1 Introduction

1.1 Overview

MOVES is the U.S. Environmental Protection Agency's (EPA) Motor Vehicle Emission Simulator (OTAQ, 2009, also available at http://www.epa.gov/otaq/models/moves/index.htm and). The purpose of MOVES is to provide an accurate estimate of emissions from mobile sources under a wide range of user-defined conditions. It helps the user answer "what if" questions, such as "How would particulate matter emissions decrease in my state on a typical weekday if truck travel were reduced during rush hour?", or "How does the total hydrocarbon emission rate change if my fleet switches to gasoline from diesel fuel?"

In the modeling process, the user specifies vehicle types, time periods, geographical areas, pollutants, vehicle operating characteristics, and road types to be modeled. The model then performs a series of calculations, which have been carefully developed to accurately reflect vehicle operating processes (such as cold start or extended idle) and provide estimates of bulk emissions or emission rates. Specifying the characteristics of the particular scenario to be modeled is done by creating a Run Specification, or runspec (discussed further in Section 3).

The outputs from MOVES are emissions estimates or emissions factors. These can be used as inputs to larger emissions modeling systems that model many different types of emissions (e.g., stationary point sources, area sources and biogenic sources) and provide results that are used in performing air quality modeling. An example of such systems is the Sparse Matrix Operator Kernel Emissions (SMOKE) model, which is widely used in both the public and private sectors (IE, 2009a). Detailed information about SMOKE is available at http://www.smoke-model.org.

In December 2009, a new version of MOVES was released: the Motor Vehicle Emission Simulator 2010 model (MOVES2010). An important feature of MOVES2010 is that it allows users to choose between (1) the “Inventory” calculation type, which provides emission rates in terms of total quantity of emissions for a given time period; and (2) the “Emission Rate” calculation type, which gives emission rates in terms of grams/mile or grams/vehicle/hour. For large-scale emissions modeling such as that needed for regional- and national-scale air quality modeling projects, it is desirable to use the Emission Rate calculation type, which populates emission rate lookup tables that can then be applied to many times and places, thus reducing the total number of MOVES runs required.

Under Work Assignment (WA) 3-03 of EPA contract EP-D-07-102 to the University of North Carolina at Chapel Hill (UNC), EPA’s Office of Transportation and Air Quality (OTAQ) has tasked UNC and ENVIRON with developing tools to facilitate the process of using MOVES to create emissions estimates appropriate for air quality modeling.

The successful use of the MOVES Emission Rate calculation type requires careful planning and a clear understanding of emission rates calculation in MOVES. To reduce the time and effort required of the user for this process, and to help the user obtain more accurate modeling results, UNC and ENVIRON created a new tool called the SMOKE-MOVES integration tool (Figure 1). This tool consists of a set of scripts that automate the proper use of the Emission Rate calculations for the purpose of estimating mobile-source emissions for air quality (AQ) modeling. The SMOKE-MOVES tool provides three major functions:

1. Meteorological data preprocessor:

· The meteorological data preprocessor program (Met4moves) prepares spatially and temporally averaged temperatures and relative humidity data to set up the meteorological input conditions for MOVES and SMOKE using the Meteorology-Chemistry Interface Processor (MCIP) output files.

2. MOVES model processing:

· The MOVES driver script creates data importer files and the MOVES input file (runspec), which specifies the characteristics of the particular scenario to be modeled.

· The MOVES postprocessing script formats the MOVES emission rate lookup tables for SMOKE.

3. SMOKE model processing:

· The SMOKE postprocessing program (Movesmrg) estimates emissions from on-road mobile sources based on MOVES-based emission rate lookup tables and meteorology data from Met4moves.

· Creates hourly gridded speciated air quality model-ready input files.

· Produces various types of reports for users.

This user’s guide describes the major functions of the SMOKE-MOVES integration tool.

Figure 1. Flow diagram of overall SMOKE-MOVES integration tool.

1.2 Key Concepts and Terminology

Before we provide more information about the tool, this section discusses some key concepts and terminology that will help users in understanding this document.

1.2.1 Reference Counties and Fuel Months

Reference county – The approach for running MOVES for SMOKE relies on the concept of reference counties. These are counties that are used during the creation and use of emission rates to represent a set of similar counties (i.e., inventory counties) called a county group. The purpose of the reference county is to reduce the computational burden of running MOVES on every county in your modeling domain. By using a reference county, the user generates key emission rates for the single county in MOVES and then utilizes these factors to estimate emissions for all counties in the county group through SMOKE. The reference county is modeled at a range of speeds and temperatures to produce emission rate lookup tables (grams/mile or grams/vehicle/hour, depending on mobile emission process). The variables that are assumed to be constant across the county group members (and the reference county) are fuel parameters, fleet age distribution and inspection/maintenance (I/M) programs. The variables that can vary within the county group are vehicle miles traveled (VMT), source type vehicle population, roadway speed, and grid cell temperatures. Determining the reference counties and their respective county groups is a key aspect of utilizing the SMOKE-MOVES tool. It is ideal for the user to create each county group based on the similarity between the county characteristics (e.g., urban and rural) and the meteorological conditions (e.g., temperature and relative humidity). The user should avoid grouping counties that have significantly different meteorological conditions.

Fuel month – The concept of a fuel month is used to indicate when a particular set of fuel properties should be used in a MOVES simulation. Similar to the reference county, the fuel month reduces the computational time of MOVES by using a single month to represent a set of months. To determine the fuel month and which months it corresponds to, the user should review the State-provided fuel supply data in the MOVES database for each reference county. If the fuel supply data change throughout the year, then group the months by fuel parameters. For example, if the grams/mile exhaust emission rates in January are identical to February’s rates for a given reference county, then use a single fuel month to represent January and February. In other words, only one of the months needs to be modeled through MOVES.

1.2.2 Emissions Processes

When the MOVES model runs as a part of the SMOKE-MOVES tool, it runs for all emissions processes (or modes), including on-road and off-network emissions processes, for the selected pollutants. Off-network emission processes (e.g., parked engine-off, engine starts, and idling, and fuel vapor venting) in MOVES are hour-dependent due to vehicle activity assumptions built into the MOVES model; the emission rate depends on both hour of the day and temperature. On-roadway emission processes (e.g., running exhaust, crankcase running exhaust, brake wear, tire wear, and on-road evaporative), on the other hand, do not depend on hour. In MOVES, these emission processes are categorized into three major groups:

rate-per-distance (RPD) – The emission rate of vehicles on-network (i.e., driving) from MOVES. The rate is expressed in grams/mile traveled.

rate-per-vehicle (RPV) – The emission rate of vehicles off-network (e.g., idling, starts, refueling, parked) from MOVES. The rate is given in grams/vehicle/hour.

rate-per-profile (RPP) – The emission rate of vehicles off-network—specifically, the evaporation from parked vehicles (vapor-venting emissions) from MOVES. The rate is expressed in grams/vehicle/hour.

1.3 Level of Knowledge Expected

This SMOKE-MOVES integration tool user’s guide assumes that readers have some experiences using both the MOVES model and the SMOKE model; this type of knowledge is necessary for effective use of the SMOKE-MOVES tool. For detailed information about these models, please check their user’s guides (OTAQ, 2009; IE, 2009a).

2 Meteorological Preprocessor: Met4moves

2.1 Preprocessor Description

2.1.1 Met4moves Overview

Met4moves is a meteorological preprocessor that prepares temperature and relative humidity (RH) data for use by both MOVES and SMOKE (see Figure 2). Met4moves produces specific meteorological metrics for the reference county(s) for MOVES and additional meteorological metrics for all inventory counties in the county group for SMOKE. The meteorological metrics are specific to the emissions processes (RPD, RPV, or RPP); this is discussed in more detail below.

The inputs for Met4moves include the reference county cross-reference file (MCXREF), the reference county fuel month cross-reference file (MFMREF) for mapping reference counties to fuels and months, spatial surrogates used to identify grid cells per county (SRGLIST), a list of counties (COSTCY), the grid description (GRIDDESC), and gridded hourly temperatures output from MCIP (ORD, 2009). More information about the inputs for Met4moves is available in Section 2.3.1.

