Implementation of the SMOKE Emission Data Processor
and SMOKE Tool Input Data Processor in Models-3
William G. Benjey*
Atmospheric Science Modeling Division, Air Resources Laboratory
National Oceanic and Atmospheric Administration
Research Triangle Park, North Carolina 27711
benjey@hpcc.epa.gov
Marc R. Houyoux
Environmental Programs, MCNC North Carolina Supercomputing Center
3021 Comwallis Road
Research Triangle Park, North Carolina 27709-2889
mhouyoux@ncsc.org
Joseph W. Susick
Science Applications International Corporation
6565 Arlington Boulevard
Falls Church, Virginia 22042-3000 -
ioseph.w.susick@saic.com
ABSTRACT
The U.S. Environmental Protection Agency has implemented Version 1.3 of SMOKE (Sparse
Matrix Object Kernel Emission) processor for preparation of area, mobile, point, and biogenic sources
emission data within Version 4.1 of the Models-3 air quality modeling framework. The SMOKE
system includes MOBILE 5b to model on-road emissions and the Biogenic Emission Inventory System,
Version 2 (BEIS 2) to model biogenic emissions. Although SMOKE may be operated from scripts
outside of the Models-3 framework, integration within the system allows automatic registration and
tracking of input and output files, as well as close coupling to the Community Multiscale Air Quality
(CMAQ) modeling system and visualization tools by using the Net CDF I/O API data format
convention. Models-3 also includes the SMOKE Tool to assist in preparation of emission input data for
SMOKE. The SMOKE Tool performs basic functions for the user, including 1) import and quality
checking of emission inventory and related data, 2) preparation of user-defined model grids and
gridding of emission data and surrogates (SMOKE Tool requires a Geographic Information System
(GIS), 3) preparation of user-defined "packet" files which instruct SMOKE to apply factors including
growth, control, and reactivity analysis factors (individual species may be added or deleted by source
and location), and 4) allows the user to edit input files. Manual preparation of input files outside of the
system is possible, but increases the probability of human error and substantially increases the time
needed for data preparation.
INTRODUCTION
The Models-3 air quality modeling framework was designed to be a flexible, multi-scale, multi-
pollutant system, that would supercede the need to develop and maintain specialized regional air quality
models for individual pollutants.' One purpose of the framework design was to ensure consistency of
the data exchanged by traditionally separate modeling components (meteorology, emissions, and
*On assignment to the National Exposure Research Laboratory, U, S. Environmental Protection
Agency.
-------
chemistry and transport), with regard to the spatial and temporal domains being used. This approach
avoids many potential errors from use of inappropriate data, and allows easier tracking of input and
output data files. Accordingly, the modeling of emission data was included as an integral component of
the system. Modeling components that are fully integrated into the Models-3 system are capable of
communicating at many levels with other modules of the system using the NetCDF (Common Data
Format) I/O API (Input/Output Applications Programming Interface) format convention,
Emission Processor Development
When the Models-3 design effort began in 1992, there were no existing emission modeling and
processing systems with the attributes necessary for full integration. The SMOKE was subsequently
developed by the MCNC-North Carolina Supercomputing Center, Environmental Sciences Division,
with U.S. EPA cooperation and support, for the purpose of being able to operate as a module within the
Models-3 framework. The SMOKE is designed to be applicable to any air pollutant, and was originally
applied to ozone precursor data in support of ozone modeling. The decision to develop and use SMOKE
was based on an early recognition that the matrix computation paradigm in a Fortran program for
modeling point, mobile, area, and biogenic emissions, is much faster than earlier sequential processing
programs, i.e., one kind of operation at a time (area source processing, point source, etc.), that rely on
SAS® software.2 This is true despite the fact that the current version of SMOKE, like its predecessor,
the Models-3 Emission Processing and Projection System (MEPPS), uses Mobile 5b to model hourly
mobile emission data3, and the Biogenic Emission Inventory System, Version 2, to model hourly
biogenic emission data4. For earlier releases of Models-3, a decision was made to modify an existing
processor, the Geocoded Emission Modeling and Projection (GEMAP) system, for reasons of economy
of resources and expediency.5 That system was the basis for the first Models-3 emission processor,
MEPPS.6 During 1998 and 1999, the SMOKE code was modified to fit directly within the Models-3
framework. The SMOKE had evolved since its inception and now provides additional features.7
During 2000, SMOKE was made still more efficient, tested with additional pollutant species (including
particulate emissions), tested within the current Models-3 framework design, and equipped with a
module (Skmreport) for reporting and quality control.
