&EPA
             United States
             Environmental Protection .
             Agencv
Office of Air Quality
Planning and Standards
Research Triangle Park N'C 2771
EPA-450/4-89-012
July 1989
              PROCEDURES FOR
                  APPLYING
           CITY-SPECIFIC EKMA

-------

-------
                          EPA-450/4-89-012
  PROCEDURES FOR
       APPLYING
CITY-SPECIFIC EKMA
  Office Of Air Quality Planning And Standards
      Office Of Air And Radiation
   U. S. Environmental Protection Agency
    Research Triangle Park, NC 27711

          July 1989

-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U. S. Environmental
Eroitection Agency, and has been approved for publication.  Any mention of trade names or commercial
products is not intended to constitute endorsement or recommendation for use.
                                      EPA-450/4-89-012
                                            11

-------
                                    PREFACE


     This document  is one  of  five  related  to  application  of EKMA and the use
of OZIPM-4  (Ozone Isopleth Plotting with Optional  Mechanisms),  the computer
program used by  EKMA.   Listed below are the titles of the five  documents and a
brief description of each.       -

"Procedures for  Applying City-specific EKMA",  EPA-450/4-89-012,  July 1989

     -  Describes the procedures for  using the Empirical  Kinetic Modeling
     Approach  (EKMA)..   The major focus is  on  how to develop needed inputs for
     OZIPM-4.  In addition this document describes how to determine a control
     target once OZIPM-4 has  been  run.

"A PC Based System for  Generating  EKMA Input  Files",  EPA-450/4-88-016,
November 1988

     -  Describes a program that creates EKMA  input files using  a menu driven
     program.  This sofware is only available  for  an  IBM-PC or  compatible
     machine.  Files built  using this system  can be uploaded to  a mainframe
     computer.

"User's Manual for OZIPM-4  (Ozone  Isopleth Plotting with  Optional  Mechanisms}-
Volume 1",  EPA-450/4-89-009a, July 1989

     - Describes the conceptual basis behind OZIPM-4.   It describes  the
     chemical mechanism, Carbon Bond 4, and each of the options  available in
     OZIPM-4.  Formats  for each of the options are  outlined  so that  a user
     can create  input files using any text editor.

"User's Manual for OZIPM-4 (Ozone Isopleth Plotting with  Optional  Mechanisms)-
Volume 2:  Computer Code",  EPA-450/4-89-009b,  July  1989

     - Describes modifications to the computer code that  are necessary in
     order to use OZIPM-4 on various machines.  A complete  listing of OZIPM-4
     is also found in this publication.

"Consideration of Transported Ozone and Precursors  and Their Use  in EKMA",
EPA-450/4-89-010, July  1989

     - Recommends procedures for considering  transported ozone and
     precursors  in the design of State Implementation Plans to meet national
     ambient air quality standards  for ozone.   A computerized (PC) system for
     determining whether an ozone exceedance  is due to overwhelming transport
     is described.   This document is  necessary, only if an area  is suspected
     of experiencing overwhelming transport of ozone or ozone precursors.

-------
EKMA may be used  in several ways:  (1) as a means  for  helping to focus more
resource-intensive photochemical grid modeling analyses on  strategies most
likely to be successful  in demonstrating attainment;  (2)  as a procedure to
assist in making  comparisons between VOC and NOx  controls;  (3) in non-SIP
applications, such as  in helping to make national policy  evaluations assessing
cost/benefits associated with various alternatives and (4)  for preparation of
control estimates consistent with  limitations/provisions  identified in Clean
Air Act Amendments.
                                      iv

-------
                                TABLE  OF  CONTENTS      .


                                                                    Page

List of Tables....	.........'.	  Vll-

List of Figures	 Vii1

Acknowledgements.!		          "~              iy
                                     °*»*»»»**e«*»*»*«,»»^,^e-e^w#   l j{

1.0  Introduction			       j

2.0  The CB-4 Mechanism.	    4

     2.1  Organic Reactivity			                         4
     2.2  Use of CB-4  in OZIPM4.....	!!.'.*!.*•.'!!.*!;!!.*.'    9

3.0  Procedures for Applying EKMA/CB-4	   11

     3.1  Selection of Model ing Cases	        14
     3.2  Development of Model  Inputs	...!.*!'."!.".*.*!.*.".''  15

          3.2.1     Light Intensity	   16
          3.2.2     Dilution	     18
          3.2.3     Post-0800 Emissions	'.'.['•	   18
          3.2.4     Initial N02/N0	....	;	.'.'!.'!..*."."   21
          3.2.5     Ozone Transport	   22

               A.    Present Transport of Ozone at the  Surface	   22
               B.    Present Transport of Ozone Aloft	   25
               C.    Future Transport of Ozone		   27

          3.2.6     Precursor Transport	   29
          3.2.7 .    Organic Reactivity....,	......'..".'!!.'.'.'.!.'!   31

               A.    Surface NMOC	                  31
               B.    NMOC Aloft	......*.'!.'!.'!.'!!!!!  35

         3.2.8      Temperature	,.	"..'...	  36
         3.2.9      Water Vapor			...	.............  37*
         3.2.10     Biogenic Emission  Estimates	  37

