United States       Air and Radiation      EPA420-P-99-010
           Environmental Protection              March 1999
           Agency                    M6.EXH.001
vvEPA     Determination of
           Running Emissions as a
           Function of Mileage for
           1981 -1993 Model Year
           Light-Duty Cars and
           Trucks

           DRAFT
                              > Printed on Recycled Paper

-------
                                                                         EPA420-P-99-010
                                                                               March 1999
             of                               as a               of              for
                                                            and

                               M6.EXH.001

                                   Phil Enns
                                   Ed Glover
                                  Penny Carey
                                  Michael Sklar
                         Assessment and Modeling Division
                             Office of Mobile Sources
                       U.S. Environmental Protection Agency
                                    NOTICE

    This technical report does not necessarily represent final EPA decisions or positions.
It is intended to present technical analysis of issues using data which are currently available.
         The purpose in the release of such reports is to facilitate the exchange of
      technical information and to inform the public of technical developments which
        may form, the basis for a final EPA decision, position, or regulatory action.

-------
                                                             EPA420-P-99-010


                                   - Draft -


                  Determination of Running Emissions
                         as a Function of Mileage
       for 1981-1993 Model Year Light-Duty Cars and Trucks


                      Report Number M6.EXH.001
                             February 8,1999
                                   Phil Enns
                                   Ed Glover
                                  Penny Carey
                                 Michael Sklar

                    U.S. EPA Assessment and Modeling Division
                                    NOTICE

   This technical report does not necessarily represent final EPA decisions or positions. It is intended to present
   technical analysis of issues using data which are currently available. The purpose in the release of suchreports
   is to facilitate the exchange of technical information and to inform the public of technical developments which
   may form the basis for a final EPA decision, position, or regulatory action.
M6.EXH.001                                                            DRAFT 2/8/99

-------
                                                             EPA420-P-99-010


                                   - Draft -


                  Determination of Running Emissions
                         as a Function of Mileage
       for 1981-1993 Model Year Light-Duty Cars and Trucks


                      Report Number M6.EXH.001
                             February 8,1999
                                   Phil Enns
                                   Ed Glover
                                  Penny Carey
                                 Michael Sklar

                    U.S. EPA Assessment and Modeling Division
                                    NOTICE

   This technical report does not necessarily represent final EPA decisions or positions. It is intended to present
   technical analysis of issues using data which are currently available. The purpose in the release of suchreports
   is to facilitate the exchange of technical information and to inform the public of technical developments which
   may form the basis for a final EPA decision, position, or regulatory action.
M6.EXH.001                                                            DRAFT 2/8/99

-------
                                      - Draft -

                      Determination of Running Emissions
                             as a Function of Mileage
            for 1981-1993 Model Year Light-Duty Cars and Trucks

                          Report Number M6.EXH.001

                                  October 19,1998
                           (Tables 3 and 4 revised on 2/8/99)

                                      Phil Enns
                                      Ed Glover
                                    Penny Carey
                                    Michael Sklar

                      U.S.EPA Assessment and Modeling Division

   1.0    INTRODUCTION

          The MOBILE6 emissions inventory model will allocate vehicle exhaust emissions
   between engine start (start emissions) and travel (running emissions). This split allows the
   separate characterization of start and running emissions for correction factors such as fuel
   effects and ambient temperature.  It also enables a more precise weighting of these two
   aspects of exhaust emissions for particular situations such as morning commute, parking lot
   and freeway driving. This document describes the methodology used to calculate the in-use
   deterioration of running emissions and presents estimates for model year 1981-1993 light-
   duty cars and trucks proposed for use in MOBILE6.  The deterioration of start emissions is
   addressed in a separate document.1
          Section 2 describes the Federal Test Procedure (FTP) data sources and the model
   year and technology groups used. Section 3 presents the methodology for calculating
   running emissions from FTP bag data.  It  contains a basic overview of the FTP, defines all of
   the applicable emission terms, and provides the calculations for determining the base unit of
   engine running emissions. Section 4 describes models and results for the in-use
   deterioration of running emissions as a function of mileage. Section 5 reports on high emitter
   correction factors which are applied to the deterioration estimates.  Section 6 displays the
   final results in tabular form.
   Clover, E. and P. Carey, "Determination of Start Emissions as a Function of Mileage and
   Soak Time for 1981-1993 Model Year Light-Duty Vehicles," Report No. M6.STE.003,
   October,  1998.