Smoke Tool Functions
The SMOKE requires input in the form of emission data (often annual emission inventories)
and related data which must be in acceptable formats, with files containing a defined spatial grid and
spatial surrogates for area source emission data. Creation of these files is not trivial, and they cannot be
created by SMOKE. Therefore, SMOKE Tool was developed to provide these functionalities. At this
time SMOKE Tool is coded in the SAS® language, and has been configured to operate from the
ModelsS Study Planner and to provide input files in the required formats to SMOKE. Running from
within Models-3, SMOKE Tool shares data with SMOKE and the Models-3 system by way of
environment variables defined in the Models-3 Study Planner. Figure 1 presents an overview of the
functional location of SMOKE and SMOKE Tool in the Models-3 system.
Using the Models-3 graphical interfaces, a user may define the horizontal and vertical grids of
interest and episode (or case) in the Science Manager, and register data (including metadata) in the
Dataset Manager. All of these items are then available to other system components, including SMOKE.
This approach ensures consistency of basic information between the system components.
SMOKE
In order to use SMOKE in the Models-3 framework, it is necessary to use graphical
representations of a SMOKE "study" and "plans" (subsets of a study) in Study Planner's interactive
-------
interfaces. (Note: an emission study is a user-defined instance of a specific emission situation to be
modeled.) Models-3 is provided with a typical SMOKE study (Figure 2), which is set for a tutorial, and
several plans which address most anticipated uses of SMOKE (listed in Figure 3). This study can be
copied and reused, or reannotated and reused by most users. Each study subset, or plan, is represented by
an interactive graphical interface. Within a plan, individual algorithms or groups of algorithms, are
represented by boxes (nodes), and linked by lines defining data dependencies. Plans have been
constructed which include each core program defined in SMOKE on-line documentation.8 The
capabilities of the SMOKE core programs have been documented previously,5 with exception of
SMOKE Report (Smkreport). SMOKE Report is a post-processor to SMOKE which allows the user to
specify many combinations of emission data for analysis and quality control purposes.
Figure 4 shows an example plan in which the SMOKE programs are graphically represented in
the Models-3 Study Planner. This example shows the Skminven program, which imports emission
inventories into SMOKE. Each plan and the nodes in it are annotated with environment variables which
specify the logical names and physical locations of the needed inputs. Figure 5 shows the window listing
the environment variables in the Smkinven plan. Given appropriate input data, a user need only re-
annotate the logical or physical names of file locations in the links to fit a specific case of interest, and
initiate and control program execution from the interface. Many names and locations will remain the
same and annotations will not need to be changed. Each arrow which links nodes in a plan is annotated to
give the location of the appropriate input and output files. Output files are automatically registered by
the Models-3 Data Set Manager and placed in a standard output directory identified by the convention:
$M3DATA/studies/site_id/study_name/plan_name/run_number/node_name/
$M3DATA is an environment variable representing the location of the Models-3 data directory on a
particular system. Site id is an identifier assigned to Models-3 system by EPA, and the study name is
user defined. The plan and node names may be changed by the user, but there is usually little reason to
do so because they are descriptive of the process included.
SMOKE TOOL
Because SMOKE requires input files in particular formats, the SMOKE Tool was created to
assist the user in preparation of the files for emission modeling runs. The SMOKE Tool is in the SAS
program language, and uses the Arc/Info geographic information system to create gridded files. The
SMOKE Tool performs the following major functions:
• Creates a grid file for SMOKE, in accordance with the user-defined grid and projection
definition taken from Models-3 Science Manager.