    3.3  Predicting Peak Ozone			  38

         3.3.1      Procedures for Making Ozone Predictions	  38
         3.3.2      Comparisons  of Predictions With  Observations..  41
         3.3.3      Review and Adjustment to  Model  Inputs	  43

-------
                        tTABLE OF CONTENTS  (CONTINUED)
                                                                    Page
     3.4   Computing VOC Emission Reductions........		....   46
           3.4.1      Derivation of Empirical  Data	'..   46
           3.4.2      Daily Ozone Design Value.............	•.	   47
         •3.4.3      NMOC/NOX Ratios...	..,	   47
     3.5   Selection of  the VOC Emission Reduction  Target	:.   49
           3.5.1      Without Overwhelming Transport.	   49,
           3.5.2      Selection of a VOC Reduction Target at  Sites
                     Subject to Overwhelming  Transport.. T	   53
References	•	   55
Appendix A - Listing of CB-4  Mechanism	'.; A-l
Appendix B - Estimation of  Mixing  Heights for Use  in OZIPM4	 B-l
Appendix C - Computation of Carbon Bond  Fractions  From GC Data	 C-l

-------
                            LIST OF TABLES

 Number                                                        paqe
 2-1   Carbon  Numbers for CB-4 Organic Species...	   8
 3-1   OZIPM4/CB-4 Model  Inputs.	  17
 3-2   Example Calculations  of the Design NMQC/NOX Ratio..	  50
 3-3   Example Illustrating  Effect of Model  Predictions on
      Selection  of Control  Target		„		  52
 A-l   CB-4 Mechanism	„	A-2
 B-l   NWS Radiosonde Stations		,		 B-4
 B-2   Preferential  Order of Data  Selection.;......,.	 B-6
 B-3   Procedures  for Estimating Mixing  Heights	 B-9
 B-4   Worksheet  for Computing Mixing  Heights	.	B-ll
 B-5   Surface  and  Sounding  Data	B-13
 B-6 •'Morning  Mixing  Height  Determintion		..B-17
 B-6A  Example  (Hypothetical  Data)	B-19
 B-7  Maximum  Mixing  Height  Determination	;	B-20
 B-7A Example  (Hypothetical  Data)	•	B-24
 C-l  Species  Profiles by Bond Groups for CB-4...	  C-2
C-2  Example  Problem - Part  1	C-16
C-3  Example  Problem - Part  2	C-17
                                vii

-------
                           LIST OF  FIGURES
 Number
 2-1   Example  of Carbon Bond Lumping Procedure	.	    7
 3-1   OZIPM4 Example  Isopleth Run	...:...	   12
 3-2   Example  Determination  of Hourly Emissions	   20
 3-3   Examples of Acceptable Monitoring  Locations  for
      Estimating Transported Ozone	'f   28
3-4   Future Ozone Transport as a Function of  Present
      Transport	,	
B-l  Flow Chart for Tab! e B-3	 g_
12
                               viii

-------
                                ACKNOWLEDGEMENTS
      Contributions from authors of earlier versions of this document, served
 as  a  basis  for  this report.   These include Gerald Gipson,  and Marcia Dodge,
 Atmospheric Research  and  Exposure  Assessment Laboratory,  EPA; Bob Kelly,
 Region  II EPA;  and Henry  Hogo,  Mike  Gery,  and Gary Whitten,  Systems
 Applications.   In  addition,, Ke.lth  Baugues  and Edwin Meyer,  Jr.  should be
 acknowledged for the time spent  in reviewing  and-revising earlier drafts  of
this report.  Special recognition  is due Mrs.  Cynthia Baines  for  her  splendid
clerical support in preparing and assembling  this  report.
                                     ix

-------
  1.0   INTRODUCTION
       In March  of 1981,  the  U.S.  Environmental  Protection  Agency (EPA)  issued
  guidelines for applying the city-specific  Empirical  Kinetics  Modeling  Approach
  (EKMA) (Gipson, et al,  1981).  EKMA  is a procedure that can be  used  to
  estimate emission reductions that are needed to achieve the national ambient
  air quality standard (NAAQS) for ozone. Application  of city-specific EKMA
  according to the March 1981 guidelines entails using the  Ozone  Isopleth
  Plotting Package (OZIPP) to relate peak ozone concentrations  to  its
 precursors--nonmethane organic compounds (NMOC) and oxides of nitrogen (NOY)
                                                                           **
  (Whitten and Hogo,  1978; and EPA, 1978).  OZIPP is a computer program that
  incorporates a simplified trajectory model and a chemical  kinetics mechanism
 (known as the DODGE mechanism) that mathematically simulate ozone formation.
 After the issuance  of the'March 1981 guidelines,  the use of other chemical
.mechanisms with EKMA was suggested (Jeffries,  et al,  1981; and Carter,  et al,
 1982).  In response-,  supplemental guidance on  using other mechanisms was
 circulated to EPA Regional  Offices in December of 1981  (Rhoads,  1981).
 Specific  guidance regarding the use of one alternative  mechanism-the Carbon
 Bond  III  mechanism  (CB-S)--was issued in February of 1984  (Gipson, 1984).
      Since 1984,  newer  chemical mechanisms have been  developed (Gery, et al,
 1988;  and  Lurmann,  et al,  1987).   This document focuses on information
 necessary  to  apply  EKMA utilizing the Carbon Bond  IV  mechanism (CB-4) and  •
 provides details  on  all  necessary input  parameters.   The discussions that
 follow will focus exclusively  on  using the  CB-4 mechanism  with the OZIPM4
 program.   This  program,  Ozone  Isopleth Plotting With  Optional  Mechanisms  4
 (OZIPM4),  is  an updated  version of OZIPP which  contains the most recent

-------
 chemical kinetics  and  stoichiometric  information.   Any application of EKMA

 should be carried  out  with the  OZIPM4  code  described  in EPA (1989).