M6.EXH.001                                -2-                            DRAFT 2/8/99

-------
   2.0    FTP DATA SOURCES USED

          FTP datasets were used to determine in-use deterioration. The FTP-based emission
   estimates were then adjusted by applying high emitter correction factors derived using Ohio
   IM240 data. This section describes the FTP data sources used. Three FTP data sources were
   used: (1) the test results from the EPA laboratory in Ann Arbor, Michigan; (2) data received
   from the American Automobile Manufacturers Association (AAMA) based on testing
   conducted in Michigan and Arizona; and (3) American Petroleum Institute (API) data
   collected in Arizona.  Model years range from 1981 through 1993, and vehicles include both
   cars and trucks. Table 1 gives a breakdown for the light duty vehicle sample by vehicle type,
   model year, and technology for the three datasets combined.
          Most of the 1990 and  later model year vehicle data were supplied by AAMA, while
   most of the pre-1990 data came from EPA laboratory testing. The API sample is a relatively
   small sample (99 cars and trucks).  Its chief appeal is that the vehicles have generally higher
   mileage readings (all over 100,000 miles) than the rest of the sample. There is a general
   trend from carbureted and open loop technologies in early model years to fuel injection in
   more recent years. Port fuel injected vehicles dominate in 1990 and later model years.
   Although not explicitly shown in the table, new catalyst technology was phased slowly into
   the fleet starting in the mid 1980's.
          For this analysis, cars and trucks were each classified into the following model
   year/technology groups:

          MY Group / Technology Type - Cars

                 1988-93 Port Fuel Injection (PFI)
                 1988-93 Throttle Body Injection (TBI)
                 1983-87 Fuel Injection (PFI plus TBI)
                 1986-93 Closed Loop Carbureted/Open Loop
                 1983-85 CL Carb/Open Loop
                 1981-82 FI (PFI plus TBI)
                 1981-82 CL Carb/Open Loop

          MY Group / Technology Type - Trucks

                 1988-93 Port Fuel Injection (PFI)
                 1988-93 Throttle Body Injection (TBI)
                 1981-87 FI (PFI plus TBI)
                 1984-93 Closed Loop Carbureted/Open Loop
                 1981-83 CL Carb/Open Loop
   These groupings were selected on the basis of changes in emission standards or the


M6.EXH.001                                 -3-                            DRAFT 2/8/99

-------
   development/refinement of new fuel metering or catalyst technologies. Because of the
   relatively large amount of 1988-93 fuel injected data, this category was split into PFI
   technology and TBI technology for both cars and trucks.  This produces separate
   deterioration functions based on these fuel delivery technologies and allows the modeling of
   the future penetration of PFI technology into the in-use fleet.
   3.0    DETERMINATION OF RUNNING LA4 EMISSIONS

   3.1    Overview of the Federal Test Procedure (FTP)

          The Federal Test Procedure (FTP) is a test cycle which is used to certify new
   vehicles to emission performance standards.2  The FTP consists of a cold start segment (Bag
   1), a hot stabilized segment (Bag 2), and a hot start segment (Bag 3). Initially, the vehicle is
   stored for a minimum of 12 hours before testing to simulate a 12 hour overnight soak period.
   It is then driven over the cold start segment, which lasts 505 seconds over a length of 3.59
   miles, and the emissions collected as Bag 1. Bag 2 emissions are then immediately collected
   from the hot stabilized segment, which lasts 867 seconds over a length of 3.91 miles. After
   a 10 minute soak, the 505 seconds of the start segment is repeated and the emissions are
   collected as Bag 3.
          The FTP composite emission rate is a weighted combination of the three measured
   bags designed to represent two trips.  The first trip is a cold start after a 12  hour soak, and
   the other is a hot start after a  10 minute soak.  Each trip is a "LA4" cycle, which is a
   combination of the 505 cycle (either Bag 1 or Bag 3) and the Bag 2 cycle.  In a typical FTP
   test, the Bag 2 is only measured once and the results are used for both trips. Since the 505
   cycle is 3.59 miles long and the Bag 2 cycle is 3.91 miles long, each LA4 trip is 7.5  miles
   long. Based on findings about driving activity from the original FTP study, the cold start trip
   is weighted 43% and the hot start trip weighted 57%.  Hence the fraction of vehicle miles
   traveled (VMT) in Bag 1 (containing the cold start) is:

              FTP Bag 1 VMT Weighting = 43%*(3.59 miles / 7.5 miles) = 0.206

   Similarly, since 57% of trips  involve a hot start, the VMT weighting for Bag 3 (containing
   the hot start) is:

              FTP Bag 3 VMT Weighting = 57%*(3.59 miles / 7.5 miles) = 0.273

   The remaining VMT represents stabilized driving (Bag 2).  Since it is used for both the cold
   start and hot start trips, its VMT weighting is computed from both:

          FTP Bag 2 VMT Weighting = (43% + 57%)*(3.91 miles / 7.5 miles) = 0.521
      240 CFRPart 86, SubpartB, Section 86.144

M6.EXH.001                                  -4-                             DRAFT 2/8/99

-------
    Thus, the standard VMT weighting of the bags reported in grams per mile (g/mi) for the full
    FTP is:

                     FTP = (Bag 1*0.206) + (Bag 2*0.521) + (Bag 3*0.273)

    where the fractions represent the proportion of vehicle miles traveled within the three modes
    during the FTP trip in grams per mile.
    3.2    Overview of the Hot Running 505 and Its Use