• Creates gridded ascii spatial surrogate files, using the grid file definition, which are used
by SMOKE to apply the surrogates to emissions using spatial cross-reference files.
* Imports, checks and provides temporal and spatial cross-reference files (by source
category code) for use in SMOKE.
* Allows import, editing, and modification of source specific chemical profile files, and
grouping (lumping) to match the chemical mechanism being used in the air quality model
in Models-3. Currently, the chemical mechanism lumping in SMOKE Tool supports the
Carbon Bond 4 and RADM2 chemical mechanisms. The SMOKE will accept any species
definitions, whether specific species or lumped species.
* Allows the user to specify the emission inventory files (and geographic subset thereof) to
be used for a given emission modeling run.
* Allows the user to define "packets" of source category and geographically-specific factors
for application in SMOKE to apply controls, future projections, adjustments, and
reactivity tests (changes in specific species).
-------
The general procedure for using SMOKE Tool requires operation both from Models-3 Study
Planner (first) and from the SMOKE Tool SAS windows (second) as follows:
1) It is necessary to first create the grid and spatial surrogate grid files to be used by SMOKE and
SMOKE Tool. The grid definition must be established in the Models-3 Science Manager (see Figure 1).
The nodes and links in the SMOKE Tool plan (Figure 6) must be annotated to refer to the grid name and
to specify the spatial coverages to be used in generating the spatial surrogates by SMOKE Tool. Table 1
lists the environment variables used by the SMOKE Tool plan. Many of these variables are also used in
SMOKE. The top node (grid_def) executes Arc/Info programs to create a grid file in accordance with
the grid definition (referenced by environment variable HGREDNAME). The second node (grid_cov)
executes programs which grid the surrogate spatial coverages for the defined grid in accordance with the
coverage definition file referenced by environment variable COVER_DEF. The final node of the plan
(grid_srg) computes the spatial surrogates from the gridded data generated by the second node.
Environment variable FEATURE_SRG references the surrogate feature table file containing a table of
surrogate-specific features identifying the manner in which the surrogates will be computed. The
coverage definition file and surrogate feature table file, along with a README_data_file, are located in
the $M3DATA/nostudies/gisdb/ directory.
Table 1. Environment variables used by the Models-3 SMOKE Tool gridding plan.
Environment Variable
CENSUS.DATA
EMS_HOME
FEATURE_SRG
HGRIDNAME
Coverage Name
(example RAILROADS)
COVER_DEF
Example Value
$M3DATA/nostudies
/gisdb/census_block_
data
$M3DATA/studies
$M3DATA/nostudies
/gisdb/feature_srg.in
tut_36_c
Y
$M3DATA/nostudies
/gisdb/covdef.in
Description
Census data. Contains housing and population
data used to compute spatial surrogates in
conjunction with the CENSUS GIS coverage.
Work space path
Name of file containing the feature-surrogate
table. Table defines the manner in which spatial
surrogates are computed from GIS coverages.
Grid name (set when plan annotated with grid)
Specifies whether a named GIS coverage is
processed (Y for YES or N for NO - case
insensitive). All spatial surrogates defined in the
feature-surrogate table for the coverages selected
will be generated. Eighteen surrogates are
available for the U.S. and eleven for Canada.
Name of file defining the GIS coverages available
for the spatial surrogate calculation
2) Executing the SMOKE Tool plan generates ASCII format grid and gridded surrogate files for
use by SMOKE and other portions of SMOKE Tool. Consequently, the SMOKE Tool plan must be run
first. In order to create other input files required by SMOKE (emission inventories, cross-reference files,
and packets), it is necessary to run SMOKE Tool through its interactive SAS windows.