      The remainder of  this document  is divided  into two chapters.   Chapter 2

 contains a discussion  of the CB-4 mechanism and  its relationship to  the .OZIPM4

 program.  Chapter  3 describes the information necessary to  develop input  data

 suitable for use of EKMA/CB-4.


      In the OZIPM4 model, a column of  air containing  ozone  and precursors  is

 transported along  an assumed straightline trajectory.   The  trajectory is

 defined so that the simulated column of air  over the  city being simulated

 arrives at the site observing the daily maximum ozone  concentration  at the
 !                     '
 time of the observed maximum.  As the  column moves, it  encounters  gridded

 emissions of fresh precursors that are mixed uniformly  within the  column.  The

 column is assumed to extend from the earth's surface through the mixed layer.

 The assumed horizontal  dimensions of this column are such that the

 concentration gradients are small enough to make the horizontal  exchange of

 air between the column  and its surroundings insignificant.   The air within the

'column is assumed to be uniformly mixed at all  times.

      At the beginning of a simulation,  the column is assumed to contain some

 specified initial  concentrations of  NMOC,  NOY,  and CO.  As  the column moves
                                             A                    i

 along the assumed trajectory,  the height of the  column will  change  because of

 variations  in mixing  height;  it  is assumed to change with time during a user-

 selected  period (for  example,  8  a.m.  -  3 p.m.),  and to be constant  before  and

 after that  period.  As  the  height of  the column  increases,  its volume

 increases,  and  air  above from  the inversion  layer  is mixed  in.  Pollutants

 above the mixed layer are described as  "transported above the surface layer"

-------
or  "transported  aloft."  Any  ozone  or  ozone  precursors  above  the mixed layer
that are mixed into the  column  as  it expands are  assumed  to be  rapidly mixed
throughout the column.
     Concentrations of NMOC species, NO, N02,  CO,  and 03, within the  column
are physically decreased by dilution due to  the inversion rise,  and physically
increased both by entrainment of pollutants  transported aloft and by  fresh
emissions.  All species react chemically according to the kinetic mechanism
selected.  Photolysis rates within that mechanism  are functions  of the
intensity and spectral distribution of sunlight, and they vary diiirnally
according to time of year and location.

-------
 2.0   THE CB-4 MECHANISM
       As the  name implies,  CB-4 is the fourth in a series of evolving chemical
 kinetics mechanisms.   Each  of the successive carbon bond mechanisms contains
 revisions that reflect increased knowledge of the photochemistry leading to
 ozone formation.   The  CB-4  mechanism is currently the most recent version of
 that  generic  series.   It  has  been designed to simulate laboratory smog chamber
 experiments using detailed  data bases,  as  well  as atmospheric situations in
 which much less  information, is  typically available.   While a comprehensive
 discussion of the scientific  basis of  the,CB-4  mechanism is beyond the scope
   •  i     i           i   .     ,   |           ,
 of this  document,  some introductory material  on basic concepts  is included
 below for those  unfamiliar  with CB-4.
       A  distinguishing feature  of any  chemical  mechanism is the  manner in
 which organic'reactivity  is treated.   Because the construction and use of a
 mechanism that  includes all atmospheric species  is  virtually impossible,
 individual organic species must be combined,  or lumped,  into some sort.of
 functional group  or groups.  Thus,, the  discussion of  any chemical  mechanism
 must  necessarily  address the manner  in  which  organic  chemistry is represented
 in the mechanism.  The concepts  underlying the  treatment  of organic reactivity
 in CB-4  are discussed  in Section  2.1 below.
      As  noted in Section 1.0,  use of the CB-4 mechanism  in  a city-specific
 EKMA  analysis  is most  easily accomplished with the OZIPM4 computer program.
2.1  Organic Reactivity
      As described in Section 2.0, a characteristic that typically
distinguishes chemical  mechanisms  is the manner  in which organic compounds are
represented in the mechanism.   A number of approaches have been taken,

-------
 but most have focused on lumping similar  species  into  a  single,  identifiable
 molecular species that represents the chemistry of that  particular class  of
 compounds.  For example, propylene might  be used  to represent the chemistry of
 all alkenes.  The CB-4 mechanism is somewhat different, in that the primary
 functional organic groups are based, on various, types of  structural units
 (e.g., single-bonded carbon atoms) as opposed to molecular type  (e.g.,
 alkanes).  As will be seen below, this kind of structuring results in some
 organic species being partitioned among more than one functional group.
       In CB-4,  nine functional groups are used to represent organic species,
 each based ion various, types of carbon bonds:
       (1) single-bonded paraffinic carbon atoms,  and represented by PAR;
       (2) slowly reacting olefinic double bonds,  almost exclusively
 ethylene and represented  by ETH;
       (3) relatively reactive  olefinic double  bonds,  and  represented  by OLE;
       (4) less  reactive aromatic  compounds represented  by TOL;
       (5) more  reactive aromatic  compounds represented  by XYL;
       (6) formaldehyde represented by FORM;
       (7) acetaldehyde and  high aldehydes  represented by  ALD2;
       (8)  isoprene,  represented by ISOP;
       (9) nonreactive compounds represented  by  NR.
      Just as important as the definition  of the functional groups themselves
 is the manner in which individual organic  species  are apportioned to those
groups.  As noted above, a particular organic compound  is assigned to  a  CB-4
group, or groups, on the basis of molecular  structure.  To illustrate  the
procedure, consider the propene molecule which contains one single carbon-bond

