  SEQ CHAPTER \h \r 1 

Technical Memorandum

Development of MANE-VU Mobile Source Projection Inventories for
SMOKE/MOBILE6 Application

June 2006,    NESCAUM

I.  Background

The MANE-VU Emissions Inventory workgroup chose to utilize in-house
expertise to develop future year mobile source emission estimates for
use in Ozone, PM2.5, and Regional Haze SIP modeling.  Jung-Hun Woo,
NESCAUM, and Megan Schuster, MARAMA compiled all the information
necessary to run Mobile6 with Sparse Matrix Operator Kernel Emissions
(SMOKE) based on the state submitted data. The purpose of this document
is to describe how member states and NESCAUM/MARAMA prepared the mobile
source activity data (i.e. non-link based VMT & Speed), Mobile6 scenario
input files, and accompanying files in SMOKE/MOBILE6 format in support
of future year projection inventory development. 

II. Procedure

2.1. Base year (Y2002) inventory file 

The MANE-VU 2002 base year SMOKE/M6-ready mobile source files served as
a starting point for projection inventory development; hence, the
procedures employed for the development of these files are summarized
here.  E.H Pechan,  a contractor for MANE-VU, compiled MANE-VU 2002
mobile source inventories. The onroad emissions inventory was compiled
from data supplied by the MANE-VU State agencies in the form of onroad
emissions input data or emissions inventories either directly to MANE-VU
or to EPA through their CERR submittal.  States provided information in
one or more of the following ways:  (1) an onroad emission inventory
submittal to EPA, (2) MOBILE6 inputs and VMT data in National Mobile
Inventory Model (NMIM) format to EPA, (3) portions of MOBILE6 inputs or
full MOBILE6 input files and supporting files plus VMT to EPA, or (4)
portions of MOBILE6 inputs or full MOBILE6 input files and supporting
files plus VMT to MANE-VU.  Different procedures were followed in
developing the MANE-VU 2002 onroad emission inventory depending upon how
the data were submitted. All States provided information on 2002 onroad
activity in the form of vehicle miles traveled (VMT) estimates, although
the level of detail of the VMT data varied by State. After integration
of all these data, Pechan compiled onroad mobile source files in a
SMOKE/Mobile6 ready format. 

2.2. Preparing technical guidance to create SMOKE/Mobile6 ready
information

As described above, States had the ability and experience to develop
their mobile source inventory data in NMIM, stand-alone Mobile6, and
emissions inventory format. They, however, were not familiar with
procedures for integrating files and converting them into a model ready
format. To assist the State workgroup in the creation of an inventory in
(as close as possible to) the SMOKE/Mobile6–ready format,
NESCAUM/MARAMA developed a technical guidance document (  HYPERLINK
"http://www.marama.org/visibility/Inventory%20Summary/FutureEmissionsInv
entory.htm" 
http://www.marama.org/visibility/Inventory%20Summary/FutureEmissionsInve
ntory.htm ). In the document, we also explained how State workgroup
could download the year 2002 files (  HYPERLINK "ftp://ftp.marama.org" 
ftp.marama.org ) and update them to create projection inventory files.

In the technical guidance, States were requested to supply NESCAUM and
MARAMA with the data described below, under Required Data.  States also
had the option to supply NESCAUM and MARAMA with the information listed
under Optional Data.

REQUIRED DATA: 

SMOKE/IDA Activity File 

Go to MARAMA FTP site (above), click on Onroad/ Final Smoke Files/
MANEVU_2002_mbinv.txt

Open the MANEVU_2002_mbinv.txt file

Delete all data that is not your state’s. 

Update file with new header and annual VMT by SCC for 2009 

Name file: state_2009_mbinv.txt

Repeat steps ‘b’ through ‘e’ for years 2012 and 2018 

MOBILE6 input scenario files

Go to MARAMA ftp site, click on MANE-VU Onroad/ Final Smoke Files/
MANEVU_2002_SMOKE_M6_InputFiles_12032004.zip
/m6_2002/month/statecounty.in

Provide updated statecounty.in files for each month and year 2009, 2012
and 2018 

Use same naming convention as the 2002 files – statecounty.in

OPTIONAL DATA: 

 If a state wants to, they can update registration data to either 2004
or 2005 

Go to MARAMA FTP site - Onroad/ Final Smoke
Files/MANEVU_2002_SMOKE_M6_ExternalFiles.zip 

As an example, open file 0900002.reg (or any files with extension .reg)

There is a long header followed by the real data.  Real data starts at
the line called REG DIST. (See example file below). 

States should update the registration data by vehicle class and age for
2004 or 2005.

If you update this data and change the name of the file, make sure the
M6 input scenario file (required data from above) includes the correct
file name for registration data (REG DIST). 

2.3. States data submittal and compiling future inventory

As States submitted their projected mobile source files as listed above,
NESCAUM started to integrate Y2018 files and convert them into regional
(i.e. MANE-VU) files. The following subsections summarize the
compilation process for activity input, scenario input files, external
files and other SMOKE related files.