3) The SMOKE Tool is accessed through the Models-3 Strategy Manager menu. The initial
SMOKE Tool window presents the choices Emission Mechanism, Emission Inventory, and SMOKE
Input File. Emission Mechanism allows creation or editing of the species profiles and species lumping
(grouping) consistent with the chemical mechanism selected within the Models-3 Community Air
-------
Quality Model (CMAQ), Normally, no changes to the mechanism grouping will be needed.
4) Next, within SMOKE Tool, the Emission Inventory selection is used to import, reformat,
quality control, and export emission data (Figure 7). If a standard annual or other previously quality
controlled inventory is being used, the user need only establish a working directory (using the Setup
menu choice), define" the geographic area (which may be done by referring to the grid using the menu
option under Setup), and import the emission inventory files (usually point, area, and mobile source
emission files) of interest using the import menu selection. Files are placed in Inventory Data Analyzer
(EDA) format during export (export menu selection) before they are used with SMOKE. The IDA files
are made available to the SMOKE input program (Smkinven) by annotating the graphical links in
Smkinven plans. If emissions from more than one country are being used, they must be imported to
separate user-defined subdirectories in the working directory, and then appended together on export (the
user is prompted to append).
5) If the user wishes to define Major and/or Major Elevated (ie. sources for use with Plume in
Grid modeling) points sources for SMOKE, the Stack Splits function in the Emission Inventory window
of SMOKE Tool must be used to create two definition files (a group defnition and file and a stack
definition file) for the SMOKE laypoint program.
6) Non-inventory files, such as spatial, temporal, and speciation cross-reference files needed by
SMOKE are prepared in SMOKE Input File window of SMOKE Tool. SMOKE Input File requires the
user to establish a working directory into which files are imported. Raw data are normally stored and
imported from the $M3DATA/nostudies directory. After import, the files may be quality controlled,
edited and then exported. Importing converts files into SAS data sets. Quality control and editing are
performed on the intermediate SAS data set files. Exporting places the files in the ASCII format
required by SMOKE.
SMOKE also accepts files for projections and control of emissions in the form of packets in the
file with logical name GCNTL (page 9-48 of SMOKE 1.3 draft User's Manual).8 The packets consist of
multiplicative factors in most cases. GCNTL subset files are:
* Adjustment packets allow user-defined modifications to emission inventory values by
creating a file which applies percent factors by source category code and/or geographic
area. Adjustment packets may also applied to emission inventories in the Emission
Inventory section of SMOKE Tool.
* Allowable packets contain county, Standard Industrial Codes (SICs)(not yet supported in
SMOKE), and Source Category Code (SCC)-specific controls, caps and replacement
emission data.
» Control packets contain county and SCC-specific control levels for control efficiency, rule
effectiveness, and rule penetration.
» CTG (Control Technology Guideline controls), such as Maximum Achievable Control
technology (MACT) and Reasonably Available Control Technology (RACT)
* Projection packets use year, source-category-code, and geographically-specific percent
factors (master tables are stored under $M3DATA/nostudies/projection/) to estimate
future year emissions from a base year inventory, typically 1996.
• Reactivity packets (also user-defined) allow specification of individual chemical species
(add, omit or replace) by source category and/or geographic area.
Packets are quality controlled, edited and prepared for export from SMOKE Input file using the
Quality Control screen (Figure 9). Each packet is accessed, checked and edited as necessary, and then
made available for export. The export function prepares the packet files for the SMOKE import
-------
(Smkinven) program, which accesses them by means of annotating a graphical link.
packets, as described in the SMOKE Version 1.3 User Manual,8 are given in Tables 2 and 3,
respectively.
6) Finally, it is necessary to exit SMOKE Tool and Strategy Manager, and re-enter the SMOKE study
and inventory input plan (Smkinven) in Models-3 Study Planner. The Smkinven input links (Figure 4)
must be annotated by clicking on each of the graphical links and defining the physical location of the
input files (click "view" under the logical name of each file).
Table 2. Control Packet Format.
