting of bio_emiss_opt = 2 in your
namelist.input file. One will need to run real.exe (with bio_emi_opt=2 in the namelist)
when creating the wrfinput data file to load the static biogenic emissions variables should
now be included in the wrfinput data file.
The final option for biogenic emissions is the use of the Model of Emissions of
Gases and Aerosols from Nature (MEGAN). This global biogenic-emissions data set has
a horizontal spatial resolution of approximately 1 km so it can be used for nearly any
WRF-Chem simulation. The use of this biogenic-emissions option requires the user to
download and compile a further utility from the National Center for Atmospheric
Research (NCAR) web site http://www.acd.ucar.edu/wrf-chem/. The utility prepares
MEGAN input data files for use in WRF-Chem these files are named wrfbiochemi_d0x
files and a separate file is provided for each domain (d0x). The assimilation of MEGAN
biogenic emissions into WRF-Chem is a matter of setting the correct namelist.input
settings.
The University Corporation for Atmospheric Research (UCAR) provides Fortran source
code files to create MEGAN biogenic emission data for importing into WRF-Chem.
MEGAN is a global emissions dataset, at one-kilometer spatial resolution, compiled for
2003. A users guide and descriptions of the data set are provided at
http://cdp.ucar.edu/acp/megan.
The MEGAN toolkit for WRF-Chem preprocesses the MEGAN data set, and creates
wrfbiochem_d0x (x = domain number) input files for ingestion into WRF-Chem at model
run time.
The following instructions assume that real.exe and wrf.exe have been compiled
normally and that an initial meteorology only wrfinput_d0x file(s) have been created post
WPS. These instructions also assume you are using the provided 2003 MEGAN data
files. The tar file from UCAR includes a helpful readme file which expands on the
instructions below.
21
3.5.1 Compiling MEGAN preprocessor
The MEGAN preprocessor Fortran source code and data set have been made available
for download from NCAR/ACD at the web site http://cdp.ucar.edu/acp/megan This site
will for each visitor to register by providing some contact details. After registering just
click on the bio-emiss button on the user registration page to go to the biogenic
emissoin page. At this page one can download the MEGAN Fortran source code,
makefile and MEGAN data input files. The files can be downloaded to a directory of
your choice. Keep in mind that the MEGAN code needs to access the WRF run directory
and the wrfinput_d01 file during MEGAN data preprocessing.
Next the downloaded files need to be extracted usign the tar command. One can un-
tar the files downloaded in the MEGAN directory by issuing the commands:
This creates two other tar files, un-compress and un-tar these files:
After extracting the file be sure that the correct environment variables for your
FORTRAN compiler and netCDF libraries are set correctly by editing the make_util
script file if necessary. At this point the MEGAN source code in the MEGAN directory
can be compiled by issuing the command:
And once the code is compiled the executable file megan_bio_emiss will be created in
the MEGAN directory.
With the MEGAN executable constructed one needs to have a WRF input file
available for the desired domain (2). If a wrfinput file is not available, then run real.exe
for a meteorology only case to produce the input files. While this step will create a file
named wrfinput_d01 examination of the file will show that it does not contain any input
data arrays for MEGAN. The next step using MEGAN pre-processing tool will examine
the wrfinput_d01 for geographical and temporal parameters for the production of
biogenic emissions and use them in the construction of the WRF biogenic emission files.
Edit the MEGAN preprocessor input file (megan_bio_emiss.inp) and set the values in the
test file for:
22
1. domains the number of domains used in your WRF model simulation
2. start_lai_month - the month before the month in which your WRF run is set to
start. Typically this will be set to 1.
3. end_lai_month the month in which your WRF run is set to end. Typically this
will be set to 12.
4. wrf_dir the directory where the associated wrfinput_d01 for your WRF-Chem
simulation resides, and
5. megan_dir the directory in which the downloaded MEGAN data files reside.
Keep in mind that the leaf area index (lai) parameter data files are provided in the
MEGAN preprocessing download. There will be twelve of these files as one file exists
for each month.