1) Activity input data (VMT/Speed)

- All member states submitted VMT and speed data. There were minor
issues in data formatting but nothing major. 

- DE and MA re-submit VMT file at October, 2005 (MA change 2002 VMT,
too)

- Merge update 2002 MBINV file, ptpro/ptref, scenario input files for CT
and NJ

2) Scenario input files

- VT used the same input file for all the counties. So, only 36 input
files (i.e. 12months * 3 future years) were provided. VT set 050001 as a
reference county (MCREF).

- VT uses January as reference month for August processing. VT has
confirm that they want to change reference month for Apr-Sep as July
instead of January.

- ME submitted all monthly data in a file. NESCAUM split it by month.

- MA submitted only three input files (Avg Summer day, Typ Ozon day, Avg
Winter day) which were submitted to PECHAN to develop base year input
files. NESCAUM created *.in files by copying MANE-VU 2002 base-year *.in
files, then change years into 2009/2012/2018 and the link to the new
external files (LEV2CERT.D, LEV2EVAP.D, LEV2EXH.D).

- CT submitted *.IN files on 9/22/2005. The “> I/M DESC” file path
had been removed.

- RI scenario input files have 044yyc.IN name convention that are
supposed to be 044ccc.IN(yy stands for 2digit year and ccc stands for
3digit county FIPS). They are renamed and populated for 12months.

- Some states(MD) put "REBUILT EFFECTS" after “SCENARIO RECORD”
line. They were moved ahead of it.

- PA put “ANTI-Tamp” records in external files that need to be in
Scenario input files. NESCAUM wrote a Fortran code
(pa_input_4_override.f) and a C-shell script(cp_antitamp_4_pa.csh) to
put those records in scenario input files

- NJ has 2012 year tag in 2009 scenario input files. NESCAUM put them
back to 2009.

- For 2002 scenario input files update (10/27/2006), CT puts 0.07
Rebuild effect, NJ didn't put any. So, I put 0.1 to NJ 2002 input files.

- Putting future year rebuild effect is a modeling decision that needs
to be a constant throughout the region. So NESCAUM didn’t correct that
in file compilation. Modelers’ put 10% for the base year (2002) and
90% for the future years(2009/2012/2018). NYSDEC put 0.9 into year 2009
files and NESCAUM put 0.9 into year 2018 files before SMOKE/Mobile6
processing.

3) External files 

  - VT uses some of NY state external files (e.g. NYLEV2.D, NYL2EXH.D,
NYL2EVAP.D, and NYL2CERT.D)

  - RI didn't provide any future-specific external files. 2002 MANE-VU
M6 external files were used

- MA and ME used the same named files which have identical contents to
each other: LEV2CERT.D, LEV2EVAP.D, and LEV2EXH.D   

- DE, PA, and NJ submitted the same named files which have identical
contents to each other: NLEVNE.D

- DC file produce warnings (but not errors) from SMOKE/Mobile6 test run

 “User Supplied HC IM240 Cutpoint of  0.500 for year   1 and Vehicle
Type   1 is Too Low and has been Reset to 0.80 g/mi.”  Consulted with
DC and get a confirm that it’s ok to use it as-is.

4) Other SMOKE related files 

  - MCREF/MVREF

    Update MANEVU_2002_mcref.txt for Maine because Penobscot county
(23019) was not found from MAINE_pointerfile.xls. Tammy Gould @ Maine
confirmed that Penobscot (019) has to be added where Piscataquis (021)
is. Correct this problem on both MANEVU_2018_mcref.txt and
MAINE_pointerfile.xls. MANEVU_2018_mvref.txt is updated by deleting
non-ref counties in ME.

  - MTPRO/MTREF

  Add CT and NJ's temporal profile/x-ref to 2018 MANEVU files

  

  

III. Sample model run

Even though NESCAUM had not run SMOKE/Mobile6 for any of the complete
projection years, we ran a day worth of each year: 1) to test
compatibility of compiled data with SMOKE/Mobile6 system and 2) to
examine trends in submitted projection data. As seen in the Figure 1 and
Figure 2, VMT increases 10~20% compared to the base year (i.e. year
2002) for the most of states whereas speed remains the same except for
NY and PA. 

Figure 1. VMT(upper) and Speed (lower) for one day (August 1) of year
2002, 2009, 2012, and 2018

Figure 2. NOx(upper) and CO (lower) emissions for one day (August 1) of
year 2002, 2009, 2012, and 2018

Despite VMT increases, NOx and CO emissions decrease dramatically due to
future-year control programs. Emissions decrease a bit more slowly after
2009 given fewer anticipated controls during post-2009 years. 

NOx reductions seen from Figure 2 can be verified in the spatial (by
county) NOx emission maps which were calculated using SMOKE/Mobile6 for
August 1 of each year. As region wide emissions decreased by years,
emissions from high emission areas (e.g. urban centers) decreased, as
well. 

    

Figure 3. Spatial distribution of emissions from SMOKE/Mobile6 (August
1, NOx)

 PAGE   

 PAGE   1 