-------
   and one double carbon-carbon bond (see Figure 2-1).  In the CB-4 mechanism,
   the propene molecule is represented by one paraffin and by one olefin.  In
   essence,  the molecule  has  been apportioned on the basis of the carbon-carbon'
   bonds:   the double  bond represented by OLE,  and the one single- bond by PAR.
   Similar classifications haye been determined for hundreds  of other compounds,
   and  they  provide the basis  for establishing  the overall  reactivity of an  urban
  mix.
        In the propene example  illustrated  in  Figure 2-1,  note  that  the  number
  of carbon atoms associated with PAR  is one,  while the number  for OLE is two.
,  l\ general principle underlying use of the carbon bond mechanism is that the
  number of carbon atoms associated with any individual carbon  bond group is
  fixed.  (Table 2-1 shows these characteristic carbon numbers  for the carbon
  bond functional groups.)  By making use of the carbon numbers, concentrations
•  of each  CB-4 group can be determined from concentrations of.individual  organic
  species.   To illustrate,  consider the propene example discussed above,  and
  further  assume  that the concentration.of propene is  3 ppmC.   Since propene is
  represented  in  CB-4 by one PAR and by one OLE,  the 3  ppmC total propene
  concentration must be  apportioned  to  these two  carbon bond groups.   Of  the
  three  carbon atoms in  a propene molecule,  one is PAR  and two are OLE (see
  Figure 2-1). Thus,  one-third  of the carbon atoms can  be  thought of as PAR,  and
  two-thirds as OLE.   Since  concentration  is proportional  to the number of
  carbon atoms, the  concentrations of PAR and OLE  in the CB-4  mechanism would  be
  1 ppmC and 2 ppmC,  respectively.*  This same  concept  can  be.extended to
 mul.ticomponent mixtures as well.   In  such  cases, concentrations of  the
       i.e., CpAR =1/3x3 ppmC and CQLE =2/3x3 ppmC
                                        6

-------
H

I
C =

H   H
II
C - C

    A
                              ,
                             -  H
                        Propene
          H   H


          C = C

          H
                H

              -  C - H

                A
          1  OLE
               1  PAR
Figure 2-1.  Example of Carbon Bond Lumping Procedure
                          7,

-------
                       TABLE 2-1

        CARBON NUMBERS FOR CB-4 ORGANIC SPECIES

Carbon Bond Group             Carbon Number (carbon atoms per
                                       molecule)
      PAR                           1
      ETH                           2
      OLE                           2
      TOL                 •          7
      XYL                           8            '
      FORM                           1
   -   ALD2                           2
      NR                             !
      ISOP
                         8

-------
  individual  organic species  are first apportioned to their respective CB-4
  group.  The total  concentration of any particular CB-4 group is then obtained
  by summing  the contributions  due to the individual  organic species.   This
  procedure will be  more fully  discussed in  Appendix  C.
      In using the  CB-4 mechanism with  the  OZIPM4 program,  absolute -.
  concentrations of  the individual  CB-4  groups  are not directly  input  to  the
 model.   Rather, the total NMOC  concentration  is  specified,  and  the fraction of
 carbon attributable to each CB-4  group  is  input.  For  example,  assume that the
 total NMOC concentration is 2.0 ppmC, of which 1.4 ppmC  is  PAR  (as determined
 by the  procedure described in the preceding paragraph).  Then the apportioning
 factor,  or carbon fraction,  for PAR would be 0.70,' indicating that 70 percent
 of the  total carbon is categorized as PAR.   A special set of default values
 for CB-4 fractions  is normally used.  These defaults were derived from 1984-86
 NMOC  species data collected  in many cities  (Jeffries, 1987).
 2.2   Use of  CB-4  in OZIPM4
      The CB-4 mechanism  that is contained in OZIPM4 is  outlined in
 Appendix A.   A discussion of the development and  testing of this mechanism is
 contained  in Gery,  et  al, 1988.   More extensive  information on  the evolution
 of the carbon bond  mechanism in  general  can be found in Killus  and Whitten,
 1983; Killus  and Whitten, 1982;  Whitten and Hogo, 1977;  Whitten, et al,  1980;
 Whitten, et  al, 1979.
     The specific inputs  necessary to use CB-4 in OZIPM4 are contained  in
 EPA, 1988 and EPA 1989.  The discussions  in  Section  2.1  and  above  have
provided a general  overview of the CB-4 mechanism and its relationship to  the
OZIPM4 program.  In most  instances, consideration of the details of the