           The FTP testing method outlined above does not allow the precise separation of start
    and running emissions, since Bags 1 and 3 contain both start and running emissions. Bag 2
    of the FTP does not contain an engine start; however, the driving cycle used in the second
    bag is significantly different from the cycle used for Bags 1 and 3. Thus, to estimate the
    amount of FTP emissions that can be allocated to engine start and running emissions, the
    concept of the Hot Running 505 (FIR505) must be introduced.
           The FIR505 refers to emissions measured from a driving test performed on the 505-
    second cycle of FTP Bags 1 and 3 without an engine start.3  Appending the HR505 cycle to a
    standard three-bag FTP produces values that can be used to estimate the portions of Bags 1
    and 3 attributable to  start emissions following a 12 hour soak and start emissions following a
    10 minute soak, respectively.
           Since the  HR505 has not historically been included in FTP test programs, a method
    of estimating the  FIR505 from FTP bag data was developed using data from a special test
    program. Briefly, HR505 emissions were measured in a sample of 77 cars and trucks tested
    under EPA  contract.  The results from this sample were used to develop a correlation
    between the HR505 and FTP bag data. This correlation was then used to estimate HR505
    results for the larger FTP dataset used in this analysis.

    3.3    Basic Running LA4 Emission Rate

           The LA4  refers to a cycle comprised of the 505-second driving cycle used for Bags 1
    and 3 of the FTP  and the 867-second cycle of Bag 2. Running LA4 emissions are defined as
    emissions from this 1372-second cycle with  no engine start. For the MOBILE6 separation of
    start and running emissions, the running LA4 represents the running portion. For a given
    three-bag FTP, running LA4 emissions can be estimated using a VMT-weighted
    combination of the HR505 and the Bag 2 emissions (stabilized operation). This estimate
    contains all of the driving behavior in the LA4 cycle, without engine starts. Mathematically,
    it is given by:
       3Brzezinski, D. and P. Enns, "The Determination of Hot Running Emissions from FTP
Bag Emissions", Report No. M6.STE.002, December, 1997.

 M6.EXH.001                                 -5-                             DRAFT 2/8/99

-------
             Running LA4
              Emissions     = (HR505*(0.206+0.273)) + (Bag 2 * 0.521)
             (grams/mile)

   where 0.206, 0.273, and 0.521 are the VMT weightings for Bags 1, 3, and 2, respectively.

   Like the FTP, running LA4 emissions are measured in units of grams per mile.  This
   estimate is proposed for use in MOBILE6 as the basic exhaust emission rate from which all
   other running exhaust emission estimates are derived.
          Using the methods described in this section, all emissions measured using the FTP
   and reported by bag can be allocated to start or running emissions before analysis. Average
   running LA4 and FTP emission rate estimates for each model year are shown in Table 2 for
   the light-duty cars and trucks in the EPA-industry sample used in this study.
   4.0    FTP-BASED MODELS OF RUNNING LA4 DETERIORATION
          WITH MILEAGE

          This section describes the methodology EPA used to estimate the deterioration of
   running emissions. Deterioration of running emissions as a function of mileage was
   examined using a number of linear and nonlinear models. The goal was to develop a
   description of deterioration that is consistent with both the available test data and with
   engineering judgment of past and likely future technologies.
          In particular, for the model year/technology subfleets identified above, adequate
   data are often absent in some part of the useful lifetime mileage range. Such data gaps
   raised concerns when trying to fit a single functional form to a given data set, as it
   usually was found that no simple description of deterioration adequately describes the
   full range.  For example, a fitted least squares regression often tends to overestimate
   emissions at low mileage.
          A number of linear and nonlinear models of deterioration were examined. The
   chosen models represent a balance of simplicity and engineering judgment. They take the
   general form of expressing emissions as a piecewise linear function of mileage. At low
   mileage, emissions are assumed to equal the mean level estimated from those vehicles in
   the dataset with less than 20,000 miles of accumulated driving.  This level applies for
   mileages ranging from zero up to the mean mileage for those vehicles. This approach
   was thought to give the best prediction, since the vehicles tested at low mileage should
   not be subject to any recruitment bias influence. The 20,000 mile cutoff is somewhat
   arbitrary, and was developed in coordination with the FACA In-Use Deterioration
   Workgroup.
          At higher mileage, emissions are modeled to deteriorate linearly. While nonlinear
   models were investigated, they did not provide significant improvement over simpler
   linear forms. Two linear functions are used in the final models: (1) a least squares
   regression using all the data that is constrained to pass through the low mileage sample


M6.EXH.001                                -6-                               DRAFT 2/8/99

-------
   means; and (2) a least squares regression using all the data, with no constraints. The
   unconstrained linear model was chosen as the best representation of the data at higher
   mileages. The constrained line was chosen to provide a transition, when needed,
   between the low mileage mean and the high mileage unconstrained regression. The
   connection between these two lines is made based on their relative positions for a given
   technology/model year class. With a few exceptions, the following steps describe the
   calculation of this piecewise linear function.