The outputs are (1) a file for MOVES that contains the temperatures and RH for each reference county, and (2) a file for SMOKE that contains the temperatures and RH for each inventory county in the county group(s), supplementing the gridded hourly temperatures from MCIP. More specifically:

· For MOVES, Met4moves creates datasets that provide all minimum/maximum (min/max) temperature combinations for a reference county, reflecting all min/max combinations for all inventory counties in that county group. The associated RH values are also included in these datasets. In addition, the datasets include the 24-hour diurnal profiles needed for the RPP emission process, and contain user-specified temperature increments for use by MOVES. This is discussed in more detail at Section 2.3.3.1.

· For SMOKE, Met4moves creates datasets that contain the min/max temperatures, and averaged RH associated for each inventory county. More detailed information is available in Section 2.3.3.2.

The key difference between the datasets provided to MOVES and those provided to SMOKE is that the former includes only the reference counties, while the latter includes all of the inventory counties.

Figure 2. Flow diagram of Met4moves

2.1.2 Reference County Mapping

As described in Section 1.1.1, the SMOKE-MOVES approach is based on reference counties, each one representing a county group that shares the same fuel parameters, fleet age distribution, I/M programs, and meteorological conditions. The use of reference counties allows reductions in the MOVES processing time and in the sizes of the emission rate lookup tables output by MOVES. The reference county cross-reference file (MCXREF) defines the reference counties, and the county group that maps to each reference county. Table 1 provides the format for this input file, while Table 2 gives an example of reference county cross-reference file entries (both of these tables are shown later in Section 2.3.1.4).

2.1.3 Spatial Averaging Method for Temperature and RH Data

In order to combine gridded meteorology data with county-based mobile emissions data, a technique has been developed for calculating spatially averaged meteorology for each reference county. Because not all of the grid cells in a county contain on-road mobile emissions, this technique provides a way to select which cells should be used in determining the min/max temperatures and 24-hour temperature profiles. Spatial surrogates (see IE, 2009b) are used to select grid cells in each county group where the mobile emissions are located. Based on the definitions of reference counties and county groups, and the specific spatial surrogates, Met4moves selects the appropriate meteorology grid cells across the county groups. It uses the temperatures in the selected grid cells to create the temperature input files for both the MOVES model (using a process described in Section 3) and the SMOKE model (using a process described in Section 4). Details of defining a subset of county grid cells for MOVES modeling include the following:

· The user must specify at least one spatial surrogate for Met4moves processing.

· Using more than one surrogate could provide a proxy for grid cells with higher mobile emission activities.

· To select all grid cells within the county, the user could select a total land area surrogate.

· The grid cells selected will likely vary depending on the choice of spatial surrogate(s).

· Only the selected grid cells are used to estimate the min/max temperatures and 24-hour temperature profiles.

· When selecting the absolute min/max temperatures in any hour that are needed for the reference county, Met4moves considers all the selected grid cells in the reference county and all the selected grid cells from inventory counties in the county group. This approach is needed because the reference county could have a smaller temperature range than one of the counties that is mapped to it. The absolute min/max temperatures are used for computing RPD and RPV emissions processes.

· When calculating the diurnal temperature 24-hour profiles for the reference county, Met4moves considers all the selected grid cells in the reference county and all the selected grid cells from inventory counties in the county group. This approach is needed because the reference county could have a smaller temperature range than one of the counties that is mapped to it. The 24-hour diurnal shape profiles are used for computing various 24-hour diurnal profiles based on min/max temperature combinations for the RPP emissions process.

2.1.4 Temporal Averaging Method for County Groups

Met4moves uses hourly min/max temperatures and averaged RH over the spatial region that includes all of the inventory counties in a county group over the user-defined modeling period. The current version of Met4moves supports only the monthly averaging method (versus daily or episodic) to create min/max temperatures and averaged RH for all inventory counties in the county group(s). This means, for example, that if you process an entire year using the monthly averaging method, then Met4moves will produce 12 calendar months of min/max temperatures and averaged RH for all of the inventory counties in the county group for SMOKE.

When computing RH, Met4moves defaults to using only the hours from 6 AM through 6 PM, in order to exclude hours with little traffic that would artificially skew the values. Users can override the default and change the hours of the day to use for this calculation, if desired. Detailed information for this setting is available in Section 2.3.2.

2.1.5 Temporal Optimization for Reference County Fuel Month Groups

As described in Section 1.1.1, the concept of a fuel month is used to indicate when a particular set of fuel properties should be used in a MOVES simulation. To group months by fuel properties, the user must create and input a fuel month file (MFMREF) to Met4moves. The fuel month file is a text file that contains reference county FIPS codes, monthly fuel type identification (ID) codes, and the months that use each fuel type (Tables 3 and 4, shown later in Section 2.3.1.5). If a fuel month file containing more than one fuel month entry is provided to Met4moves, fuel-month-specific temperature outputs will be created for the MOVES model. For example, if a reference county has four fuel months representing the entire year with the monthly averaging method, then Met4moves will produce four sets of temperatures and averaged RH outputs for the reference county, as opposed to 12 calendar months of outputs for the county group. The outputs for the reference county are used as input to the MOVES driver script (Section 3), while the outputs for the county group are used as input for SMOKE processing (Section 4).

2.1.6 Temperature Increments

Temperature increments are used by MOVES to determine the number of emission rates needed in the various lookup tables. The user can define three different temperature increments, which control the RPD, RPV, and RPP emissions processes, respectively. MOVES will calculate emission rates at the various temperatures (determined by the temperature increment) and bounded by the range of absolute min/max temperatures. This provides some control over the number of MOVES runs. Note that all temperatures produced by Met4moves are in ºF.

Examples: If the absolute min/max for an averaging period and reference county is 68/94 ºF, the temperatures associated with the RPD, RPV, and RPP calculations are as follows:

a. For RPD, a temperature increment of 5 ºF would require emission rates at all temperature from 65 ºF to 95 ºF in 5º increments.

b. For RPV, a temperature increment of 10 ºF would require emission rates for each hour at 60, 70, 80, 90, and 100 ºF. (Note that these are needed for each hour because RPV depends on hour of day as well as temperature.)

c. For RPP, with a temperature increment of 10 ºF, Met4moves will create a set of 24-hour temperature profiles based on the normalized 24-hour shape profile. This set of profiles will cover all combinations of min/max values within the absolute min/max range. In this example, the set of profiles (min, max) are: (60, 100), (70, 100), (80, 100), (90, 100), (100,100), (60, 90), (70, 90), (80, 90), (90,90), (60, 80), (70, 80), (80,80), (60, 70), (70,70), and (60, 60) ºF.

2.2 Met4moves Processing

2.2.1 Overview

Using the specified county groups and temporal averaging approach for temperature and RH data (Section 2.1.4), Met4moves determines the min/max grid cell temperatures and associated RH for both SMOKE and MOVES, and computes average 24-hour temperature profiles for use in MOVES.

The 24-hour temperature profiles are averaged over a user-specified time period and grid cells for the reference county. Each profile is assigned to a profile ID code that identifies the combination of minimum and maximum temperatures. Note that there could be several temperature profile IDs used by the MOVES driver scripts (discussed in Section 3) for a single iteration of MOVES.

2.2.2 Met4moves Processing Sequence

Met4moves must be run on a Linux / Unix computer after running MCIP and before running MOVES and SMOKE.

The following are the major processing steps that Met4moves performs:

1. Read the reference county cross-reference file (MCXREF) that contains a list of reference counties and the county groups that map to those reference counties.

2. Read the surrogate description file (SRGDESC) and a list of associated spatial surrogate(s) chosen for use in selecting grid cells.

3. Determine a list of grid cells for each county. Only the selected grid cells are used to estimate the min/max temperatures, 24-hour temperature profiles, and RH over the user-specified modeling period.

4. Set the dates of the modeling episode in local time using the flags STDATE and ENDATE.

5. Determine the averaging method (AVERAGING_METHOD) chosen by the user to create 24-hour temperature profiles (i.e., MONTHLY).

6. Determine the fuel month for the reference county using the MFMREF input file.

7. Read the country/state/county (COSTCY) file to define the time zones for county groups.

8. Read the meteorology data that have been processed by MCIP.

9. Calculate the min/max temperatures hourly and over the modeling period.

10. Calculate average RH for the specified hour range over the modeling period. The default hour range is from 6 AM to 6 PM local time).