-------
mechanism will  not  be  required  in  any particular model  application.  The major
concern  in most  applications  is  the  determination of the  total  NMOC
concentration.   In  cases where defaults  are  not  used,"specification  of the
carbon bond fractions  required to  apportion  the  total  carbon  concentration  to
the individual carbon  bond groups  (i.e., PAR,  ETH, OLE, TOL,  XYL,  FORM,  ALD2,
ISOP,  and NR) is also  of concern.  The procedures  for developing these  and
other model  inputs for an EKMA application are the subject of the  next
chapter.
                                     10

-------
 3.0  PROCEDURES FOR APPLYING EKMA/CB-4
      Although the March 1981 guidelines deal explicitly with OZIPP and the
 DODGE mechanism,  many of the concepts described in that document are relevant
 to the use of EKMA with OZIPM4 and the CB-4 mechanism.  For example,  selecting
 the cases  to  model  is unaffected by choice of chemical  mechanism.   Neverthe-
 less,  use  of  CB-4 with OZIPM4 does require some special  considerations.   This
 chapter will  focus  primarily upon these circumstances,  but will  also describe
 all  other  facets  of conducting an EKMA modeling analysis.
      The ensuing  discussion  of using CB-4 with EKMA can  perhaps  be facili-
 tated  by a brief  overview of the general  modeling  procedure.   While  the
 following  section describes  EKMA in terms of ozone isopleth diagrams,  it  is  no
 longer necessary  to develop  these diagrams in.order to determine the VOC
 control requirements.   By using  the EKMA  option the control  requirement  is   .
 determined without  drawing isopleth diagrams.   The OZIPM4  program  is used to
 generate ozone  isopleth  diagrams that  explicitly relate  peak  hourly  ozone
 concentrations  to. initial  (i.e.,  8 a.m.)  ambient levels  of the ozone pre-
 cursors NMOC  and  NOX  (see  Figure 3-1).  The  diagrams are used with a measured
 peak ozone  concentration  and  a city's  NMOC/NOX  ratio to  compute, on  a  percent
 basis, the  VOC  emission reduction  needed  to  lower  the observed peak  to the
 level  of the  standard.  While  isopleth diagrams  are explicit functions of
 initial NMOC  and  NOX, the positioning  of  the ozone  isopleths on  the  diagram  is
 affected by model input variables  that are related to meteorology, emissions
 occurring throughout the day, and  pollutants transported from areas  upwind of
the city under review.  Because  these factors vary from day to day,  the
                                      11

-------
        21*0    OlvQ    QO'O    90-0
CO
               01-0  •
90*0    9Q-0
    (Wdd) XON
00-0°
                               -12-

-------
 highest VOC emission reduction estimate will  not necessarily correspond to the
 highest,  observed  ozone peak (Killus and Whitten,  1983;  and Killus and
 Whitten,  1982).  To account for this phenomenon,  the modeling approach
 recommended in the March 1981  guidelines consisted of:
      (1)   modeling a number of high, observed ozone peak concentrations;
     '(2)   computing the amount of VOC emission reduction needed tb lower each
           peak to  the level  of the standard;
      (3)   selecting a final  VOC emission reduction target that  is  consistent
           with the statistical  form of the  ozone standard.
        i            •                      .    •          '•'.,..         i
      Subsequent  to the distribution of these  recommendations, EPA  issued     '
 supplemental guidance further  recommending  that  predictions  of  peak 'ozone  bei
        1                                                            i
 compared to observed  levels  (Rhoads,  1981).   If  the  agreement between
 predictions and  observations is  found  to  be poor,  review  and possible
 adjustment to key  model  inputs  are  suggested  prior to computing  VOC emission
 reductions.  While  good  agreement between predictions and observations does
 not completely ensure  accurate  control  estimates,  successful prediction  of
 observed ozone peaks  does provide some  confidence  that the chemical and
 physical processes  leading to ozone  formation are  being adequately simulated.
     The modeling  procedure described  in the  preceding paragraphs can be
divided into five basic steps which  should be followed:
     (1) selecting the observed ozone peaks to model;
     (2) formulating the model   inputs;
  «
     (3) predicting peak ozone;
     (4) computing VOC emission reductions;  and
     (5) selecting the overall  VOC emission  reduction target.
                                      13