   1. For each of the model year/technology groups listed in Section 2.0, the mean CO, HC,
   and NOX emissions, and the mean mileage were  computed for all vehicles with an
   odometer reading of less than 20,000 miles. This value is used to model the group's low
   mileage emission level from zero miles up to the mileage determined in step 2. An
   exception occurs when the mean emissions for the entire sample is less than the mean of
   the low mileage subsample, a case that is discussed below.

   2. For each group, an (unconstrained) regression line was estimated for emissions versus
   mileage.

      a. If this line has positive slope and its intercept is less than the low mileage mean
      emissions from (1) above, it defines estimated emissions beginning at the mileage
      where it intersects the low mileage mean.

      b. If the (unconstrained) regression has positive slope but the intercept is greater than
      the low mileage mean, the constrained line defines emissions from the mean of the
      low mileage subsample to the mileage at which it intersects the unconstrained line.
      Beyond that mileage, emissions are estimated by the unconstrained line. Thus, the
      constrained line links the other lines for an intermediate range of mileages.

      c. If the unconstrained regression has negative slope or the mean of the full  sample is
      less than low mileage mean, emissions for all mileages are set equal to the mean
      emissions for the full sample. This assures that negative deterioration cannot occur.

   While these rules do not encompass  all possible  scenarios, they do cover all situations
   arising with the FTP data on which this analysis  is based. The majority of cases are
   covered by option 2(a), giving a simple two-piece function.  The three-piece function of
   2(b) applies to several situations, usually with only a small slope change from the
   constrained to unconstrained line. Finally, the simple horizontal deterioration line of
   option 2(c) is needed for the CO fits of the 1988-93  TBI cars and the NOx fits  of the
   1981-83 carbureted trucks. The underlying numerical estimates are  listed in Table 3.
          For the FTP data set, these rules appear to produce reasonable emission
   projections in most cases.  The two cases  in which the full sample mean is less  than the
   low mileage mean are caused by a few low mileage  outliers.
M6.EXH.001                                -7-                               DRAFT 2/8/99

-------
   5.0    HIGH EMITTER CORRECTION FACTORS

          Since the estimates of running emissions deterioration are based on FTP tests
   obtained from public vehicle recruitment programs, there is some concern that low
   vehicle recruitment acceptance rates (typically less than 25%) in these programs may
   introduce recruitment bias. Whether such bias results in overestimation or
   underestimation of the true emissions deterioration is a matter of debate. This section
   addresses this issue, describes the methodology for adjusting emission factor estimates to
   account for bias, and presents the results.
          Most of the 1990 and later model year vehicle data for this analysis were supplied
   by the domestic automobile manufacturers (the AAMA dataset).  The manufacturers
   have expressed the opinion in FACA meetings and MOBILE6 workshops that owners of
   vehicles experiencing problems would be more likely to respond to the manufacturers'
   recruitment efforts, especially  considering that repairs were included as  an incentive to
   participate.  The AAMA dataset is also composed of vehicles tested when they were
   roughly 2-3 years of age, when gross emitters should be few in number and any
   recruitment bias influence should be minimal.
          Most of the pre-1990 data were collected by EPA; the average age of the vehicles
   was roughly 3-5 years at the time of testing. In this case, tampered vehicles or vehicles
   with problems should be greater in number, but owners may be more reluctant to
   participate in a program run by a regulatory agency, resulting in an underestimation of
   high emitters. The California Air Resources Board (CARB) has tested this hypothesis by
   comparing estimates from its CALEVIFAC emissions inventory model,  which are based
   on surveillance programs similar to those run by EPA, with emissions obtained from a
   California Pilot Project fleet with a high (60%) vehicle capture rate.  In general, the
   comparison  showed that the modeled estimates tend to underestimate emissions in older
   model year vehicles and  slightly overestimate the emissions of newer vehicles. CARB
   developed high emitter adjustment factors (HECFs) for use in its EMFAC model to
   account for these discrepancies.
          EPA developed high emitter correction factors using EVI240 data collected in
   Dayton, Ohio during 1996-97. Like other inspection and maintenance (I/M) data, these
   form a large sample of vehicles within their geographical region, and are considerably
   less subject to sources of bias found in non-mandatory programs. The data and their
   translation to running LA4 estimates are described in more detail in a separate
   document4.
          Because of problems with the Ohio data odometer readings, the data were
   condensed to their mean running LA4 values by age, which then were associated with
   the corresponding region-specific mileage accumulations obtained from 1995
   Nationwide Personal Transportation Survey (NPTS) data. After smoothing these values
   4Enns, P., E. Glover, P. Carey and M. Sklar, "Analysis of Emissions Deterioration
   Using Ohio and Wisconsin EVI240 Data," Report Number M6.EXH.002, October, 1998.