11. Once min/max temperatures and averaged RH are estimated for all reference counties and all inventory counties in the county groups, estimate diurnal 24-hour temperature profiles for use by the MOVES driver script (Section 3). The result is a normalized 24-hour shape profile over the user-specified period or fuel month.

2.3 Files and Environment Variables for Met4moves

2.3.1 Input Files

Met4moves requires several input files (Figure 3). This includes the reference county cross-reference file (MCXREF), the reference county fuel month cross-reference file (MFMREF) for mapping reference counties to fuels and months, spatial surrogates used to identify grid cells per county (SRGLIST), a list of counties (COSTCY), the grid description (GRIDDESC), and gridded hourly temperatures MCIP output.

Figure 3. A flow diagram of Met4moves input and output files.

2.3.1.1 GRIDDESC: Grid Description File

The GRIDDESC input file describes the modeling grid. See the SMOKE user’s manual (http://www.smoke-model.org) for details.

2.3.1.2 SRGDESC: Surrogate Description File

The SRGDESC input file includes a surrogate list, description, and surrogate file name used in the modeling grid. Note: The user must set SRGPRO_PATH to define the location of the spatial surrogate coefficient file(s). See the SMOKE user’s manual for details.

2.3.1.3 METLIST: Spatial Surrogate File

The METLIST input file contains a list of MCIP meteorology files, including their full paths.

2.3.1.4 MCXREF: Reference county cross-reference File

The MCXREF input file defines the reference counties, and the county group that maps to each reference county. In defining the reference counties and county groups, the following guiding principles should be taken into account:

· The members of the county group should have the same fuel characteristics, the same distribution of fuels over the year, the same I/M programs, and the same fleet age distribution.

· Since RH is averaged over a county group, grouping counties with reasonably similar daytime RH is advisable.

· Since min/max temperature and diurnal temperature profiles are calculated over a county group, grouping counties with reasonably similar temperature ranges is advisable. Optionally, the shape of the diurnal temperature distribution can be considered for defining county groups. The shape of the diurnal temperature profile is created based on intersections with all inventory counties in the county group.

· Optionally, the ratio of VMT to vehicle population can be considered in the definition of county groups, since the ratio affects the off-network emissions factors. This could be a minor factor in the county grouping, but it would be incomplete not to mention it.

The MCXREF file is a comma-separated-values (CSV) file. Table 1 provides the file format, and Table 2 gives an example set of entries for this file. The user can either include or exclude leading zeroes. For example, California could be represented by ‘06’ or ‘6’.

Table 1. Format of reference county cross-reference file (MCXREF)

Line	Col	Description
1+	A	One-digit country FIPS code for inventory county (Integer)
	B	Two-digit state FIPS code for inventory county (Integer)
	C	Three-digit county FIPS code for inventory county (Integer)
	D	One-digit country FIPS code for reference county (Integer)
	E	Two-digit state FIPS code for reference county (Integer)
	F	Three-digit county FIPS code for reference county (Integer)

Table 2. Example of reference county cross-reference file entries

County Groups			Reference Counties
Country	State	County	Country	State	County
0	13	101	0	13	121
0	13	102	0	13	121
0	13	103	0	13	121
…	…	…	…	…	…
0	13	121	0	13	121
0	13	123	0	13	217
0	13	125	0	13	217
0	13	127	0	13	217
…	…	…	…	…	…

2.3.1.5 MFMREF: Reference County Fuel Month File

The MFMREF input file serves the purpose of grouping months of the year by fuel parameters. The file specifies representative fuel months (the fuelMonth field in the MOVES database) to assign to the calendar months being simulated (the Month field) for each reference county.

As with MCXREF, MFMREF is a CSV file. Table 3 provides the file format. Table 4 is an example that illustrates a situation in which there are three fuel formulations in a given calendar year to be modeled in SMOKE-MOVES. In this example, reference county 13121 uses the fuel formulation mix in January (1) for modeling months November (11), December (12), January (1) and February (2). The April (4) fuel formulation mix is used for March (3), April (4), and May (5). The June (6) fuel formulation mix is used for simulating June (6), July (7), August (8), September (9), and October (10). The user can either include or exclude leading zeros for the country codes; for example, USA country code ‘0’ could be excluded.

Table 3. Format of reference county fuel month file (MFMREF)

Line	Position	Description
1+	A	Six-digit county-specific code for reference county (Integer)
	fuelMonth	Reference county fuel month (Integer)
	Month	Month (Integer)

Table 4. Example of reference county fuel month file entries

RefCounty	fuelMonth	Month
013121	1	1
013121	1	2
13121	4	3
13121	4	4
13121	4	5
13121	6	6
13121	6	7
13121	6	8
13121	6	9
13121	6	10
13121	1	11
13121	1	12

2.3.2 Input Environment Variables

Environment variables are used to provide settings to Met4moves that control the functioning of the program. These settings are below.

· STDATE: [default: 0]

Sets the overall episode start date; Julian format (YYYYDDD).

· ENDATE: [default: 0]

Sets the overall episode end date; Julian format (YYYYDDD).

· AVERAGING_METHOD: [default: MONTHLY]

Sets averaging method to create 24-hour temperature profiles based on STDATE and ENDATE

o MONTHLY: Average data and profiles for each month within the user-specified modeling episode. MONTHLY is the only allowable method at this time.

· MOVES_OUTFILE: [default: none]

Defines the output filename for MOVES.

· SMOKE_OUTFILE: [default: none]

Defines the output filename for SMOKE.

· PD_TEMP_INCREMENT: [default: 5]

Defines the temperature increment (in ºF) for RPD lookup table (described in Section 3.1.1.2) to create combinations of min/max temperature bins.

· PV_TEMP_INCREMENT: [default: 5]

Defines the temperature increment (in ºF) for RPV lookup table (described in Section 3.1.1.2) to create combinations of min/max temperature bins.

· PP_TEMP_INCREMENT: [default: 10]

Defines the temperature increment (in ºF) for RPP lookup table (described in Section 3.1.1.2) to create combinations of min/max temperature bins for normalized 24-hour diurnal temperature profile.

· RH_STR_HOUR: [default: 60000]

Defines the start hour in local time for average RH over the user-specified modeling episode.

· RH_END_HOUR: [default: 180000]

Defines the end hour in local time for average RH over the user-specified modeling episode.

· SRG_LIST: [default: none]

Specifies the name(s) of the spatial surrogate(s) to be used in selecting the grid cells for the county (example: setenv SRG_LIST ‘100, 230’).

· SRGPRO_PATH: [default: none]

Defines the location of spatial surrogate coefficient files.

· TMPVNAME: [default: TEMP2]

Specifies the variable name for the temperature to extract from MCIP files.

2.3.3 Output Files

Met4moves produces two output files (Figure 3). The main difference between the output files is that the output file created for input to the MOVES driver script (Section 3) includes data for the reference counties only, while the output file created for input to SMOKE includes data for all of the counties within the modeling domain.

2.3.3.1 MOVES_OUTFILE: Output File for MOVES Driver Script

The output file created for use by the MOVES driver script contains the absolute minimum and maximum temperatures and average RH values associated with each reference county. The scope of these min/max temperatures extends across the selected grid cells in the county group associated with that reference county. The min/max temperatures determine the MOVES runs that are needed for generating the RPD and RPV emission rates. The temperature increments listed in the header are used to define the temperature bins used to optimize the MOVES runs (Table 5). In addition, the output file prepared for the MOVES driver script contains sets of diurnal temperature profiles based on combinations of min/max temperature bins for each reference county; these are necessary for the MOVES vapor-venting emissions calculation that is performed for the RPP emissions process.

This output file contains all of the temperature and RH values for all reference counties. If, for example, the duration of the episode is annual, there are four fuel months, and the averaging method is monthly, Met4moves outputs four sets of monthly average RH, min/max temperatures, and 24-hour temperature profiles in local time for all reference counties into one output file. An example of this file is provided in Table 5. When each set of fuel month min/max temperatures begins with the record “min_max” in the temperatureProfileID column, the “Temp1” and “Temp2” fields can be referred to as minimum and maximum temperatures, respectively. The remaining records for the specific reference county and fuel month (the records between this “min_max” and the next “min_max”) are the 24-hour temperature profiles. The profile names, temperatureProfileID, are a combination of the averaging type (M is for monthly), the last Julian date of the averaging period, and an index of the profiles (e.g., M2009180003 is the third monthly profile for the 2009180 averaging period). For a specific reference county and fuel month, the monthly average RH value is identical for all the records.