-------
 3.1   Selection  of  Modeling  Cases
       As  noted  in Section  3.0,  the  highest  VOC  control  estimate may not
 correspond to the  highest observed ozone concentration.   Further,  the
 statistical form of the ozone  NAAQS permits on  average,  one daily  maximum  1-
 hour .average ozone concentration above 0.12 ppm per calendar year  at each
 site.  Consideration of these  two.  factors Jed to the recommendation that a
 number of observed peaks  above 0.12 ppm be modeled for each site.  The  VOC
 emission reduction target is then  selected from these results  in a manner that
 is consistent with the statistical form of the ozone NAAQS'i  For an EKMA/CB-4
 analysis, the same procedure is recommended.               . .   - ,   ,       ;   ,
      Recommended Procedure:   The five (5) highest daily, maximum ozone   '
 concentrations at each site  should be selected as candidates for modeling.
 Only ozone peaks that occur  within or downwind of the urban area under review
 should be included.  The five highest such values should generally be chosen
 from the most recent 3 years during which measurements were made at a site.
      A State  may choose to include an additional year if data from another
 ozone season  become available between the time of the SIP call  and the  time
 when the analysis is conducted.  While an additional  year may be added, a year
 may not be replaced (i.e., 4 years  of data must be used). If 4  years of data
 are included,  the six  (6)  highest  daily,  maximum ozone concentrations  at each
 site should be  selected as candidates  for modeling.   If there  is a tie  for  the
 last daily maximum  value,  both  days should  be  modeled.   In the  event that a
..significant amount  of  time (a few years)  passes  between  the time of the SIP
 call  and  the  start  of  the  modeling  analysis, the appropriate  EPA Regional
 office should be contacted to determine the appropriate  years to model.
      In some  cases,  it  may happen that on days  initially selected  as
 candidates for modeling, daily  maximum ozone (03)  is most likely the result  of
 "overwhelming transport" from upwind areas.  That  is,  it  is unlikely that  •
 locally generated emissions  have an appreciable  effect on  the observed  daily
maximum.  Procedures for determining whether an  observed  daily  maximum  results
from overwhelming transport  are described in detail by Meyer and Baugues
 (1989).   In general, overwhelming transport is a strong possibility if the
                                      14

-------
 daily maximum occurs before 10 a.m. or if the timing of the observed maximum
 is inconsistent with available ,10 a.m. - 4 p.m. surface wind data, and the,
 orientation of the monitoring site with respect to the Metropolitan
 Statistical  Area (MSA)  under review.   Even if it is likely that an, observed
 daily maximum 03 concentration results from overwhelming transport, if is
 possible  that the selected day should still  be modeled.   This would be
 appropriate if the following occurred:,  (a)  concentrations greater than  0.12
 ppm  occurred at other times of the day; (b)  surface wind data and monitor
                                                i
 orientation  were consistent with  impacts from the local  MSA at these times;
                                                 ,i    ,  ,           ,
 and  (c) the  63 concentration judged'to result (from'local  emissions was on$  of
 the  top five local  peaks.   Unless all  of the preceding  three conditions  are
 met,  the  day should be  discarded  and  replaced by the previously unselected  day
 having the  highest observed daily maximum.   Overwhelming  transport can also
 affect.selection of which  estimate for VOC  and/or NOY controls is needed to
                                                     J\     ^       •
 attain the NAAQS.   This latter issue  is addressed in Section 3.5.
 3.2   Development of-Model  Inputs    .                .         . •
      As just described, the five  or six highest,  daily maximum ozone peaks  at
 each  site are selected for  modeling.   Two basic  objectives  of the. modeling
 analysis  are  to  predict the observed  ozone peaks, and to  compute  the VOC
 emission  reductions  needed  to  lower each observed peak to the level  of the
 ozone NAAQS.   To  best accomplish  these  objectives,  the model  inputs  should be
 based on the  atmospheric conditions associated with  each observed  peak.   Thus,
 their derivation  ought to be done  on  a  case-by-case  basis.   In  some  instances,
 however,   insufficient or inadequate data preclude such a determination,  and
appropriate approximations  or defaults  are needed.   The major  purpose  of  this
                                      15

-------
                                   I
 section  is to describe the methodologies recommended for deriving the model
 input  values  under both sets of circumstances.
     Table 3-1 summarizes the model  input variables that require
 consideration,  regardless of the intended purpose of the model  simulation.
 Before discussing  each of the model  input variables, one additional  point
 should be  added.   The  recommendations  discussed below deal  with model  inputs
 that correspond to conditions associated with the observed  ozone peak.(i.e.,
 so called  base-case conditions).   Some of these conditions  might be  'expected
                                                                          i
 to change  in  future years subsequent to the  implementation  of VOC control
 programs'.   Factoring these potential changes  into the. modeling  analyses  wi'll
 be discussed  in Section 3.4.   Thus, the recommendations  discussed below
 concerning  the  derivation of  model input values will, necessarily focus on  data
                                                     i
 corresponding to emissions  and  atmospheric conditions  associated with  a
 particular  ozone peak  observed  in  the  base case.
     3.2.1  Light  Intensity
          The OZIPM4 program  uses  a city's latitude^  longitude,  and  time
 zone,  and the "day  of the  year being modeled to  generate  the appropriate1
diurnal pattern of  photolytic reaction  rates.   While updates  have  been; made  to
some of the photolytic rates, these have been incorporated  in the  OZIPM4
computer code.  Thus, no  changes need be made for this set of model  inpu;ts.
     The correct set of numerical time  zones for the continental Unitedl      '
States  is as follows:                           •                       i   •
               Numerical  Time Zone          Common Name
                    4.0                   Eastern Daylight Time  .
                    5.0                   Central Daylight Time
                    6.0                   Mountain Daylight Time
                    7.0                   Pacific Daylight Time
16

-------
Model Input Variables
Sunlight Intensity
Dilution
Post-0800 Emissions
Ozone Transport
Precursor Transport
Reactivity
Temperature
Water Vapor.
Biogenic Emissions
                              Table 3-1
                       OZIPM4/CB-4  MODEL  INPUTS
Section
Addressed
3.2.1     ,
3.2.2
           i
3.2.3
3.2.'5
3.2:6  .
3.2.4, 3.2.7
3.2.8
3.2.9
3.2.10      i
                i  i
                                 17