With the input file constructed the MEGAN bioegenic-emissions data can be created
for the WRF-Chem domain(s) and time frame by running the MEGAN utility. The utility
us executed by issuing the command:
Here the execution command uses redirection to supply the executable with the input file
and saves all of the screen output to a log file called megan_bio_emiss.log. Whereas the
reading and writing of data files is a serial process, the running of this program can take a
relatively long time depending on a range of factors such as computer power and
available memory. Once the execution of the program is finished it is always best to
examine the log file and the output for any possible errors or issues. You might review
the wrfbiochemi_d0x files with ncview to ensure the correct geographical bounds have
been applied and that the MEGAN data sets are included. The MEGAN data elements in
the WRF-Chem biogenic emission data files are:
With the successful running of the MEGAN preprocessor, the created WRF-Chem
biogenic input files (wrfbiochemi_d0x files) should be copied (or linked) into the WRF-
Chem directory WRFV3/test/em_real.
23
At this point a normal WRF-Chem simulation can be made. This includes the
running of real.exe to create the input and boundary files as well as combine any static
emission fields to the appropriate input file. When running the WRF-Chem code with
MEGAN biogenic emissions data be sure to set the appropriate namelist values. In this
case one will use bio_emiss_opt=3 and will need to set the parameter ne_area to a
value equal to or greater than the total chemical species used. See chapter 4 of the WRF-
Chem Users Guide for more inforamtion about the ne_area parameter.
Keep in mind that after running real.exe the wrfbiochemi_d01 and wrfinput_d01 files
could also include variables for running BEIS3.14, but the values for BEIS variables will
all be all zero. If one is wanting to use BEIS biogenic emisisons, then the user will need
to create a different wrfbiochemi_d01 file using a different source for the biogenic
emisisons input data.
With the static MEGAN biogenic emissions data loaded into the wrfinput file and the
WRF biogenic emissions data file availble for reading during the run, the WRF
executable can be run. During execution of the WRF code the MEGAN emissions will be
calculated online using the simulated meteorological conditions along with the biogenic
emissions.
24
Chapter 4: Generation of Biomass Burning Emissions
Data Files
Table of Contents
4.1 Preparation of biomass burning emissions ................................................................. 25
4.2 Using prep_chem_sources for biomass burning emissions ........................................ 25
4.3 Using FINN biomass burning emissions .................................................................... 27
Once the prep_chem_sources executable has been run the binary intermediate
files it produces for WRF-Chem will need to be converted to WRF input data files. The
conversion code is discussed in more detail in chapter 8 of this Guide so the running of
the code for the biomass burning emissions will be discussed next.
25
For each simulation that you are going to use biomass burning data, you need to
run the prep_chem_sources and convert_emissions programs. This program will also
need a wrfinput data file for the desired date as well. So to begin, one can run the
real.exe with chem_opt=0 and start and end time set to your year, month, day, and start
hour generate a wrfinput_d01 file. Then the output from prep_chem_sources is linked to
the WRF running directory (e.g., WRFV3/test/em_real) with commands similar to the
following:
And then before one can run the convert emissions program convert_emiss
(convert_emiss.exe) the namelist.input settings for the chemistry needs to be turned back
on (e.g., chem_opt=301, biomass_burn_opt = 1). Be sure to double check your other
namelist settings as well or you might not get the desired data files.
&time_control
io_form_auxinput5 =2
io_form_auxinput6 =2
io_form_auxinput7 =2
io_form_auxinput8 =2
io_form_auxinput12 = 2,
io_form_auxinput13 = 0,
auxinput6_inname = 'wrfbiochemi_d01',
auxinput7_inname = 'wrffirechemi_d<domain>',
auxinput8_inname = 'wrfchemi_gocart_bg_d<domain>',
auxinput12_inname = 'wrf_chem_input',
auxinput13_inname = 'wrfchemv_d<domain>'
auxinput5_interval_m = 1440,1440
auxinput7_interval_m = 1440,1440
auxinput8_interval_m = 1440,1440
auxinput13_interval_m = 1440,1440
frames_per_auxinput6 = 1,1
frames_per_auxinput7 = 1,1
frames_per_auxinput8 = 1,1
frames_per_auxinput13 = 1,1
/
&chem
kemit = 1,
chem_opt = 301, 2,
io_style_emissions = 2,
emiss_inpt_opt = 1, 1,
emiss_opt = 5, 5,
emiss_opt_vol = 0, 1,
26
biomass_burn_opt = 1, 1,
plumerisefire_frq = 120, 120
And when complete and the converted biomass burning files is verified you can rename
output, or just right away use the biomass burning emissions file (wrffirechemi_d01) in a
WRF-Chem simulation.