Table 5. Format of reference county minimum/maximum temperatures, relative humidity, temperature increments, and temperature profiles used as input to MOVES driver script

# DESC Sample Met input file for MOVES Driver script PD_TEMP_INCREMENT 5 PV_TEMP_INCREMENT 5 PP_TEMP_INCREMENT 10
Ref. County	fuelMonth	temperatureProfileID	RH	Temp1 (Min)	Temp2 (Max)	Temp3	…	Temp24
13121	1	min_max	66.82	31.21	89.98
13121	1	M2009120001	66.82	37.80	35.69	34.50	…	41.07
13121	1	M2009120002	66.82	46.50	44.75	43.75	…	49.23
13121	1	M2009120003	66.82	55.20	53.80	53.00	…	57.38
…	…	…	…	…	…	…	…	…
13121	4	min_max	66.21	45.21	90.2
13121	4	M2009180001	66.21	47.88	45.12	44.52	…	51.06
13121	4	M2009180002	66.21	56.51	54.45	53.35	…	59.24
13121	4	M2009180003	66.21	65.94	63.55	63.15	…	67.65
…	…	…	…	…	…	…	…	…

2.3.3.2 SMOKE_OUTFILE: Output File for the SMOKE Model

The output file created for use by the SMOKE model contains county-specific min/max temperatures and averaged RH values in local time for every inventory county and averaging period in the modeling inventory. Table 6 gives an example of this file. SMOKE adjusts the native MCIP time zone (GMT) to local time in order to properly use the Met4moves lookup tables, which are given in local time. This output contains the actual month (Month), the fuel month (fuelMonth) and the Julian date (julianDate) used by MOVES to generate the emission rate lookup tables. If the averaging method is set to monthly, julianDate will contain the last Julian date of the averaging month. There are no 24-hour profiles, since SMOKE does not need them.

Table 6. Format of inventory county-specific minimum/maximum temperatures
and temperature profiles used as input to SMOKE.

Inventory County	fuelMonth	Month	julianDate	RH	Temp (Minimum)	Temp (Maximum)
13001	1	3	2009060	51.1	25.2	65.1
13002	1	3	2009060	55.2	29.1	58.9
13003	1	3	2009060	52.6	21.4	59.3
13005	1	3	2009060	51.7	25.8	62.1
…	…	…	…	…	…	…
13001	4	4	2009090	61.1	44.2	75.1
13002	4	4	2009090	66.6	39.9	63.7
13003	4	4	2009090	61.1	45.1	80.5
13005	4	4	2009090	56.2	46.2	79.5
…	…	…	…	…	…	…

3 MOVES Model Processing

The second major component of the SMOKE-MOVES tool is MOVES model processing. This involves two scripts: The MOVES driver script (Section 3.1) creates data importer files and the MOVES input file (runspec), which specifies the characteristics of the particular scenario to be modeled. The MOVES postprocessing script (Section 3.2) formats the MOVES emission rate lookup tables for SMOKE.

Unlike the Met4moves meteorology preprocessor and the SMOKE modeling system, the MOVES driver script is typically run on Windows XP/Vista/7. However, a version that can run on Linux may become available later in 2010 After you finish running Met4moves (Section 2) on Linux, you must copy the Met4moves output file to system on which you are running MOVES (typically a Windows computer). Next, run the MOVES driver script to perform the MOVES runs and the postprocessing script to process the data output from MOVES on that system to create the MOVES-based lookup tables. These tables will be used as inputs to SMOKE that runs on Linux.

When you run the MOVES driver script, it generates the MOVES driver files (both run specifications and data importers) based on the Met4moves output temperature list by reference county, as follows:

Reads the output file from Met4moves and two other user input files.
Assembles instructions for MOVES to create MySQL input databases from XML files (data importer).

Assembles run specification (runspec) XML files to run MOVES for a wide range of conditions.
Generates the run-specific temperature and humidity CSV file.
Assembles a batch list of data importer files, runspec files, and also a list of the MySQL. output database names to be postprocessed.

Once the MOVES driver script has completed, the postprocessor script extracts the emissions factor tables from the MOVES databases; maps MOVES source, fuel, and road types to Source Classification Codes (SCCs); and formats the emissions factor tables for use as SMOKE inputs, as follows:

Reads the list of databases created by the MOVES driver script.
Maps from the MOVES source types, fuel types, and road types to SCC, and aggregates SCC emissions factors over model years.
Performs operations to drop and add fields and reduce the database size by performing a cross-tab query that places all pollutants on a single row.
Writes final processed lookup tables to CSV files that are formatted for SMOKE.

3.1 MOVES Driver Script

The MOVES driver script, which is written in Perl, generates outputs for each reference county. The goals of the driver script are (1) to create the MOVES input files called “runspecs”; (2) to create all required MOVES data importer files, which are the instructions to MOVES on how to build county-level MySQL input databases; and (3) to create batch files of MOVES commands that run MOVES from the Windows command prompt. When the user launches the importer batch file though the Windows command prompt, the MOVES model imports data files into MySQL tables, after which the user needs to launch the runspec batch file so that MOVES calculates emission rates for all the conditions specified in the runspecs.

This section of the user’s guide describes how to coordinate the use of input data from Met4moves and raw MOVES inputs (e.g., prepared by States or EPA) to create and organize the runspec files, and how to choose some user selections (e.g., calendar year, pollutants of interest) and modify hard-coded features (e.g., inclusion of all required emission processes for a given pollutant).

The MOVES model can be run at any of the three domains/scales: national, county, or project. Refer to the MOVES2010 user’s guide for detailed descriptions of the different scales. The SMOKE-MOVES tool always uses the county domain/scale because this level of model detail is required by EPA for SIP and conformity analyses. For this scale, MOVES requires a MySQL input database containing local data for a single county. Every runspec file contains the name of a MySQL input database.

The MOVES driver script requires the following four types of inputs:

1. A run control file containing basic user selections for MOVES (example format in Appendix A)

2. Meteorological outputs from the Met4moves preprocessor for each reference county and fuel month:

a. Minimum and maximum temperatures

b. Average daytime relative humidity

c. Multiple sets of 24-hour temperature profiles

d. Temperature increments

3. Reference county file indicating the filenames by file type keyword for each of the files specified in item 4 (see example format in Appendix A).

4. Reference county inputs in CSV files formatted for the County Data Manager in MOVES that specify:

a. Vehicle age distribution

b. Fuels (parameters and market shares, by month[1])

c. I/M programs

d. VMT

e. Vehicle population

Using the MOVES driver script involves two major processing steps: creation of the runspec files (Section 3.1.1), and user preparation of the data inputs needed for using the driver script to assemble the MySQL input data tables (Section 3.1.2). After the driver script approach is explained, Section 3.2 describes the MOVES postprocessing script that is used with output from a MOVES batch file run to generate the SMOKE-ready lookup table input files.

3.1.1 Creation of Runspec Files

The first action of the driver script is to create runspec files for MOVES that contain all of the selections required to execute a run. The MOVES driver script requires as input two files: a run control file in which the user specifies calendar year, pollutants, and day type (weekday or weekend); and the Met4moves output temperature file for the batch run of MOVES. The user selections in the run control file determine the information printed by the driver script in runspec files for MOVES.

3.1.1.1 Specifying Pollutants

The user must specify in the run control file one or more groups of pollutants to model. There are four groups available: ozone precursors, toxics, PM, and greenhouse gases (GHG). The MOVES pollutants and pollutant groups for the four types of modeling are provided below in Table 7. The choice of pollutant group(s) determines what pollutants are included in the three emission rate lookup tables output by MOVES (RPD, RPV, and RPP), which are described in Section 3.3. The letter ‘X’ marks the key pollutants for inclusion, and a ‘d’ signifies that the pollutant is included in the MOVES run because a key pollutant depends on it.