-------
      To produce standard time simulations, even though the output will  show
 daylight time units, a false time zone of one unit (hour) more can be used.
 Thus, Pacific Standard Time photolysis constants would be generated  if  a 8.0
 were entered instead of the correct 7.0 time zone.
      Recommended Procedure:  To properly simulate light intensity in OZIPM4,
 input the city's latitude-,  longitude, time zone, and the day of the year beinq
 modeled.   .  ^' .  -                           ,    •           .
      3.2.2  Dilution
                              \
           In the OZIPM4 model,  dilution occurs as a result of the rise  in
 atmospheric mixing height that  typically1 occurs between early morning and mid-
 afternoon.   The mixing height can be viewed as the top of a surface based
                               ;   •         I     i                  •           i  .
 layer of  air which is  well-mixed due to^mechanical  and thermal  turbulence.
 Specific  inputs to OZJPM4 include the early morning  mixing height,  the maximum
                                          i
 afternoon  mixing height,  the  time that the  mixing  height  rise begins, and the
 time  at-which the  maximum mixing height is  finally attained.   Procedures for
                                                !
 estimating  the  early morning  mixing  height  and maximum afternoon  mixing  height
 from  available  radiosonde measurements!are  outlined  in Appendix B of  this
 document.   The  OZIPM4  program will  internally  calculate,the rate  of rise in
 mixing height based  upon  a  characteristic curve developed  by  Schere and
 Demerjian  (EPA  1981; Schere,  et  al,  1977).
      Recommended Procedure:   It  is recommended that city-specific estimates
 of 0800 LCT mixing height and maximum  afternoon mixing  height be  computed
 using procedures outlined in Appendix  B.|  Minimum 0800  LCT mixing heiqtit used
 should be 250 meters.                   I
     3.2.3  Post-0800 Emissions
          Post-0800  emissions refer  to  emissions occurring along the
trajectory subsequent to the start of the model simulation.  The actual  model
 inputs are expressed as emission densities (kg/km2) of NMOC, NOX,  and carbon
                                      18

-------
 monoxide (CO) concentrations that should be added each hour to represent the


 effect of fresh precursor emissions.  The requirements have changed from the


 March 1981 guidelines in two respects:  (a) large point sources of NOY are
                                                                      A

 differentiated from other sources and (b) CO emissions are considered.


      The following example illustrates how to determine which county's
 T.   ^               '                 *

 emissions should be used for each hour.   Figure 3-2 shows an example


 trajectory.   In this case the peak ozone occurs in Rockdale County between 3
                          1       i
 and  4  p.m.   The parcel  starts at 8 a.m.  in Fulton County.   Between 8 a.m.  and

                i           >  ''            ,        -  ''
 4  p.m.  are 8 hours,  indicated pq the straight-line trajectory in  Figure 3-2.


 Each value on'this'line  represents the location of the parcel  at  1 .through 8

 hours.


     During  hour 1  (8-9  a.m.), ,the parcel  is entirely in  Fulton County.


 Emissions for 8-9  a.m. from  Fulton County should be used  in OZIPM4.   During


 the  second hour the  parcel moves into  Dekalb County,   Emissions for 9-10  (hour


 2) should' be averaged between  Fulton and  Dekalb Counties.   During  hours 3
          i                 I              •                          .

 through  6, the  parcel is  entirely in Dekalb  County and emissions should be
                             1  I                               :
 based upon Dekalb County,  buying hour. 7,  the  parcel  crosses  into  Rockdale


 County.  Emissions for hour  7  (2-3 p.m.)  should be  averaged between  Dekalb  and
                                   !     1     .  • .

 Rockdale counties.   Emissions for hour-8  should be  from Rockdale County.


     In developing post-8 a.m. emission densities  for  EKMA,  it  is  necessary
                             I     !

to determine whether significant  NOX point sources  have effective  stack


 heights greater than the  initial  mixing height  (usually 250 meters).   If the


effective stack height is greater than the initial mixing height,  the  NOY
                                                                        A

emissions from that source will not  be contained within the mixed  surface
                                      19

-------
LU
                                        CO

                                        O
                                        CO
                                        CO


                                        LJJ


                                        5:

                                        O

                                        u_
                                        O


                                        O

                                        I


                                        s
                                        tr
                                       o
                                       UJ

                                       a.
                                       CJ
                                       UJ
                                       cc
     20

-------
•layer during the first few hours of the OZIPM4 simulation.  At sometime later
 in the day, as the mixing height rises, these NOX emissions will be entrained
 into the mixed surface layer.  Sources which should be reviewed to determine
 effective stack height include large industrial process boilers and power
 plant boilers.  Only significant saurces of NOX (those greater than 5 percent
 of the total NOX emission inventory) should be reviewed to estimate effective
 stack height.
      Each day is modeled to determine hourly mixing heights.  The days must
 then be rerun  adding in the tall  stack NOX emissions..   These emissions should
 be added to the hour (and only to that hour) where the mixing height first
 equals or exceeds the effective stack height.
      The recommended procedure for estimating  effective stack height is to
 run PTPLU (EPA,  1982).   This  model  requires stack height,  stack gas
 temperature, stack gas  velocity,  and inside stack diameter.   The
 meteorological  conditions of  interest are C stability  and  5  m/s wind speed (at
               •
 10 meter height).
      An alternative procedure is  contained on  the following  pages for those
 individuals who  do not  have access  to PTPLU.
      Recommended Procedure:   Post-0800  emissions  should be entered as
 emission densities (kg/km*1).   Large NOX sources may  require  a review of
 effective stack  height  to determine if  the NOX  emissions are within the mixed
 surface layer.   Emission  densities  are  required for  NMOC,  NOV,  and CO.
                                                             A
      3.2.4  Initial  NDo/NO..       .                                  '
           The  March 1981  guidelines recommend  a default value of 0.25.
 Review of recent data indicates that  this  ratio may  vary over a wide range
 (.1-.9)  and  that "median"  ratios  for  individual cities  may also vary
                                      21