The Fire INventory from NCAR (FINN) program (Wiedinmyer et al., 2011) can
be used to provide fire emissions data for a WRF-Chem simulation. FINN uses daily,
one kilometer resolution, global estimates of the trace gas and particle emissions from
open burning of biomass, which includes wildfire, agricultural fires, and prescribed
burning, but does not include biofuel use and trash burning. Emission factors used in the
calculations have been updated with recent data, particularly for the non-methane organic
compounds (NMOC). The resulting global annual NMOC emission estimates are as
much as a factor of 5 greater than some prior estimates. Chemical speciation profiles,
necessary to allocate the total NMOC emission estimates to lumped species for use by
chemical transport models, are provided for three widely used chemical mechanisms:
SAPRC99, GEOS-CHEM, and MOZART-4. Using these profiles, FINN also provides
global estimates of key organic compounds, including formaldehyde and methanol.
Any uncertainties in the emissions estimates arise from several of the method
steps. The use of fire hot spots, assumed burned area, land cover maps, biomass
consumption estimates, and emission factors all introduce error into the model estimates.
The uncertainty in the FINN emission estimates are about a factor of two; but, the global
estimates agree reasonably well with other global inventories of biomass burning
emissions for CO, CO, and other species with less variable emission factors.
Near-real-time fire emissions from the FINN, based on MODIS Rapid Response
fire count (FIRMS) are made available at the NCAR/ACD web site
http://acom.ucar.edu/acresp/dc3/AMADEUS/finn/emis and the FINN fire emissions for
previous years are available from http://bai.acom.ucar.edu/Data/fire.
27
28
Chapter 5: Generation of Volcanic Emissions Data Files
Table of Contents
5.1 Preparation of volcanic emissions for use with WRF-Chem ...................................... 29
5.2 Volcanic ash emissions ............................................................................................... 29
5.3 Volcanic SO2 degassing emissions ............................................................................. 30
In this section the ash and SO2 emissions from prescribed volcanic activities is
described. The volcanic ash emissions can either be for a single volcano producing just
ash, or for a single volcano emitting ash and sulfurdioxide. Both methods use the
prep_chem_sources utility program for specifying the location of the volcano on the
WRF-Chem domain and the type of emissions from the volcano. For additional
information on volcanic emissions that are produced when running prep_chem_sources
the reader is referred to Freitas et al. (2011).
29
5.3 Volcanic SO2 degassing emissions
30
Chapter 6: Generation of Tracer Emissions Data Files
Table of Contents
6.1 Preparation of tracer emissions for use with WRF-Chem .......................................... 31
6.2 The CO2 tracer option ................................................................................................. 32
6.3 The Greenhouse Gas tracer option .............................................................................. 32
At this time there is no single program available that will allow the user to
construct tracer emissions for their domain and directly import them into the simulation.
However, there are relatively simple methodologies that can be undertaken to allow the
user to generate their tracer emissions.
Probably the simplest way to produce tracer emissions is to modify the sample
program provided in chapter 8.2 to suit your needs. Upon examination of the sample
program one will see that it can be modified to ignore the reading, or setting of
anthropogenic emissions data sets and instead simply fill the EMISS3D array. The three-
dimensional emissions array (EMISS3D) can be filled with user-specified values at the
desired grid location with the user-specified emitted tracer amount. The user needs to be
aware of not only where the emitted species are to be located in their simulation domain
(grid indexes I,K and J), but also which chemical emissions index, N, corresponds to the
emitted species used as a tracer in their simulation. To get the emissions correct, start by
examining the registry.chem file to determine which species are used in the tracer
package (e.g., chem_opt=13). Then use the array ename in the sample program to get
the name and order of the emitted species. Knowing the chemical species index in
ename, one can set the emissions for the chemical index, N, in the output array. For
example, if the tracer emissions is to be defined for CO at the surface for grid point
(18,23), then the read and setting section of the SAMPLE program would be:
N=11
DO IHR=1,12 ! file contains 12 hours of data (00 to 11 UTC)
EMISS3D(18,1,23,N, IHR) = 1.E-06
ENDDO
Once the SAMPLE program has been modified, compiled and run, a binary
intermediate file containing the tracer emissions will be produced. This binary file
containing will need to be converted as a NEI emissions data set to a WRF netCDF input
file using the convert_emiss.F program discussed in the previous section. Another
example routine using this methodology has been made available on the WRF-Chem ftp
server at: ftp://aftp.fsl.noaa.gov/divisions/taq/emissions_data_2005/emiss_v03_tracer.F
31
If using the sample programs is not functional for the users problem another
methodology can be used. It has ben assumed that an option that allows users to use pre-
existing data and code with only small modifications needed to produce input or output
files for their needs would be the best way for students to begin using the model.