Table 7. MOVES Pollutants Available for Inclusion in the Lookup Tables Output by MOVES

pollutantID	pollutantName	Pollutant group
pollutantID	pollutantName	Ozone	Toxics	PM	GHG
1	Total Gaseous Hydrocarbons	d	d	d
79	Non-Methane Hydrocarbons	d	d	d
80	Non-Methane Organic Gases	d	d	d
86	Total Organic Gases	X	X	X
87	Volatile Organic Compounds	X	X	X
2	Carbon Monoxide (CO)	X
3	Oxides of Nitrogen	X		X
30	Ammonia (NH3)			X
32	Nitrogen Oxide	X		X
33	Nitrogen Dioxide	X		X
31	Sulfur Dioxide (SO2)			X
100	Primary Exhaust PM10 – Total		d	X
101	Primary PM10 - Organic Carbon		d	X
102	Primary PM10 - Elemental Carbon		d	X
105	Primary PM10 - Sulfate Particulate		d	X
106	Primary PM10 - Brakewear Particulate			X
107	Primary PM10 - Tirewear Particulate			X
110	Primary Exhaust PM2.5 - Total			X
111	Primary PM2.5 - Organic Carbon			X
112	Primary PM2.5 - Elemental Carbon			X
115	Primary PM2.5 - Sulfate Particulate			X
116	Primary PM2.5 - Brakewear Particulate			X
117	Primary PM2.5 - Tirewear Particulate			X
91	Total Energy Consumption		d	d	X
92	Petroleum Energy Consumption				X
93	Fossil Fuel Energy Consumption				X
5	Methane (CH4)	d	d	d	X
6	Nitrous Oxide (N2O)				X
90	Atmospheric CO2				X
98	CO2 Equivalent				X
20	Benzene		X	X
21	Ethanol
22	MTBE		X
23	Naphthalene		X
24	1,3-Butadiene		X
25	Formaldehyde		X
26	Acetaldehyde		X
27	Acrolein		X

3.1.1.2 Driver Script Approach for Organizing Emission Processes

When the MOVES driver script creates the runspec files, it includes all emissions processes (or modes) for the selected pollutants. To optimally implement this design, different emission processes are selected in separate runspecs but sent to the same output database of lookup tables. The remainder of this subsection explains the grouping of emissions processes (RPD, RPV, RPP).

For RPD lookup tables: Off-network emission processes in MOVES are hour-dependent due to vehicle activity assumptions built into the MOVES model; the emission rate depends on both hour of the day and temperature. On-roadway emission processes, on the other hand, do not depend on hour. Thus, the total MOVES run time can be reduced by implementing the temperatures of interest in hour spots for the on-road running processes. The MOVES driver script therefore groups together the on-roadway emission processes that occur in a single runspec file. Processes that fall into this category include the following:

Running Exhaust

Crankcase Running Exhaust

Tire Wear

Brake Wear

On-road Evaporative Permeation (roadTypeID=2,3,4,5)

On-road Evaporative Fuel Leaks (roadTypeID=2,3,4,5)

On-road Evaporative Vapor Venting (roadTypeID=2,3,4,5)

These processes have in common that they have no dependence on hour of the day and their emission rates are all written to the RPD lookup table in grams/mile traveled. Due to the lack of dependence on hour of the day, individual temperature bins in the min/max temperature range can be placed in hours 1 through 24 sequentially. For emission rate calculation, MOVES automatically generates the emission rates for all 16 speed bins (see Table 9) at every temperature bin. The temperature increment (#PR_TEMP_INCREMENT) used to divide temperature bins between the minimum and maximum is provided through the header of the Met4moves temperature preprocessor (Table 5). Potentially, there could be more than 24 temperature bins, in which case a second runspec file for these processes would be created.

For RPV lookup tables: The second grouping of emission processes occur from parked cars. These are run together in a MOVES runspec file for the same reference county. The emission processes in this grouping are listed below.

Start Exhaust

Crankcase Start Exhaust

Off-network Evaporative Permeation (roadTypeID=1)

Off-network Evaporative Fuel Leaks (roadTypeID=1)

Crankcase Extended Idle Exhaust

Extended Idle Exhaust

The MOVES emission rates for these processes are all written to the RPV lookup table and have units of grams/vehicle/hour. The rates depend on hour for various reasons. Start exhaust depends on engine soak time, which varies by hour. Additionally, the lookup table hourly emission rates already incorporate an assumption about the number of starts per vehicle by hour. Similarly, for off-network evaporative emissions and extended idle exhaust, hourly emission rates contain activity assumptions for parked time and idling time activity. As such, this group of emission processes must be run at a single temperature for each of the 24 hours of the day, and the entire range of temperatures will be modeled by using a separate runspec file for each temperature. The number of runspec files equals the number of temperature increments (#PV_TEMP_INCREMENT) (Table 5). MOVES does not calculate these processes at different speeds.

For RPP lookup tables: The third and final group of emissions processes needing its own runspec file for the same reference county includes just one process: off-network evaporative fuel vapor venting. This emission rates for this process are written to the RPP lookup table and have units of grams/vehicle/hour. This emission process is unique in that it depends on the full 24-hour diurnal temperature profile. Temperature inputs for these runspec files are based on a period-specific normalized 24-hour profile shape that is spatially specific to the group of inventory counties mapped to the reference county, as described in Section 2.3.3.1. Sets of 24-hour temperature profiles from Met4moves are used for the vapor venting MOVES runs using the number of temperature increment (#PP_TEMP_INCREMENT). The number of runspec files equals the number of unique temperature profiles per reference county, obtained from Met4moves analysis of the county group for a given averaging period (Table 5).

It is important to note that the MOVES hourly emission rates in all three of the lookup tables (RPD, RPV, RPP) are in local time. When SMOKE uses emission rates from these tables (where hour and temperature are key lookup fields), SMOKE scripts need to account for SMOKE’s meteorology data time zone (GMT) and adjust to the local time of the actual county in order to properly use the lookup tables.

3.1.1.3 Driver Script Generation of MOVES Runspec Batch File

As each runspec file is created by the MOVES driver script, its filename and path are appended to a text file that becomes the batch file of all the runspec files needed for all reference counties. For logistical reasons, the three groups of emissions processes for a reference county are split in the runspec setup portion of the MOVES driver scripts. The first group of emission processes is set up with a different temperature bin increment at each hour. Splitting out this group saves run time compared to running these emission processes in the same runspec file as the second group of emission processes: starts, idle, off-network evaporatives. Vapor venting requires its own third group due to the uniqueness of the temperature inputs—diurnal 24-hour profiles rather than temperature bins in the min/max range. Temperature and speed bins are not the only inputs required to set up a MOVES county domain scale run; other key inputs include vehicle age distribution, fuels, and I/M programs, which are defined based on the reference county. Temperature is one of the tables in the MySQL input database for a county-level MOVES run that changes by runspec. Therefore, each runspec file requires a unique MySQL input database. The other data sources required for inclusion in a MySQL input database are described in Section 3.1.2.

This batch file of all the MOVES runspec files needed for all reference counties resides in the output directory specified in the run control file. The user can review this file before launching the script via the Windows command prompt. However, prior to submitting the MOVES runspec files, the user must first generate the MySQL input databases required for each runspec. This is our next topic of discussion.

3.1.2 Data Sources for Both the MOVES Driver Script and the MOVES Model

This section begins with some background on the data required for a county-level MOVES run. Manually, a user can import county-level data from Excel^® or an ASCII format using the MOVES County Data Manager in the Geographic Bounds panel of the MOVES graphic user interface (GUI). The County Data Manager importer tool transforms these data into a MySQL input database that is named in the MOVES runspec file. As an alternative to using the GUI, a user can manually create an XML file that lists the directory path and filenames of the county-level data. The user can then launch that XML file using a MOVES java command at the Windows command prompt, and the data importer tool included in the MOVES model will build the MySQL input database to be used by MOVES.