Line
1
2+
3
Columns
A
A
B
C
D
E
F
G
H
I
J
K
L
M
N
A
Description
/CONTROL/
Country/State/County code (Integer)
10-character SCC (Character)
Pollutant ID (Character)
Primary control equipment code (PCEC); -9 or 0 applies to all equipment
(not yet supported in SMOKE)
Control efficiency (1-100 %) (Real)
Rule effectiveness (1-100 %) (Real)
Rule penetration rate (1-100 %) (Real)
Standard Industrial Category (SIC) (Integer) or -9 for missing
Plant ID for point source (Character)
Characteristic 1 (for IDA format, this is Point ED -9) (Character)
Characteristic 2 (for IDA format, this is Stack ED or -9) (Character)
Characteristic 3 (for EDA format, this is segment or -9) (Character)
Characteristic 4 (blank for EDA format) (Character)
Characteristic 5 (blank for EDA format) (Character)
/END/
Matching using PCEC or SIC is not yet supported. The valid key combinations are:
1) County, plant, Character (Char) 1, Char 2, Char 3, Char 4, Char 5, SCC, pollutant
2) County, plant, Charl, Char 2, Char 3, Char 4, Char 5, pollutant
3) County, plant, Charl, Char 2, Char 3, Char 4, pollutant
4) County, plant, Charl, Char 2, Char 3, pollutant
5) County, plant, Charl, Char2, pollutant
6) County, plant, Charl, pollutant
7) County, plant, pollutant
8) County, SCC, pollutant
Reproduced from
best available copy.
PROTECTED UNDER INTERNATIONAL COPYRIGHT
ALL RIGHTS RESERVED
NATIONAL TECHNICAL INFORMATION SERVICE
U.S. DEPARTMENT OF COMMERCE
-------
9) Country/state, SCC, pollutant
10) County, pollutant
11) State, pollutant
Table 3. Projection Packet Format.
Line
1
2+
3
Columns
A
A
B
C
D
A
Description
/PROJECTION <4-digitfrom yearx4-digit to year>
Country/State/County code or 0 (Integer)
10-digit SCC, or left SCC with remaining digits filled in with zeros or 0
(Integer)
Projection factor (Real)
4-digit SIC, or 2-digit with remaining digits filled in with zeros or blank for
missing (Integer)
/END/
The valid key combinations are:
1) Country, state, county, SCC
2) Country, state, SCC
3) Country, state, county
4) Country, state
EMISSION INPUT DATA
The Version 4.1 release of Models-3 with SMOKE 1.3 contains the most recent United States
emission inventories and processing input files. In each case, users may add or substitute input files of
their own. These input files are located in the Models-3 data structure as follows:
$M3DATA/nostudies/l kmjanduse/
/inventories/
/gisdb/
/packets/
/projection/
/smoke_tut/
/ge_dat/
/inventory/
/met/
/run_m3demo/
/tigerdb/
where:
$M3DATA is an environment variable representing the location of the Models-3
dataset directory.
"1 kmjanduse" contains new 1km resolution gridded land cover for North
America for use with the BEIS model in SMOKE.
-------
* "inventories" contains annual national emission inventories including the revised
1990 NET inventory, 1995 revised NET inventory, and 1996 NET inventory
(version 3.12), and an estimated 1995 Canadian inventory, all in IDA format.
Other inventories may be added by users.
• "gisdb" contains the Geographic Information System (GIS) coverages used for
creating emission spatial surrogate files. SMOKE Tool has been modified to
allow users to add coverages and replace or modify a spatial surrogate cross-
reference file.
• "packets" is the location of user-generated adjustment, control, projection,
reactivity and related files used to adjust emissions (SMOKE logical file name
GCNTL).
• "projection" contains the factor files used for future year projections by source
category and geographic area through 2020, based on the recently updated
Emission Growth and Analysis System, Version 4.9
• "smoke_tut" and its subdirectories contain specific pre-prepared input files needed
for the tutorial study supplied with the Models-3 Version 4.1 release. Subdirectory
ge_dat contains related general SMOKE input files such as spatial and temporal
cross-references to source category codes.