The CO2 tracer option allows a user to simulate biospheric CO2 fluxes using
Vegetation Photosynthesis and Respiration Model (VPRM) coupled to WRF-Chem. The
VPRM model uses satellite data and WRF meteorology to derive biospheric CO2 fluxes
from different biomes. All the chemistry species such as CO2 and CO in this option are
treated as passive tracers. No chemistry and removal processes are considered for this
option. The option simulates transport and mixing of several CO2 and CO tracers from
different sources such as background, anthropogenic and biospheric. More information
regarding VPRM is provided by Pillai et al. (2011) and Ahmadov et al. (2009).
The Greenhouse Gas tracer option allows a user to perform simulations of CO2,
CO, and CH4 within WRF-Chem. All chemistry species are treated as passive tracers.
The option includes all tracers from the CO2 tracer option but also includes CH4 tracers
(biogenic, anthropogenic, background) and biomass burning tracers for CO2, CO, and
CH4 as well. Biogenic CH4 fluxes are simulated with the Kaplan wetland inventory, the
soil uptake model from Ridgwell, and the termite database of Sanderson coupled to
WRF-Chem. The flux models use WRF meteorology and soil characteristics to calculate
the CH4 fluxes. The wetland inventory additionally relies on an external wetland
inundation map and carbon pool. Similar to chem_opt=16 no chemistry and removal
processes are considered. Detailed information is found in Beck et al. (2011).
32
Chapter 7: Other WRF-Chem Utilties
Table of Contents
7.1 Boundary conditions with WRF-Chem ...................................................................... 33
7.2 The wrfchembc utility ................................................................................................. 33
7.3 The mozbc utility ........................................................................................................ 34
7.4 Including an upper boundary bounary condition for chemical species ...................... 34
ftp://aftp.fsl.noaa.gov/divisions/taq/Boundary/wrfchemv2.2_bcond_code_09Apr07.tar
You can then modify the Makefile to use your desired compile options and
compile to generate the wrfchembc executable. You must also modify the
wrfchembc_namelist.input file to have the correct data directories and species added to
the wrfbdy file.
Run the wrfchembc program after real.exe and before wrf.exe to add the global
model data to the lateral boundary data file (wrfbdy_d01). In addition, before running
wrf.exe, modify the namelist.input to set have_bcs_chem = .true..
33
7.3 The mozbc utility
The mozbc utility requires users to have a WRF-Chem initial condition file
(wrfinput_d<domain>) for each domain of interest and/or a WRF-Chem boundary
condition file (wrfbdy_d01) for the first WRF-Chem domain. The MOZART datasets are
interpolated in space (bilinearly in longitude, longitude and linearly in pressure), but not
in time. Therefore, users need to ensure that the times in the MOZART and WRF-Chem
files are matching for the period of interest. Otherwise, if a non-matching time is found,
the mozbc will abort the run with an error statement.
An upper boundary condition for select gas species may be specified by setting
the have_bcs_upper in the chemistry namelist. The namelist variable have_bcs_upper
defaults to .false. meaning that no chemical species concentrations are specified near the
upper boundary. By setting have_bcs_upper to .true. the model will specify the o3, no,
no2, hno3, ch4, co, n2o, and n2o5 concentrations at the top of the model. These values
will override the original values as defined in the idealized chemical profile (section 4.5).