The approach of the MOVES driver script is to build the input databases and to create the XML data importer files at the same time it generates the runspec files. The driver script also generates a batch file of the data importer XML files. This batch file must be executed, and the log file reviewed, prior to submitting the MOVES runspec batch file. The XML file identifies the filename and path of required input data (in Excel^® or CSV format). Table 8 lists all of the data types required as input to the MOVES driver script for running MOVES for a reference county. The rightmost column names the sources of the data. Where the data source listed is “Dummy data,” the user does not need to provide this information; the MOVES driver script handles those inputs to MOVES by calling a file that applies for all runs. Meteorology, speeds, vehicle age distribution, fuels, inspection/maintenance (I/M) programs, population, and VMT are discussed individually below.

Table 8. Inputs for MOVES at the County Domain Scale, Emission Rate Calculation

Data Type	MOVES Table Name	Description	Source of Data
Temperature bins	Zonemonthhour	Temperature and relative humidity inputs by hour	Met4moves run by user
Speed bins	Avgspeeddistribution	Speed bin distribution by roadway type and vehicle class	Dummy data
Vehicle age distribution	sourceTypeAgeDistribution	Age distribution by source type over 30 vehicle model years	User
Fuel properties	fuelSupply (required) fuelFormulation (optional)	Fuel properties and their market shares	User
I/M programs	IMcoverage	Inspection & maintenance program inputs	User
Population	sourceTypeYear	Vehicle population by MOVES source type	User
VMT	HPMSVTypeYear	Annual VMT by Highway Performance Monitoring System (HPMS) vehicle type	User
	roadTypeDistribution	Fractions allocating annual VMT to the five MOVES roadway types	Dummy data
	monthVMTFraction	Fractions allocating annual VMT to individual months	Dummy data
	dayVMTFraction	Fractions allocating month VMT to day type (weekday or weekend)	Dummy data
	hourVMTFraction	Fractions allocating day-type VMT to individual hours	Dummy data

The MOVES driver script assembles a text file in the format needed for the County Data Manager import tool in MOVES to create the zonemonthhour MySQL table for the input database. The zonemonthhour table contains temperature and RH data from the Met4moves meteorological preprocessor (Section 2). Relative humidity is a single value averaged over the time period selected by the user in Met4moves (default 6 AM to 6 PM local time) and averaged over the entire group of inventory counties that map to a reference county.

To create the avgspeeddistribution MySQL table, the MOVES driver script assembles dummy inputs in the format needed for the County Data Manager import tool in MOVES. This input is meaningful only for the Inventory calculation type, where the fraction of VMT at each of the 16 speed bins described in Table 9 would need to be specified. For the SMOKE-MOVES tool, however, MOVES is run using the Emission Rate calculation type where, by default, MOVES calculates for every speed bin (Table 9) at every hour.

Table 9. MOVES Default Speed Bins

avgSpeedBinId	avgBinSpeed	avgSpeedBinDesc
1	2.5	Speed< 2.5mph
2	5	2.5mph ≤ speed < 7.5mph
3	10	7.5mph ≤ speed < 12.5mph
4	15	12.5mph ≤ speed < 17.5mph
5	20	17.5mph ≤ speed <22.5mph
6	25	22.5mph ≤ speed < 27.5mph
7	30	27.5mph ≤ speed < 32.5mph
8	35	32.5mph ≤ speed < 37.5mph
9	40	37.5mph ≤ speed < 42.5mph
10	45	42.5mph ≤ speed < 47.5mph
11	50	47.5mph ≤ speed < 52.5mph
12	55	52.5mph ≤ speed < 57.5mph
13	60	57.5mph ≤ speed < 62.5mph
14	65	62.5mph ≤ speed < 67.5mph
15	70	67.5mph ≤ speed < 72.5mph
16	75	72.5mph ≤ speed

The next category of inputs in Table 8 includes the three inputs that define a reference county: vehicle age distribution, fuel properties, and I/M programs. These inputs are expected to be provided by the user in the exact formats discussed below. Tables 10 through 13 list the field descriptions of required user-provided data for a county-level MOVES run.

Table 10. Field Descriptions of MOVES Age Distribution Inputs

MOVES Table Name	Field Heading
sourceTypeAgeDistribution	sourceTypeID
	yearID
	ageID
	ageFraction

Table 11. Field Descriptions of MOVES Fuels Inputs

MOVES Table Name	Field Heading
fuelSupply	countyID
	fuelYearID
	monthGroupID*
	fuelFormulationID
	marketShare
	marketShareCV

*monthGroupID is not currently a supported feature in MOVES

Table 12. Field Descriptions of a Second (Optional) MOVES Fuel Inputs

MOVES Table Name	Field Heading
fuelFormulation	fuelFormulationID
	fuelSubtypeID
	RVP
	sulfurLevel
	ETOHVolume
	MTBEVolume
	ETBEVolume
	TAMEVolume
	aromaticContent
	olefinContent
	benzeneContent
	e200
	e300
	bioDieselEsterVolume
	cetaneIndex
	PAHContent

Table 13. Field Descriptions of MOVES I/M Program Inputs

MOVES Table Name	Field Heading
IMCoverage	polProcessID
	stateID
	countyID
	yearID
	sourcetypeID
	fuelTypeID
	IMProgramID
	inspectFreq
	testStandardsID
	begModelYearID
	endModelYearID
	useIMyn
	complianceFactor

Tables 11 and 12 show the fuel input formats, which are two separate input tables. The fuelSupply table allows users to specify a fuelFormulationID corresponding to a list of nearly 9,000 predefined fuels in MOVES. If the user decides that none of the MOVES fuels describes the local fuel for the reference county of interest, then the user has to modify an existing, or define a new, fuelFormulationID, and list all of the properties (except the last three) shown in the fuelFormulation table fields (Table 12). Users should refer to the MOVES2010 user’s guide for additional information on how to populate these input tables in the format for the County Data Manager importer tool in MOVES. In the fuelSupply table (Table 11), users share the same fuel month in the monthGroupID field for different fuelFormulationIDs with the fuelMonth in the Met4moves output file in Table 5 (Section 2.3.3.1). The same fuel month must be specified in both places.

Tables 14 and 15 list the required fields for population and VMT inputs, respectively. Note that VMT inputs are input by HPMSVType, rather than by MOVES sourceType. Users should refer to Table 3.3 of the MOVES2010 technical guidance document (http://www.epa.gov/otaq/models/moves/420b10023.pdf) to understand the relationship between HPMSVType and sourceType (OTAQ, 2010).

Table 14. Field descriptions of MOVES Population Inputs

MOVES Table Name	Field Heading
sourceTypeYear	yearID
	sourceTypeID
	sourceTypePopulation

Table 15. Field descriptions of MOVES VMT Inputs

MOVES Table Name	Field Heading
HPMSVTypeYear	HPMSVtypeID
	yearID
	HPMSBaseYearVMT
	baseYearOffNetVMT

The remainder of the MOVES input tables in Table 8 relate to VMT fractions by month, day, hour, and road type. All of these are irrelevant to MOVES emission rate outputs as long as they sum to 1, yet these are still required inputs to MOVES. Users do not need to specify the dummy values. The MOVES driver script uses a default file that contains default distributions.

3.2 MOVES Postprocessing Script

Once a MOVES batch file run completes, MOVES will have populated the three output lookup tables. Table 16 lists the field format of the raw tables as they are output by MOVES, prior to any postprocessing steps.

Table 16. Columns Included in the Three MOVES2010
Emission Rate Lookup Tables (RPD, RPV, RPP) before Postprocessing

rateperdistance (grams/mile)	ratepervehicle (grams/vehicle/hour)	rateperprofile (grams/vehicle/hour)
MOVESScenarioID	MOVESScenarioID	MOVESScenarioID
MOVESRunID	MOVESRunID	MOVESRunID
yearID	yearID	temperatureProfileID
monthID	monthID	yearID
dayID	dayID	dayID
hourID	hourID	hourID
linkID	zoneID	pollutantID
pollutantID	pollutantID	processID
processID	processID	sourceTypeID
sourceTypeID	sourceTypeID	fuelTypeID
fuelTypeID	fuelTypeID	modelYearID
modelYearID	modelYearID	Temperature
roadTypeID	Temperature	ratePerProfile
avgSpeedBinID	ratePerVehicle
Temperature
relHumidity
ratePerDistance

Part of the SMOKE-MOVES tool is the MOVES postprocessing script, which transforms the tables output by MOVES into new tables, as described in this section. The modifications to the raw tables are necessary to remove unnecessary information, create fields that are missing, and reduce the overall table size. The postprocessing changes enable SMOKE to more easily use the MOVES emission rate results. The following list shows the input and output table names for this postprocessing:

rateperdistance ® rateperdistance_smoke
ratepervehicle ® ratepervehicle_smoke
rateperprofile ® rateperprofile_smoke

The following postprocessing steps will be performed:

· Create a new field for ‘SCC’ and aggregate emission rates (Section 3.2.1).