M6.EXH.001                                -8-                               DRAFT 2/8/99

-------
   in the manner required for use in MOBILE6, these points were graphed with the
   emission rates fitted from the FTP data as described in Section 4. For each pollutant and
   within each model year/technology group, the difference between the Ohio mean and
   FTP-based fit was computed. These values were regressed through the origin against
   mileage. (The line was forced through the origin so that at zero miles the difference is
   zero.) Finally, the fitted differences were added to the fitted FTP-based values to obtain
   corrected values.
          In a few model year/technology groups, the Ohio adjustment is negative and,
   when applied to the deterioration line, causes negative deterioration. For these cases,
   deterioration is held equal to zero up to the mileage at which the adjusted emissions
   exceed the low mileage constant level.
           Figures 1 and 2 illustrate how the adjusted and original values compare for each
   model year/technology group as a function of mileage for the car sample. Ninety-five
   percent confidence bands for the unadjusted lines are drawn to help judge the impact of
   the corrections. If the adjusted values fall inside these bands, it suggests that the Ohio
   EVI240 data agrees fairly closely with the FTP data, i.e., bias is not a large problem.
   Otherwise, the recruitment bias is more serious. The graphs show varying levels of
   disagreement between the two data sources. In these graphs, the mileage interval for a
   given set of lines corresponds to the average mileages assigned in the NPTS survey to
   the model years for that group of vehicles. For example, the  1990 to 1993 cars range in
   mileage from about 45,000 to 70,000. Thus, the graphs show line fits for those vehicles
   in that interval.
          Figures 3 to 9 present emission estimates for each model year/technology group
   as a function of mileage both with and without the  high emitter correction factors for
   cars and trucks. For MOBILE6,  deterioration estimates with the high emitter corrections
   will be used.
   6.0    RESULTS

          Results for each vehicle type/model year/technology group are presented in
   Tables 3 and 4. Included are the slopes and intercepts of the constrained and
   unconstrained regression lines, low mileage emissions and mileage intervals for each line
   segment. Table 3, described in Section 4, gives the unadjusted slopes. Applying the
   adjustment factors effectively changes the line segment slopes. The high emitter
   correction factors and the corresponding adjusted slopes are displayed in Table 4.

          Shown below is a sample calculation of running emissions. It illustrates how the
   model coefficients given in Table 4 are used.

   Example:     Calculate HC running emissions for a 1985 model year Fl-equipped car
                 with: a) 15,000 miles, b) 75,000 miles, and c) 125,000 miles.
   From Table 4:
M6.EXH.001                                 -9-                               DRAFT 2/8/99

-------
   a)     At 15,000 miles, mileagesecond corner, therefore:

          Running (g/mile) = ZML + (First Slope * First Corner) +
                            (Second Slope) * (Second Corner - First Corner) +
                            (Third  Slope) * (Mileage - Second Corner)
                          = 0.1479 + (0.0000*18.89) + (0.0078)*(81.38-18.89) +
                            (0.0059)*(125-81.38)
                          = 0.8927 g/mile
M6.EXH.001                                -10-                               DRAFT 2/8/99

-------
               Hgura 1: RUNNING LM: FTP-BASED MOBILES and OHIO IM240 ADJUSTMENTS
                                    1861-83 FUEL INJECTION CARS
       HC (g/ml)
        ao

        24

        2.0

        14

        1.0

        05

        ao
                                     80
                                                               120
                                                   Oeiooo)
       CO 
-------
               Figure 2: RUNNING LM: FTP-BASED
                                    1961-88
         and OHIO
          CARS
                     IM240 ADJUSTMENTS
       HC (gftnQ
        1-;
                                   120
                                                             •MO
                                                                                        •WO
                                                  «cMOO)
       CO (8/ml)
        10
                                                              MO
                                                  tClOOO)
       NOK to/mo
        1.6
        O6
          100
                                                                                        •MO
                         LOWBt8B%CL
                         UPPER MK OL
     I98MCL    —
UPPER MKOL    — '
                      90WCL
                      •BKOL
M6.EXH.001
-12-
                                        DRAFT 2/8/99

-------
       HC (o/mi)
               Figure 3: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                 RUNNING LA4, 1988-88 PR CARS
        0.0
                             SO                100
                                            MLE8 £1000)
                                • •• FTP—BASED   nnn ADJUSTED
                                                                                     200
M6.EXH.001
-13-
DRAFT 2/8/99

-------
       HC (o/mi)
               Figure 4: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                 RUNNING LA4, 1868-83 TBI CARS
        OO
                             SO                 100
                                            MLE8 00000)
                               • •• FTP—BASED   nnn ADJUSTED
                                                                                    200
M6.EXH.001
-14-
DRAFT 2/8/99

-------
       HC (o/mi)
               Figure 5: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                RUNNING LA4. 1986-93 GARB CARS
        0.0
                             SO                100
                                            MLE8 £1000)
                                • •• FTP—BASED   nnn ADJUSTED
                                                                                    200
M6.EXH.001
-15-
DRAFT 2/8/99

-------
       HC (o/mi)
               Figure ft FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                  RUNNING LA4, 1983-87 Fl CARS
                             SO                100
                                            MLE8 £1000)

                                • •• FTP—BASED   nnn ADJUSTED
                                                                                    200
M6.EXH.001
-16-
DRAFT 2/8/99