To use the upper boundary contisions, the user is required to provide two additional
input data files:
34
Climatologies for 4 different time periods derived from WACCM RCP simulations have
been made available to users from the NCAR/ACD website (www.acd.ucar.edu/wrf-
chem). These files are named: ubvals_b40.20th.track1_1950-1959.nc,
ubvals_b40.20th.track1_1980-1989.nc, ubvals_b40.20th.track1_1996-2005.nc, and
ubvals_rcp4_5.2deg_2020-2029.nc where the years used to produce the climatology are
specified in the file names. Additional output variables are included in the model when
using the upper boundary conditions. These tropopause diagnostics (TROPO_P,
TROPO_Z, TROPO_LEV) are listed in the registry and a user should verify that they
will included in the output file before running the model. Additional information about
the upper boundary condition scheme was provided in presentation 8A.2 (Barth et al.)
given at the 2011 WRF User Workshop. The presentation can be accessed online at
http://www2.mmm.ucar.edu/wrf/users/workshops/WS2011/WorkshopPapers.php.
35
Chapter 8: WRF-Chem Conversion Utility
Table of Contents
8.1 Converting binary intermediate file to a WRF-Chem data file................................... 36
8.2 Binary data file format ................................................................................................ 37
8.3 Building the WRF-Chemistry emissions conversion code ......................................... 38
8.4 Namelist settings for running convert_emiss .............................................................. 38
The final step in the process for some of the utility programs is to produce a
WRF-Chem netCDF input data file containing all of the required metadata (map
projection data, simulation start time, etc.) for the simulation. Ideally the metadata
contained in the WRF-Chem emissions input file would be generated by the emissions
mapping step. However, if not provided by the user, the WRF input data file (e.g.,
wrfinput_d01) can be read and the metdata information extracted and added to the
anthropogenic-emissions input data file.
When selecting a file name for the WRF chemical emissions, it was decided to
use the prefix of wrfchemi for anthropogenic emission. Using this prefix, the name of the
final netcdf data file(s) can either include the hour of the day
wrfchemi_<hour>_d<domain_id> or the full date and time
wrfchemi_d<domain_id>_<date/time> depending upon the users intent. The chemistry
namelist variable io_style_emissions allows the model run to switch between the two
naming conventions. The nameing option including the date and time stamp is designed
for daily varying emissions (io_style_emissions=2). When selecting this option the
date/time specified in the WRF-Chem emissions data file must match the simulation
date/time or no emissions data will be read potentially resulting in a run-time error. The
emissions file name option that includes just the hour (io_style_emissions=1) assumes
you are creating two netcdf 4-D emission files named wrfchemi_00z_d<domain_id> and
wrfchemi_12z_ d<domain_id>. Each of these files contain hourly emissions data with
hours ranging from 00:00 to 11:00 UTC for the wrfchemi_00z_d<domain_id> file, or
from 12:00 to 23:00 UTC for the other. This latter naming convention is used by
convert_emiss.F when using the provided NEI-05 emissions inventory.
36
anthropogenic emissions for the start of the simulation and the update interval would be
defined as a time period greater than the simulation time.
Perhaps the best way to explain the format of the intermediate binary data file is
to provide a sample FORTRAN program that demonstrates how it is written. The arrays
are defined in the program as cell center data on an Arakawa C grid. Next the names of
the emitted species are defined and the 3D emissions array is filled. Once this is
accomplished all of the 3D emissions data is written to file with a header defining the
number of emissions fields and their names.
PROGRAM SAMPLE
PARAMETER(IX=39,JX=39,KX=19)
PARAMETER(NRADM=30, IHOUR=12)
CHARACTER (len= 9), DIMENSION(NRADM) :: ENAME
REAL, DIMENSION(IX2,KX,JX2,NRADM,IHOUR) :: EMISS3D
! Temporary array used for output of data
REAL, DIMENSION(IX2,KX,JX2) :: TMP
! Names of emission variables and the order they are written to file
DATA ENAME / &
'e_so2 ','e_no ','e_ald ','e_hcho','e_ora2', &
'e_nh3 ','e_hc3 ','e_hc5 ','e_hc8 ', &
'e_eth ','e_co ','e_ol2 ','e_olt ','e_oli ','e_tol ','e_xyl ', &
'e_ket ','e_csl ','e_iso ','e_pm25i','e_pm25j', &
'e_so4i','e_so4j','e_no3i','e_no3j','e_orgi','e_orgj','e_eci', &
'e_ecj','e_pm10'/
! < Read and set all emissions here and place data in EMISS3D array>
OPEN(19,FILE='wrfem_00to12Z',FORM='UNFORMATTED')
WRITE(19)NRADM
WRITE(19)ename
37
ENDDO
ENDDO
At this point all of the WRF-Chemistry conversion utility has been built and the
emissions input data can to be generated.