· Create a new field for ‘countyID’ (Section 3.2.2).

· Cross-tab pollutantID to reduce output table size. (Section 3.2.3)

· Augment the speciated PM emissions factors to reflect the PM species needed for modeling (Section 3.2.4).

· Create final emissions rate lookup tables (Section 3.2.5).

3.2.1 Postprocessing an SCC Field

The SMOKE model processes county-level emissions by Source Classification Code (SCC). Currently in MOVES2010, output detail by SCC is not available when running MOVES for the Emission Rate calculation tables. For Emission Rate calculation, emission rates are stored in the lookup tables by source type, fuel parameters, road type, and emission process. ENVIRON and EPA OTAQ agreed on an approach for mapping from these parameters to SCC. The approach mirrors the methodology used in the MOVES Inventory calculation with output by SCC. There are 156 different SCCs; they are described in Appendix B.

The MOVES default database (MOVES20091216) table scc decodes SCC as SCCRoadTypeID, SCCVtypeID, and SCCProcID (OTAQ, 2010). For example, the MOVES SCC 223000127X corresponds to SCCRoadTypeID=27 (Urban Principal Arterial), SCCVtypeID=6 (Light Duty Diesel Vehicles), and SCCProcID=X (exhaust emission processes). The MOVES postprocessor script does not use SCCProcID in the construction or of SCCs. Instead “processName” is listed in a separate field from SCC.

Mapping MOVES roadway and emission process to SCC classification requires disaggregation from the 5 MOVES roadway types to 13 SCC roadway types. The fractions for disaggregation are found in the sccroadtypedistribution table in MOVES20091216. Fractions are applied at the county level. An example set of disaggregation factors for Wayne County, Michigan, is presented in Table 17 to illustrate the mapping. Per communication with OTAQ, these factors allocate both the emissions and the activity.

Table 17. Example Mapping of MOVES roadTypeID to SCCRoadTypeID

(Example fractions are specific to countyID 26163.)

roadTypeID	roaddesc	SCCRoadTypeID	SCCRoadTypeDesc	SCCRoadTypeFraction
1	Off-Network	1	Off-Network	1
2	Rural Restricted Access	11	Rural Interstate	1
3	Rural Unrestricted Access	15	Rural Minor Arterial	0.270
		21	Rural Local	0.139
		17	Rural Major Collector	0.520
		13	Rural Principal Arterial	0
		19	Rural Minor Collector	0.072
4	Urban Restricted Access	25	Urban Freeway/Expressway	0.215
4	Urban Restricted Access	23	Urban Interstate	0.785
5	Urban Unrestricted Access	33	Urban Local	0.181
		29	Urban Minor Arterial	0.287
		31	Urban Collector	0.061
		27	Urban Principal Arterial	0.471

Mapping MOVES vehicle types and fuel types to SCCvtypeID requires that the MOVES emission rate output be disaggregated into the 31 vehicle model years in MOVES. Fractions that map MOVES sourceTypeID by model year and fuelTypeID to SCCVtypeID are found in the sccvtypedistribution table of MOVES20091216 (OTAQ, 2010). Table 18 illustrates an example of how source type, model year, and fuel map to SCCVtype.

Table 18. Example mapping of MOVES sourceType and fuelType to SCCVtype
for model year 2000 and 2001 Single Unit Short-Haul Trucks

sourceType ModelYearID	fuelTypeID	fueltypedesc	SCCVtypeID	part5sccvtypedesc	SCCVtypeFraction
522000	1	Gasoline	4	HDGV	1
	2	Diesel Fuel	10	MHDDV	0.757
	2	Diesel Fuel	11	HHDDV	0.243
	5	Placeholder Fuel Type	4	HDGV	1
	9	Electricity	4	HDGV	1
522001	1	Gasoline	4	HDGV	1
	2	Diesel Fuel	10	MHDDV	0.778
	2	Diesel Fuel	11	HHDDV	0.222
	5	Placeholder Fuel Type	4	HDGV	1
	9	Electricity	4	HDGV	1

As noted earlier, SCCVtypeID (Table 18) and SCCroadtypeID (Table 17) are the only components needed to determine the applicable SCC. The postprocessing script maps the SCC value to each emission rate.

After the process of mapping the columns of the emission rate tables to SCCs based on Tables 17 and 18, the lookup table emission rates are listed by SCC and model year. The final step of the SCC mapping portion of the postprocessing script is to aggregate those SCC emission rates over model years using travel fraction. Travel fractions are simply weighting factors that sum to 1 over all model years; they are used to aggregate emission rates over model years to produce a single SCC-wide emission rate instead of a rate for each of the 31 vehicle model years. Travel fractions are based on mileage accumulation by model year and/or age distribution by model year. Travel fractions for the RPD table use Equation (1a), while travel fractions for the RPV and RPP tables use Equation (1b).

Equation (1a): (gram/mile emission rates)

Equation (1b): (gram/vehicle emission rates)

where:

f_travel,MY is the travel fraction by vehicle class for model year MY (sums to 1 over all model years)

p_MY is the fractional population in model year MY from the vehicle age distribution for the reference county

relMAR_MY is the relative mileage accrual for a vehicle of model year MY from the MOVES2010 default table MOVES20091216.sourcetypeage

Equation (2) shows how the travel fractions are then used to weight each model year emission rate to compute emission rates for each vehicle class. The calculation is simply a weighted sum of emission rates over all model years using the travel fractions as the weights.

Equation (2):

where:

EF_veh is the SCC-specific emissions factor of a particular pollutant (grams/mile)

f_travel,MY is the SCC-specific travel fraction for model year MY (sums to 1 over all model years)

EF_MY is the emissions factor for an SCC of model year MY of a particular pollutant, in units of grams/mile or grams/vehicle/hour, depending on lookup table

The relative mileage accumulation rate (Relative MAR) is listed for each MOVES sourceTypeID and ageID in the sourcetypeage table in MOVES20091216. The SourceType age distribution reflects the fraction of the vehicle fleet population in a particular model year. Both Relative MAR and the reference county’s sourcetypeagedistribution will need to be mapped to SCCVtypeID using the procedure illustrated by example in Table 18. Relative MAR and the age distribution for SCCVtype are then used to aggregate all the SCC emission rates over model year. There are 156 different SCCs described in Appendix B.

3.2.2 Creating a County Field

None of the MOVES2010 lookup tables (refer back to Table 16) contain a countyID field solely dedicated to a state-county FIPS code, but SMOKE needs this information. For this reason, the MOVES postprocessing script creates a county field by extracting it from other fields that include county codes as part of their value. The RPD table has a linkID field that contains countyID in the leading five digits followed by more numbers to allow for future implementation of link modeling. The RPV table has a zoneID field that is countyID with a trailing zero. The RPP table has a TemperatureProfileID field, which also leads with countyID and is trailed by other numbers to indicate month and other identification. The postprocessing script parses countyID from linkID, zoneID, and TemperatureProfileID and stores it as a separate field.

3.2.3 Cross-tabbing the pollutantIDs

Each of the three MOVES2010 lookup tables contains a pollutantID field (refer to Table 16), and associated emission rates for each are listed in the ratePerDistance field or ratePerVehicle field (depending on lookup table). This table structure repeats the leftmost fields (e.g., MOVESRunID, yearID, monthID) for each pollutantID, which greatly increases the total number of rows. The cross-tabbing postprocessing step replaces the column for ‘ratePerDistance’ (or ‘ratePerVehicle’, ‘ratePerProfile’) with additional pollutant-specific columns containing values of emission rates. For example, new field headings would include ‘CO_ratePerDistance’, ‘NOX_ratePerDistance’, ‘BENZENE_ratePerDistance’, etc. Performing this cross-tab on the pollutant emission rates significantly reduces the size of the tables for use by SMOKE.