« "tigerdb" contains optional coverages of roads for the United States based on the
Tiger data base from the Bureau of the Census. These are not in the current spatial
surrogate assignments.
The 1990 and 1995 inventories for the United States have been updated to include off road
mobile emissions for sulfur dioxide and paniculate matter. The 1996 United States inventory is the 1996
NET Version 3.12 inventory recently released. The emission inventory file formats are EDA.
Temporal profiles, spatial surrogate coverages, and temporal and spatial cross-reference files are those
developed by the U.S. EPA Office of Air Quality Planning and Standards for recent (late 2000 and 2001)
modeling runs using SMOKE, One of the main changes for input data is the change from county level
aggregate land cover (used by SMOKE to model biogenic emissions with Biogenic Emission Inventory
System Version 2) to 1 km spatial resolution gridded land use data for North America. Coverages and
principal features (with the names that need to be used in the system) used as spatial surrogates for the
United States include:
• AGRICULTURE (US)
• AIRPORTS US
* AIRPORTS CANADA
• FHAROAD (US)
• LAND_WATER (US and CAN AD A) (separate features for land and island area)
• CENSUS (TOTHU as the feature for total housing units)
• CENSUS (TOTPOP as the feature for total population)
• ROADS (Major highways, urban primary roads)
• ROADS (Rural primary roads)
« PORTS (US and CANADA)
• RAILROADS (US and CANADA)
• LAND.WATER (Water area)
» URB AN_RURAL (US and CANADA)(feature 1 urban areas, feature 2 rural
areas)
• FOREST (US and CANADA)
• ROADS (US and CANADA) (feature 11 urban primary, roads 12 rural primary,
roads 21 urban secondary, 22 rural secondary)
• CENSUS (URBPOP urban population)
-------
CENSUS (RURPOP, rural population)
COUNTY (county area)
FUTURE PLANS
The next full release of the Models-3 air quality modeling framework will entail a completely
new framework, as well as updates to the modeling components. We will replace the current framework,
which is based on the C programming language, Object Store® database, Orbix® object communicator,
and Galaxy® interface, with a Java-based framework being developed for the Multimedia Integrated
Modeling System (MIMS). In fact, the Models-3 components other than the existing framework will
become one of the initial portions (air medium) of MIMS. A description of MIMS may be found on the
Internet at http://www.epa.gov/asmdnerl/mims/. Non-Java model components including SMOKE and
CMAQ will be incorporated by the use of Java "wrappers" around them. We will replace the SMOKE
Tool functions with other tools in the new framework, but not in time for the next Models-3 release. In
the meantime, SMOKE Tool will remain available in its current form.
Near-term plans for SMOKE include incorporating the Mobile610 mobile source emission model
to replace Mobile 5, and updating the Biogenic Emission Inventory System, Version 2, to Version 3. In
addition, we will add the ability to define major and major elevated point sources within SMOKE (as
opposed to within SMOKE Tool), and the ability to import and use Continuing Emission Monitoring
(CEM) data from large point sources (usually electric utilities).
DISCLAIMER
The paper has been reviewed in accordance with the U.S. Environmental Protection Agency's
peer and administrative review policies and approved for presentation and publication. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
REFERENCES
1. Dennis, R.L., Byan, D.W., Novak, J.H., Galluppi, K.J., Coats, C.J., and Vouk, M.A. "The next
generation of integrated air quality modeling: EPA's Models-3", Atmospheric Environment, 1996,20,
12,1925-1938.
2. Coats, C.J., Jr. and Houyoux, M.R. "Fast Emissions Modeling with the Sparse Matrix Operator
Kernel Emissions Modeling System", Presented at the Emissions Inventory: Key to Planning, Permits,
Compliance and Reporting, Air and Waste Management Association, New Orleans, LA, September
1996.