Before the conversion from binary intermediate to netcdf file emissions can
begin, you need to change the namelist.input file in the WRFV3/test/em_real directory to
the emissions data file settings. The settings that will generate 24 hours of hourly-
emissions (i.e., io_style_emissions=2) data are:
&time_control
run_days = 1,
run_hours = 0,
run_minutes = 0,
run_seconds = 0,
start_year = 2008,
start_month = 07,
38
start_day = 14,
start_hour = 00,
start_minute = 00,
start_second = 00,
end_year = 2008,
end_month = 07,
end_day = 15,
end_hour = 12,
end_minute = 00,
end_second = 00,
auxinput6_inname = wrfbiochemi_d<domain>,
auxinput7_inname = wrffirechemi_d<domain>,
auxinput8_inname = wrfchemi_gocart_bg_d<domain>,
auxinput12_inname = wrf_chem_input,
auxinput13_inname = wrfchemv_d<domain>,
auxinput5_interval_m = 60,
auxinput7_interval_m = 1440,
auxinput8_interval_m = 1440,
io_form_auxinput2 = 2,
io_form_auxinput5 = 2,
io_form_auxinput6 = 2,
io_form_auxinput7 = 2,
io_form_auxinput8 = 2,
io_form_auxinput12 = 0,
io_form_auxinput13 = 2,
io_form_auxinput14 = 0,
io_form_auxinput15 = 0,
/
&chem
chem_opt = 1,
emiss_opt = 3,
chem_in_opt = 0,
bio_emiss_opt = 1.
Since version 3.3, the name list setting for the auxiliary input port time interval is
now dependent upon each input stream. That is, for the emissions conversion program
convert_emiss.exe, the settings for each auxiliary input port specifies whether it is turned
on, the IO format, the time interval for the emissions data updates and the number of
frames in each file (not shown).
When running the emissions utility programs and converting the input data into a
WRF-Chem input data files, there are several items to keep in mind. First, it is assumed
that one has already configured and compiled the WRF-Chem model. This needs to be
accomplished before building the emissions conversion program with the command
39
compile emi_conv in WRFV3 directory. If the WRF-Chem code is not constructed
then several required modules will not be found and a compile error will result.
Second, the user should run convert_emiss.exe for a 24 hour period starting at
0000 UTC and ending at 0000 UTC. The netCDF output files named wrfchemi_d01 will
need to be saved under a file with a different name such as wrfchemi_d01_2008-07-
14_00:00:00. The filename(s) need to match the name of the input data file as specified
in the WRF module mediation_integrate.F (see convert_emiss.F program inside the
WRFV3/chem directory). The file(s) should be transferred (or linked) to your
WRFV3/test/em_real directory so that they can be read in during the WRF-Chem
simulation.
Finally, once the wrfchemi files are generated, it is best to plot fields or use a
program like ncview to examine the files that have been generated. Look at the files and
confirm that the emissions appear to match the WRF forecast domain you previously
generated. When looking at anthropogenic emissions (e.g., co) the surface emissions
should look similar to a map with cities and possibly roads showing. Be sure to confirm
that the fields are consistent with your expected emissions fields in both space (vertically)
as well as in time. If the emissions do not match, then a dimension error is likely
happened in your namelist.input file.