-------
       HC (a/ml)
               Figure 7: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                RUNNING LA4. 1963-86 GARB CARS
                             SO                 100
                                            MLE8 £1000)

                               • •• FTP—BASED   nnn ADJUSTED
                                                                                    200
M6.EXH.001
-17-
DRAFT 2/8/99

-------
       HC (a/ml)
               Figure 8: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                  RUNNING LA4, 1901-82 R CARS
                             SO
                                               100
                                                0C1000)

                                                nn n ADJUSTED
                                                                                    200
M6.EXH.001
-18-
DRAFT 2/8/99

-------
       HC (a/ml)
               Figure 9: FTP-BASED MOBILES PROJECTIONS and OHIO IM240 ADJUSTMENTS
                                RUNNING LA4, 1981-82 GARB CARS
        0.0
                             SO                100
                                            MLE8 £1000)
                                • •• FTP—BASED   nnn ADJUSTED
                                                                                    200
M6.EXH.001
-19-
DRAFT 2/8/99

-------
                                                    Table 1
                  Distribution of Vehicles by Model Year and Technology for the Combined FTP Dataset

81
82
83
84
85
86
87
88
89
90
91
92
93
TOTAL
CARS
TECHNOLOGY
GARB
657
71
57
30
74
34
17
15
22




977
OPLP
367
71
63
5
24
7
1






538
PFI
29
8
62
35
66
92
106
113
103
250
426
347
366
2,003
TBI
15
74
127
46
56
60
76
69
38
160
91
57
29
898
SUB
TOTAL
1,068
224
309
116
220
193
200
197
163
410
517
404
395
4,416
TRUCKS
TECHNOLOGY
GARB


3
22
30
9







64
OPLP
124
45
8
26
33
14







250
PFI



1
6
41
4


1
144
92
93
382
TBI




13
23
6


144
141
92
90
509
SUB
TOTAL
124
45
11
49
82
87
10


145
285
184
183
1,205
TOTAL
1,192
269
320
165
302
280
210
197
163
555
802
588
578
5,621
M6.EXH.001
-20-
DRAFT 2/8/99

-------
                                                   Table 2
              Mean Running LA4 and FTP Emission Levels by Model Year for Light-Duty Cars and Trucks
                                         for the Combined FTP Dataset


MYR
81
82
83
84
85
86
87
88
89
90
91
92
93

HC_RUN
0.421
0.588
0.230
0.533
0.355
0.759
0.456
0.212
0.152
0.109
0.078
0.094
0.061

HCFTP
0.706
0.789
0.431
0.756
0.533
0.926
0.656
0.406
0.311
0.274
0.237
0.267
0.225
a
CO_RUN
6.489
5.394
2.760
7.622
5.561
8.738
7.005
3.344
2.645
2.087
1.572
2.599
0.977
\RS
COFTP
9.667
8.318
5.073
9.968
6.935
10.432
8.366
4 .574
3.911
3.614
3.145
4.327
2.551

NOX_RUN
0.795
0.750
0.677
0.785
0.687
0.612
0.698
0.564
0.553
0.400
0.353
0.322
0.286

NOFTP
0.897
0.872
0.806
0.893
0.770
0.713
0.789
0.668
0.652
0.633
0.524
0.508
0.466

HC_RUN
0.759
1.163
0.865
0.419
0.923
0.561
0.164


0.163
0.187
0.152
0.137

HCFTP
1.275
1.732
1.361
0.802
1.281
0.823
0.401



0.800

0.420
TRUC
CO_RUN
10.876
8.987
5.759
3.597
8.999
6.248
2.959


2.245
2.228
2.172
1.668
:KS
COFTP
18.158
16 .774
13.225
10.633
14.465
8.789
4.610



9.510

5.363

NOX_RUN
1.662
1.740
1.405
1.387
1.354
1.006
0.531


0.376
0.486
0.469
0.459

NOFTP
1.752
1.732
1.435
1.405
1.388
1.057
0.605



0.885

0.847
M6.EXH.001
-21-
DRAFT 2/8/99

-------
                                                         Table 3
                          Running Emission Deterioration Model Coefficients for HC (Unadjusted)

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82FI
8 1-82 CARS
ZML Mean
Emissions
(gr/m)
0.0516
0.0843
0.1479
0.0815
0.1691
0.1240
0.2108
First
Slope
(gr/m/ 1000m)
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
First
Corner
(1000 miles)
20.03
34.39
14.10
19.83
25.24
11.29
10.18
Second
Slope
(gr/m/ 1000 m)
0.0023
0.0020
0.0079
0.0019
0.0095
0.0038
0.0110
Second
Corner
(1000 miles)
N/A
N/A
81.38
N/A
N/A
70.55
N/A
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
0.0060
N/A
N/A
0.0037
N/A
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARS
81-87 FI
81-83 CARS
0.0932
0.0783
0.2495
0.2927
0.6587
0.0000
0.0000
0.0000
0.0000
0.0000
23.40
16.24
22.03
29.38
15.99
0.0025
0.0043
0.0136
0.0136
0.0110
N/A
55.16
N/A
N/A
N/A
N/A
0.0042
N/A
N/A
N/A
    Note: The first slope is zero, since it is assumed that the ZML emission rate is constant from zero miles to the first corner. For
    the cases with a single corner, the second slope is determined from the unconstrained regression and the corner occurs at the
    mileage where that line intersects the ZML mean emissions. For the case with two corners, the second slope was obtained
    using a regression line constrained to pass through the ZML mean emissions-mileage. The third slope is for the unconstrained
    regression line and applies at mileages above the second corner. (Unadjusted refers to estimates obtained using the FTP
    dataset only.)
M6.EXH.001
-22-
DRAFT 2/8/99