3.2.4 Adjusting PM species Emission Rates for Modeling

The following information developed by OTAQ and OAQPS to speciate the partially speciated PM2.5 exhaust emissions from MOVES2010. The advantage of using this approach over the approach used for speciating total PM2.5 is that it allows the speciated emissions from MOVES; i.e., elemental carbon and particulate sulfate to be retained and only the remainder of the PM2.5 to rely on speciation profiles.

The MOVES pollutants of interest for this section are summarized in Table 19, which was provided by OTAQ. We need to further disaggregate the MOVES species “PM25OM” into the CMAQ model species.

MOVES species are related as follows:

PM25_TOTAL = PM25EC + PM25OM + PSO4

The five CMAQ species Primary or Particulate Organic Carbon (POC), Primary Elemental Carbon (PEC), Primary Nitrate (PNO3), Primary Sulfate (PSO4), and other primary 2.5 particulate matter (PMFINE) also sum to total PM2.5:

PM2.5 = POC+PEC+PNO3+PSO4+PMFINE

Table 19. MOVES PM species

MOVES PollutantId	Data Transfer PollutantCode
101	PM10OM
102	PM10EC
105	PM10SO4
111	PM25OM
112	PM25EC
115	PM25SO4

The purpose of the equations given above is to fully speciate the MOVES2010 partially speciated exhaust PM2.5 to create the model species needed for CMAQ model. These equations are used only when there are any exhaust-related emissions (e.g., Running Exhaust, Start Exhaust, Crankcase Start/Running Exhaust, and Idle Exhaust). The equations utilize the following MOVES 2010 outputs:

· PM25EC, which is identical to the element carbon portion of PM2.5, or PEC

· PM25SO4, which is identical to the sulfate portion of PM2.5, or PSO4

· PM25OM, which contains all components of PM2.5 other than PEC and PSO4(PM25OM=POC+PNO3+PMFINE)

For gasoline vehicles, MOVES applies a temperature adjustment factor that accounts for the impact of cold temperatures on PM25OM and PM25EC with decreasing temperature below 72 °F. At 72 °F or higher, there is no dependency of any component of PM2.5 on temperature. There is also no dependency of any component of PM2.5 on temperature for diesel vehicles. At temperatures lower than 72 °F, the temperature dependence is different for start emissions (including crankcase starts) versus running emissions (including crankcase running).

For gasoline vehicles, the unadjusted PEC is needed to compute the five components of PM2.5 that are not impacted by temperature. We denote unadjusted PEC as: PEC_72.

There are two ways to determine PEC_72:

Run MOVES at 72 °F or higher
Calculate it by “backing out” the temperature adjustment from the adjusted MOVES PEC. This is done by dividing PEC by the MOVES cold-temperature adjustment factor, PEC_T_adj: PEC_72 = PEC/ PEC_T_adj

The option #2 “backing out” is chosen for the SMOKE-MOVE tool to eliminate the need to specify a temperature bin for the MOVES runs that is greater than or equal to 72 °F. To compute PEC_72, the cold temperature adjustment factor (PEC_T_adj) can be calculated based on type of exhaust and temperature (Table 20). For diesel vehicle, PEC_T_adj is constant to 1.0 since there is no dependency of any component of PM2.5 on temperature for diesel vehicles.

Table 20. The MOVES temperature adjustment factors (PEC_T_adj) to PEC.

Vehicle Type	Temperature (°F)	Temperature Adjustment Factor (PEC_T_adj)
Vehicle Type	Temperature (°F)	Start Exhaust / Crankcase Start Exhaust	Running Exhaust / Crankcase Running Exhaust
Gasoline Vehicles	72 °F > T	28.039 * exp(-0.0463*T)	9.871 * exp(-0.0318*T)
Gasoline Vehicles	T ³ 72 °F	1.0	1.0
Diesel Vehicles	All temperature	1.0	1.0

The equations are as follows:

(1) PEC = PM25EC

(2) PSO4 = PM25SO4

(3) PNO3 = (PEC / PEC_T_adj) × FNO3 / FEC

(4) METAL = (PEC / PEC_T_adj) × FMETAL / FEC

(5) NH4 = (PNO3/62.0049 + 2×PSO4/96.0576) × 18.0383

(6) POC = 5/6 × (PM25OM – METAL – NH4 – PNO3)

(7) PMFINE = METAL + NH4 + 0.2 × POC

(8) PMC = (R_PM10-to-PM25-1) × (PMFINE + PEC + POC + PSO4 + PNO3)

where

Gasoline SCCs	All SCCs begin with “2201”
Diesel SCCs	All SCCs begin with “2230”
PEC	Mass of primary elemental carbon, a species needed for CMAQ
PM25EC	Mass of primary elemental carbon provided by the MOVES model
PM25SO4	Mass of primary sulfate provided by the MOVES model
PSO4	Mass of primary sulfate, a species needed for CMAQ
PNO3	Mass of primary nitrate, a species needed for CMAQ
PEC_72	Mass of primary elemental carbon when MOVES runs at 72°F or higher temperature; calculated by “backing out” the temperature adjustment factor, PEC_T_adj
PEC_T_adj	The MOVES cold-temperature adjustment factor to PM25EC from gasoline vehicles.
FNO3, FEC, FMETAL	Percentages of nitrate, elemental carbon, and metal derived from the vehicle-type-specific speciation profile; values are provided in Table 21
METAL	Mass of metal component of PM2.5, which is a component of PMFINE
NH4	Mass of ammonium component of PM2.5, which is a component of PMFINE
62.0049	Molecular weight of nitrate
96.0576	Molecular weight of sulfate
18.0383	Molecular weight of ammonium
POC	Mass of primary organic carbon, a species needed for CMAQ
PM25OM	Mass of organic material provided by the MOVES model;this actually includes more than organic matter: it includes the mass of all components of PM2.5 other than PEC and PSO4
PMFINE	Mass of other primary PM2.5 not accounted for in PEC, POC, PSO4, and PNO3; this is a species needed for CMAQ. This mass includes the ammonium, metals, water, and the mass of the noncarbon material, i.e., hydrogen, oxygen, and other atoms attached to the organic carbon
PMC	Mass of the coarse fraction of PM10; defined as PM10 – PM2.5; this is a species needed for CMAQ
R_PM10-to-PM25	Ratio of PM10-to-PM2.5, which is a constant that is dependent upon fuel type; values are provided in Table 21

Table 21 gives the values for FNO3, FEC and FMETAL and R_{PM10-to-PM25.}They are based on the vehicle type (first seven digits of the SCC), except that R_PM10-to-PM25is based solely on fuel type.

Table 21. Values and basis for fractions used to compute PNO3, METAL, and PMC

3.2.5 Final Postprocessed Emission Rate Lookup Tables

In addition to the changes specified in Section 3.2.1 to 3.2.4, a few minor cosmetic changes (described in this section) are made to each table. The three postprocessed tables named rateperdistance_smoke, ratepervehicle_smoke, and rateperprofile_smoke are exported from MySQL into ASCII files for their use in SMOKE. The fields are listed by lookup table type in Table 22.

Table 22. Fields included in the three MOVES2010 Emission Rate
Lookup Tables after Postprocessing

rateperdistance_smoke (RPD)	ratepervehicle_smoke (RPV)	rateperprofile_smoke (RPP)
(grams/mile)	(grams/vehicle/hour)	(grams/vehicle/hour)
MOVESScenarioID	MOVESScenarioID	MOVESScenarioID
yearID	yearID	yearID
monthID	monthID	monthID
FIPS (countyID)	dayID	dayID
SCC	hourID	hourID
smokeProcID	FIPS (countyID)	FIPS (countyID)

Vehicle Type	SCC list	*FEC (%)*	*FNO3 (%)*	*FMETAL (%)*	*R_PM10-to-PM25*
LDDV	All SCCs that begin with: 2230001, 2230002, 2230003, 2230004, 2230005, 2230006	57.4805	0.2300	0.6513	1.0309
HDDV	All SCCs that begin with: 223007	77.1241	0.1141	0.2757	1.0309
LDGV HDGV	All SCCs that begin with 2201	20.8011	0.1015	2.2256	1.0860