40
Chapter 9: Summary
Table of Contents
9.1 Summary ..................................................................................................................... 41
9.2 WRF-Chem and utility program related publications ................................................. 41
9.1 Summary
Presented in this WRF-Chem Emissions are the descriptons and directions for
setting-up and using the utilities associated with the WRF-Chem model. This guide is
not intended to be an exhaustive report about all that is available to aid in the set up and
use the WRF-Chem model emissions utilities. While it does attempt to provide the latest
and most accurate information about the configuration and running of the various
emission utiltities, errors or incomplete information may have been unintentionally
presented. Also, due to the complexity of the each utility and the WRF-Chem model
along with the diverse needs of each user, there may be insufficient information for your
particular research or operational application. If a user has questions regarding the model
that this document fails to answer, the one should contact the WRF-Chem help desk at
wrfchemhelp.gsd@noaa.gov, or explore the WRF community forum and see if the user
community can povide an answer.
As was stated in beginning chapters, you will need to consider your needs and/or
requirements for the domain of interest before beginning the simulation. This includes,
but is not limited to the available meteorological and emissions data sets. Also, the WRF
model, and likewise the WRF-Chem model, is being continuously updated. Therefore,
you are advised to stay involved in the WRF-Chem user community to be made aware of
any and all updates to or issues with the code.
All WRF-Chem users are also advised to link their web browser to the WRF-
Chem user group web page (http://www.wrf-model.org/WG11) and periodically scan the
pages for changes and/or updates to the model. These web pages contain answers to
frequently asked questions, or FAQs. So this is a good place to start when you have a
question regarding the setup, use, or performance of the WRF-Chem model. Finally, this
web page contains the most up to date list of relevant publications regarding the WRF-
Chem model. When presenting, or publishing results from studies using the WRF-Chem
model, it is requested that you cite the Grell et al. (2005) and Fast et al. (2006)
manuscripts provided in the relevant publications section of this chapter. For any
application that uses the indirect effect, please also cite Gustafson et al. (2007). And
likewise, when using other significant features in the WRF-Chem model, the user should
examine the reference list on the WRF-Chem web page and cite the developers paper(s)
(http://ruc.noaa.gov/wrf/WG11/References/WRF-Chem.references.htm). A more detailed
model description with a series of papers is in the works and may appear in a new journal
that is intended for model description papers only.
41
Ahmadov, R., et al., 2012: A volatility basis set model for summertime secondary organic
aerosols over the eastern United States in 2006, J. Geophys. Res., 117, D06301,
doi:10.1029/2011JD016831
Barnard, J.C., J.D. Fast, G.L. Paredes-Miranda, and P.W. Arnott, 2008:
Closure on the single scattering albedo in the WRF-Chem framework using data
from the MILAGRO campaign. Atmos. Chem. Phys. Discuss., 9, 5009-5054.
Barth, M., C., J. Lee, A., Hodzic, G. Pfister, W. C. Skamarock, J. Worden, J. Wong, and
D. Noone, 2012: Thunderstorms and upper troposphere chemistry during the early
stages of the 2006 North American Monsoon, Atmos. Chem. Phys. Discuss., 12,
16407-16455, doi:10.5194/acpd-12-16407-2012.
Chapman, E.G., W.I. Gustafson Jr., R.C. Easter, J.C. Barnard, S.J. Ghan, M.S. Pekour,
and J.D. Fast, 2008: Coupling aerosols-cloud-radiative processes in the WRF-
chem model: Investigating the radiative impact of large point sources. Atmos.
Chem. Phys., 9, 945-964.
Darby, LS; McKeen, SA; Senff, CJ; White, AB; Banta, RM; Post, MJ; Brewer, WA;
Marchbanks, R; Alvarez, RJ; Peckham, SE; Mao, H; Talbot, R, 2007: Ozone
differences between near-coastal and offshore sites in New England: Role of
meteorology. J. Geophys. Res.-Atmos., 112 ( D16 ), Art No. D16S91, issn: 0148-
0227, ids: 208QX, 31-Aug 2007.
42
2006. Atmos. Environ., 44 (4) 455-467, issn: 1352-2310, ids: 556TD, doi:
10.1016/j.atmosenv.2009.11.007
Eder, B., D. Kang, A. Stein, J. McHenry, G. Grell, and, S. Peckham, 2005: The New
England Air Quality Forecasting Pilot Program: Development of an Evaluation
Protocol and Performance Benchmark. Journal of the Air and Waste Management
Association, 55, 20-27.