3. U.S. EPA. User's Guide to Mobile 4.1 (Mobile Source Emission Factor Model) (Chapter 2 updated
1995). U.S. Environmental Protection Agency, Office of Mobile Sources, Ann Arbor, MI, 1991; EPA-
AA-TEB-91-01,354pp.
4. Guenther, A., Geron, C., Pierce, T., Lamb, B., Harley, P., Fall, R. "Natural Emissions of Non-
Methane Volatile Organic Compounds, Carbon Monoxide and Oxides of Nitrogen from North
America", Atmospheric Environment, 2000, 34: 2205-2230.
5. Wilkinson J.G., Loomis, C.F., McNally, D.E., Emigh, R.A., and Tesche, T.W. "Technical
Formulation Document: SARMAP/LMOS Emissions Modeling System (EMS-95)" Final Report AG-
90/TS26 and AG-90/TS27, Prepared for the Lake Michigan Air Directors Consortium and the
California Air Resources Board by Alpine Geophysics, Pittsburgh, PA, 1994.
-------
6. Benjey, W.G., Godowitch, J.M., and Gipson, G.L. Emission Subsystem (Chapter 4), in Science
Algorithms of the EPA Models-3 Community Multiscale Air Quality (CMAQ) Modeling System, U.S.
Environmental Protection Agency, Washington, D.C., 1999: EPA-600/R-99/030, pp. 4-1 - 4-107.
7. Houyoux, M. R. and Vukovich, J.M "Updates to the Sparse Matrix Operator Kernel Emissions
(SMOKE) Modeling "System and Integration with Models-3", Presented at The Emission Inventory:
Regional Strategies for the Future, Air and Waste Management Association,-Raleigh, NC, October,
1999, pp. 430-440.
8. Houyoux, M.R., Vukovich, J.M. and Brandmeyer, I.E., SMOKE User Manual, (on-line document
updated for SMOKE Version 1.3, available at http://envpro.ncsc.org/products/smokeA. MCNC
Environmental Sciences Division, Research Triangle Park, NC, 2001.
9. E.H. Pechan and Associates, Inc. "Economic Growth Analysis System: Version 4 User's Guide" Final
Draft, Prepared for the U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, by E.H. Pechan and Associates, Inc., Durham, NC. 2001.
10. U.S. EPA. User's Guide to Mobile6: Mobile Source Emission Factor Model( Draft), U.S.
Environmental Protection Agency, Office of Air and Radiation, Office of Transportation and Air
Quality, Ann Arbor, MI, 2001, EPA420-D-01-002a, -186-pp
Figure 1. SMOKE and SMOKE Tool functions in Models-3.
Source: Houyoux and Vukovieh, 1999
-------
Figure 2. SMOKE study interface.
Figure 3. SMOKE plan listing
-------
Figure 4. SMOKE inventory input plan.
^m^n^^^p*ixeiHvitjmti,jimfaiiiw-f^t .-ju^.' "iu** j"1"* ', «*»
•-. *'!;
1 1-amM •.,»« i, , *»•,«, M>_ _»« »,.,,«, «Am
8% * " #
'-'
Figure 5. Environment variable screen
-------
Figure 6. Models-3 SMOKE Tool gridding Study
Planner interface.
$f
PI • rJ
»^Ri( :pipii H^rf
*\l" ^
- ';
-'' -,"^-, ,-,.>>,. v,I,«%;47*,.''%^™^!-'; | '"""* ,f": "
"1* «t a ' iii ' i »( -=\ >i, • ^ jjB>^fc SlMnSdtaiA, ** isr • '* *; i '•"^^ '•*•? ' *\}Wl ' f V ' ^v ^ 'fin'"' i' *V * lh < (
«k*»*«y4iM tuaf^k'nJr^.'i^JB « '
Figure 7. Emission Inventory window.
SMOKE Tool Inventory Analyzer V4.1
Setup Import QA/QC Export Analyze Compare Visualize Help
Welcome to the SMOKE Tool Inventory Analyzer. The analyzer has the
processors that can be executed Interactively:
Setup Create/Belete working directory.