-------
                                                      Table 3 (cont.)
                          Running Emission Deterioration Model Coefficients for CO (Unadjusted)

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82 FI
81-82 CARS
ZML Mean
Emissions
(gr/m)
0.7983
2.5684
2.1416
0.6910
1.0983
1.7270
2.9361
First
Slope
(gr/m/ 1000 m)
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
First
Corner
(1000 miles)
13.78
N/A
14.10
21.13
22.69
16.62
8.79
Second
Slope
(gr/m/ 1000 m)
0.0397
N/A
0.1142
0.0307
0.1739
0.0585
0.1494
Second
Corner
(1000 miles)
N/A
N/A
69.78
N/A
N/A
N/A
15.02
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
0.0898
N/A
N/A
N/A
0.1459
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARS
81-87FI
81-83 CARS
0.9017
1.1439
1.5384
5.2337
9.0704
0.0000
0.0000
0.0000
0.0000
0.0000
16.80
17.54
19.30
55.03
18.86
0.0357
0.0491
0.1986
0.0644
0.0635
58.68
N/A
N/A
N/A
N/A
0.0297
N/A
N/A
N/A
N/A
    Note: The first slope is zero, since it is assumed that the ZML emission rate is constant from zero miles to the first corner. For
    the cases with a single corner, the second slope is determined from the unconstrained regression and the corner occurs at the
    mileage where that line intersects the ZML mean emissions. For the case with two corners, the second slope was obtained
    using a regression line constrained to pass through the ZML mean emissions-mileage. The third slope is for the unconstrained
    regression line and applies at mileages above the second corner. (Unadjusted refers to estimates obtained using the FTP
    dataset only.)
M6.EXH.001
-23-
DRAFT 2/8/99

-------
                                                      Table 3 (cont.)
                         Running Emission Deterioration Model Coefficients for NOx (Unadjusted)

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82FI
8 1-82 CARS
ZML Mean
Emissions
(gr/m)
0.2582
0.2931
0.5976
0.5522
0.5614
0.6370
0.6121
First
Slope
(gr/m/ 1000 m)
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
First
Corner
(1000 miles)
18.58
21.55
34.25
26.12
12.52
16.36
8.79
Second
Slope
(gr/m/ 1000 m)
0.0048
0.0047
0.0042
0.0023
0.0059
0.0129
0.0063
Second
Corner
(1000 miles)
N/A
N/A
N/A
N/A
N/A
N/A
17.00
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
N/A
N/A
N/A
N/A
0.0060
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARS
81-87 FI
81-83 CARS
0.3782
0.3346
1.3234
0.5388
1.6660
0.0000
0.0000
0.0000
0.0000
0.0000
21.20
16.24
22.20
21.43
N/A
0.0044
0.0040
0.0040
0.0084
N/A
N/A
55.16
N/A
N/A
N/A
N/A
0.0032
N/A
N/A
N/A
    Note: The first slope is zero, since it is assumed that the ZML emission rate is constant from zero miles to the first corner. For
    the cases with a single corner, the second slope is determined from the unconstrained regression and the corner occurs at the
    mileage where that line intersects the ZML mean emissions. For the case with two corners, the second slope was obtained
    using a regression line constrained to pass through the ZML mean emissions-mileage. The third slope is for the unconstrained
    regression line and applies at mileages above the second corner. (Unadjusted refers to estimates  obtained using the FTP
    dataset only.)
M6.EXH.001
-24-
DRAFT 2/8/99

-------
                                                          Table 4
                                  High Emitter Adjusted Running Emission Deterioration
                                                 Model Coefficients for HC