Freitas, S. R., Longo, K. M., Alonso, M. F., Pirre, M., Marecal, V., Grell, G., Stockler,
R., Mello, R. F., Sanchez Gacita, M., 2011. PREP-CHEM-SRC 1.0: a
preprocessor of trace gas and aerosol emission fields for regional and global
atmospheric chemistry models. Geosci. Model Dev., 4, 419-433.
43
press. Integrated Systems of Meso-Meterological and Chemical Transport
Models.
Grell, G. A., S. R. Freitas, M. Stuefer and J. Fast, 2011: Inclusion of biomass burning in
WRF-Chem: impact of wildfires on weather forecasts, Atmos. Chem. Phys., 11, ,
5289-5303. doi:10.5194/acp-11-5289-2011.
Gustafson Jr., W.I., E.G. Chapman, S.J. Ghan, and J.D. Fast, 2007: Impact on modeled
cloud characteristics due to simplified treatment of uniform cloud condensation
nuclei during NEAQS 2004. Geophys. Res. Lett., 34, L19809.
Kim, S.-W., A. Heckel, S.A. McKeen, G.J. Frost, E.-Y. Hsie, M.K. Trainer, A. Richter, J.
Burrows, S.E. Peckham, and G.A. Grell, 2006: Satellite-Observed US Power
Plant NOx Emission Reductions and Impact on Air Quality, Geophysical
Research Letters, 33, L22812, doi:10.1029/2006GL026310, 2006.
Ntelekos, A., J.A. Smith, L. Donner, J.D. Fast, E.G. Chapman, W.I. Gustafson
Jr., and W.F. Krajewski, 2009: Effect of aerosols on intense convective
44
precipitation in the northeastern U.S. Q. J. Roy. Meteor. Soc.,135, 1367-1391. doi:
10.1002/qj.476.
Pagowski, M; Grell, GA; McKeen, SA; Peckham, SE; Devenyi, D., 2010: Three-
dimensional variational data assimilation of ozone and fine particulate matter
observations: some results using the Weather Research and Forecasting -
Chemistry model and Grid-point Statistical Interpolation. Q. J. R. Meteorol.
Soc., 136 Part B (653) 2013-2024, issn: 0035-9009, ids: 694QO, doi:
10.1002/qj.700.
Pagowski, M; Grell, GA, 2006: Ensemble-based ozone forecasts: Skill and economic
value. J. Geophys. Res.-Atmos.: Vol. 111
Pagowski, M., G.A. Grell, D. Devenyi, S.E. Peckham, S.A. McKeen, W. Gong, L. Delle
Monache, J.N. McHenry, J. McQueen and P. Lee, 2006: Application of dynamic
linear regression to improve the skill of ensemble-based deterministic ozone
forecasts, Atmos. Environ., 40, 3240-3250.
Shrivastava, M., Fast, J., Easter, R., Gustafson Jr., W. I., Zaveri, R. A., Jimenez, J. L.,
Saide, P., and Hodzic, A., 2011: Modeling organic aerosols in a megacity:
comparison of simple and complex representations of the volatility basis set
approach, Atmos. Chem. Phys., 11, 6639-6662, doi:10.5194/acp-11-6639-2011.
Wiedinmyer, C., Akagi, S. K., Yokelson, R. J., Emmons, L. K., Al-Saadi, J. A.,
Orlando, J. J., and Soja, A. J.: The Fire INventory from NCAR (FINN): a high
45
resolution global model to estimate the emissions from open burning, Geosci.
Model Dev., 4, 625-641, doi:10.5194/gmd-4-625-2011, 2011.
Yang, Q., W.I. Gustafson Jr., J.D. Fast, H. Wang, R.C. Easter, H. Morrison Y.-N. Lee,
E.G. Chapman, S.N. Spak, and M.A. Mena-Carrasco, 2011: Assessing regional
scale predictions of aerosols, marine stratocumulus, and their interactions during
VOCALS-REx using WRF-Chem. Atmos. Chem. Phys., 11, 11951-11975.
Zhao, C., X. Liu, L.R. Leung, B. Johnson, S. McFarlane, W.I. Gustafson Jr., J.D. Fast,
and R. Easter, 2010: The spatial distribution of dust and its short wave radiative
impact over North Africa: Modeling sensitivity to dust emissions and aerosol size
treatments. Atmos Chem. Phys., 10, 8821-8838.
46