Exit system,
1
Import Import data Into working directory
QA/QC Perform QA/QC and correction of data 1n
working directory.
Export Export data from working directory to
standard formats
Analyze Generate statistics
Compare Compare data 1n two working directories
Visualize CIS visualization of emission data.
View grid.
Click on a Menu Bar item to Select a Processor.
-------
Figure 8. SMOKE Input File window.
SAS:AF
SMOKE Tool
Input File Generation V4.1
Setup Import QC Export Help
Welcome to the SMOKE Tool Input Pile Generator. The generator has the
processors that can be executed Interactively:
Setup Create/Delete workspace,
Exit system,
Import Import data Into workspace
QC Perform QC and correction of data 1n
workspace
Export Export data from workspace to standard
format files
Click on a Menu Bar Item to Select a Processor,
Figure 9. Packet QC window.
SftS: AF
Input File Gen - QC Control Factor:
Workspace Path:
Model: POINT AREA
Packet: CTG CONTROL ALLOWABLE REACTIVITY
Action to Perform: Generate QC Report Report Limit- 5000 lines
(Click on Action) Edit —• >
Create a Backup File > All Records
Restore File from Backup
s With Oats
Click on OK to perform action, CANCEL to exit this screen.
|M>MHCT.I,IHII«.iDllTMlllU.*-«T| £****•».••» ™»-B.-K_J.
j OK I t CANCEL |
This screen allows the user to check and correct data 1n the
control factor packet files contained 1n the workspace.
-------
NERL-RTP-AMD-01-033
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-01/027
4. TITLE AND SUBTITLE
William G. Benjey* et al (see title pige)
U.S, EPA
National Exposure Research Ltboratoty
Atmosphenc Modeling Division (MD-80)
Research Trimgle Park. NC 27711
S.REPORT DATE
6.PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
W, G Benjey, M. R. Houyoux, and J. W. Susick
8.PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Atmospheric Modeling Division, NERL
Research Triangle Park, NC 27709
MCNC North Carolina Supercomputing Center
Research Triangle Park, NC 27709
Science Applications International, Inc.
Falls Church, VA 22042
10.PROORAM ELEMENT NO.
11. CONTRACT/GRANT NO.
8D-0155-NANX
68-W-99-002
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Research Triangle Park, NC 27711
13.TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/9
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The U.S. Environmental Protection Agency has implemented Version 1.3 of SMOKE (Sparse Manix Object Kernel Emission) processor
for preparation of area, mobile, point, and biogenic sources emission data within Version 4.1 of the Models-3 air quality modeling framework. The
SMOKE system indudes MOBILE 5b to model on-road emissions and the Biogenic Emission Inventory System, Version 2 (BEIS 2) to model
biogenic emissions. Although SMOKE may be operated from scripts outside of the Models-3 framework, integration within the system allows
automatic registration and tracking of input and output files, close coupling to the Community Multiscale Air Quality (CMAQ), and modeling system
and visualization tools by using the Net CDF I/O API data format convention. Models-3 also includes the SMOKE Tool to assist in preparation of
emission input data for SMOKE. The SMOKE Tool performs basic functions for the user, including 1) import and quality checking of emission
inventory and related data, 2) preparation of user-defined model grids and gridding of emission data and surrogates (SMOKE Tool requires a
Geographic Information System (OIS), 3) preparation of user-defined "packet" files which instruct SMOKE to apply factors including growth,
control, and reactivity analysis factors (individual species may be added or deleted by source and location), and 4) allows the user to edit input files.
Manual preparation of input files outside of the system is possible, but increases the probability of human error and substantially increases the time
needed for data preparation.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTiFlERS/ OPEN ENDED TERMS
c.COSATI
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21.NO. OF PAGES
20. SECURITY CLASS (This Page)
UNCLASSIFIED
22. PRICE
------- |