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82FI
8 1-82 CARS
ZML Mean
Emissions
(gr/m)
0.0516
0.0843
0.1479
0.0815
0.1691
0.1240
0.2108
First
Slope
(gr/m/ 1000 m)
0.0013
0.0013
0.0000
0.0039
0.0003
0.0094
0.0048
First
Corner
(1000 miles)
20.03
34.39
18.89
19.83
25.24
11.29
10.18
Second
Slope
(gr/m/ 1000 m)
0.0036
0.0033
0.0078
0.0058
0.0098
0.0132
0.0158
Second
Corner
(1000 miles)
N/A
N/A
81.38
N/A
N/A
70.55
N/A
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
0.0059
N/A
N/A
0.0131
N/A
Adjustment
Additive
(gr/m/ 1000m)
0.0013
0.0013
-0.0001
0.0039
0.0003
0.0094
0.0048
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARB
81-87 FI
81-83 CARB
0.0932
0.0783
0.2495
0.2927
0.6587
0.0013
0.0013
0.0000
0.0000
0.0018
23.40
16.24
36.01
40.58
15.99
0.0038
0.0056
0.0083
0.0099
0.0127
N/A
55.16
N/A
N/A
N/A
N/A
0.0055
N/A
N/A
N/A
0.0013
0.0013
-0.0053
-0.0038
0.0018
    Note: Adjusted refers to estimates obtained using the high emitter correction factors. To obtain the adjusted values, the
    additive adjustments given in this table were applied to the unadjusted slopes in Table 3. Slope values of zero were assigned
    in cases where the additive adjustments would have resulted in negative deterioration.
M6.EXH.001
-25-
DRAFT 2/8/99

-------
                                                      Table 4 (cont.)
                                  High Emitter Adjusted Running Emission Deterioration
                                                 Model Coefficients for CO

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82FI
8 1-82 CARS
ZML Mean
Emissions
(gr/m)
0.7983
2.5684
2.1416
0.6910
1.0983
1.7270
2.9361
First
Slope
(gr/m/ 1000 m)
0.0310
0.0310
0.0000
0.0727
0.0000
0.1817
0.1414
First
Corner
(1000 miles)
13.78
N/A
19.04
21.13
25.68
16.62
8.79
Second
Slope
(gr/m/ 1000 m)
0.0707
N/A
0.1091
0.1034
0.1537
0.2401
0.2908
Second
Corner
(1000 miles)
N/A
N/A
69.78
N/A
N/A
N/A
15.02
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
0.0846
N/A
N/A
N/A
0.2873
Adjustment
Additive
(gr/m/ 1000m)
0.0310
0.0310
-0.0051
0.0727
-0.0203
0.1817
0.1414
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARB
81-87 FI
81-83 CARB
0.9017
1.1439
1.5384
5.2337
9.0704
0.0326
0.0326
0.0000
0.0545
0.1040
16.80
17.54
28.90
55.03
18.86
0.0683
0.0817
0.1327
0.1190
0.1675
58.68
N/A
N/A
N/A
N/A
0.0623
N/A
N/A
N/A
N/A
0.0326
0.0326
-0.0660
0.0545
0.1040
    Note: Adjusted refers to estimates obtained using the high emitter correction factors. To obtain the adjusted values, the
    additive adjustments given in this table were applied to the unadjusted slopes in Table 3. Slope values of zero were assigned
    in cases where the additive adjustments would have resulted in negative deterioration.
M6.EXH.001
-26-
DRAFT 2/8/99

-------
                                                      Table 4 (cont.)
                                  High Emitter Adjusted Running Emission Deterioration
                                                Model Coefficients for NOx

                                                     Light-Duty Cars
ModelYear/
Technology
88-93 PFI
88-93 TBI
83-87 FI
86-93 CARS
83-85 CARS
81-82 FI
8 1-82 CARS
ZML Mean
Emissions
(gr/m)
0.2582
0.2931
0.5976
0.5522
0.5614
0.6370
0.6121
First
Slope
(gr/m/ 1000 m)
0.0010
0.0010
0.0023
0.0021
0.0003
0.0000
0.0003
First
Corner
(1000 miles)
18.58
21.55
34.25
26.12
12.52
30.66
8.79
Second
Slope
(gr/m/ 1000 m)
0.0058
0.0058
0.0064
0.0045
0.0062
0.0069
0.0066
Second
Corner
(1000 miles)
N/A
N/A
N/A
N/A
N/A
N/A
17.00
Third
Slope
(gr/m/ 1000 m)
N/A
N/A
N/A
N/A
N/A
N/A
0.0063
Adjustment
Additive
(gr/m/ 1000m)
0.0010
0.0010
0.0023
0.0021
0.0003
-0.0060
0.0003
                                                    Light-Duty Trucks
88-93 PFI
88-93 TBI
84-93 CARS
81-87 FI
81-83 CARS
0.3782
0.3346
1.3234
0.5388
1.6660
0.0002
0.0002
0.0000
0.0000
0.0008
21.20
16.24
1754.24
32.21
N/A
0.0046
0.0042
0.0001
0.0056
N/A
N/A
55.16
N/A
N/A
N/A
N/A
0.0034
N/A
N/A
N/A
0.0002
0.0002
-0.0040
-0.0028
0.0008
    Note: Adjusted refers to estimates obtained using the high emitter correction factors. To obtain the adjusted values, the
    additive adjustments given in this table were applied to the unadjusted slopes in Table 3. Slope values of zero were assigned
    in cases where the additive adjustments would have resulted in negative deterioration.
M6.EXH.001
-27-
DRAFT 2/8/99

-------