-------
 significantly.  Modeling analyses have  indicated that EKMA-CB4  is not
 sensitive to this ratio.  Thus the default value of 0.25 should be used.
      Recommended Procedure:  A city-specific value for the nitrogen dioxide
 to oxides of nitrogen ratio [(N02)/NOYJ need riot be estimated. The default
 value of 0.25 should be used.
      3.2.5  Ozone Transport
           The two possible mechanisms by which .ozone is transported.into an
 urban area are:
      1.   Advection of ozone along the earth's surface,  and
      2.   Advection of ozone aloft, typically at night and during early
 morning  hours above the ground-based mixed layer,  with  downward mixing when
 the mixing layer increases  later in  the day.             .
      Ozone transported at  the surface is subject to surface reactions  and
 scavenging by other species  [e.g., nitric oxide (NO)] emitted during  the
 night.   As a  result of nighttime atmospheric  stability,  ozone transported
 aloft does not come into contact with scavengers emitted  during  the  night.
 Thus, overnight  advection of  ozone aloft is the more  significant mechanism  of
 transport  from one  urban area  to  another (EPA,  1977;  and  Chan, et  al,  1979).
      Control  strategies designed  to  attain and/or maintain  the ozone standard
 in  individual urban  areas must take  into consideration the  impact  of
 transported ozone on peak afternoon concentrations.*
      A.    Present Transport of Ozone  at  the Surface
           The, chief  impact of ozone transport near the surface is  expected  to
be the more rapid conversion of NO to N02.  Several field studies  have shown
      Recall the discussion in Section 3.1, there are days when transport is
the predominant cause of observed 03.  The recommendations in Section 3.2.6
apply for instances in which this is not the case.
                                      22

-------
                          ALTERNATIVE PROCEDURE FOR ESTIMATING
                               EFFECTIVE STACK HEIGHT
 1.    Estimate wind speed at stack height
                    —
                      h  -
u - 5.0
                               m/s
                      10  '

     where:   h  is  the physical  stack  height  in  meters

2.   Estimate F  (the  flux  parameter)

          F - g  V? d2 DT/4TS
  where:  g = 9.8 m/s2
         Vs = stack gas velocity (m/s)
          d - inner stack exit diameter (m)
        € DT = Ts - T  ( stack ogas temperature- ambient air
              temperature)  K
         Ts = ambient air temperature °K  (assume 297 °K)
3.   Estimate crossover temperature  DTC

     For F < 55          DTC = 0.0297 TS V^/Vd1/3
     For F 2: 55          DTC = 0.00575 TS Vs2/3/d1/3
    where:   TS = ambient  temperature (°K)
            Vs « stack gas velocity (m/s)
             d = stack exit diameter (m)
     If  DT <  DTC   Go to step 4
     If  DT £  DTC   Go to step .5
                                 23

-------
 4.    Estimate momentum plume rise
      Dh  = 3  d Vs/u
 where:   d =  stack exit diameter (m)
       Vs =  stack gas  velocity  (m/s)
         u =  wind  speed at  stack height  (m)[from Step 1]
       Dh =  plume rise (m)
      Go  to Step 6

 5.    Estimate buoyant  plume  rise
      For  F < 55         Dh = 21.425 F'75/u
      For  F > 55         Dh = 38.71 F'6/u
where: Dh  =  plume  rise  (m)
        F  =  flux  parameter      [from Step 2]
        u  =  wind  speed at stack height (m)[from Step  1]
6.    Estimate effective stack height
      •  H =  hs +  Dh
where:  H =  effective stack height (m)
       hs = physical stack height (m)
       Dh = plume  rise (m)                     [from Step 4 or 5]
                                 24

-------
 that ozone transported  along  the  surface  tends  to  be  minimal  (Chan,  et  al,
 1979; Decker, et al,  1977;  and  Ludwig,  1979).
           Recommended Procedure:   Based on  the  previous  discussion,  it  is
. recommended that, for most  situations,  the  value for  present  ozone transported
 at the surface be set equal to  zero for each day modeled.
           Alternate Procedure:  If ozone  levels are measured  downtown during
 6-9 a.m., surface ozone.transport  may be-  considered.   It  is recommended that
 the 6-9 a.m. LCT average ozone  concentration at an- urban  site(s) be  used as
 the estimate of the concentration  of ozone  transported into the urban area
 along the surface for the given day.
      B.   Present Transport of  Ozone Aloft
           As noted above, it  appears that unscavenged  ozone, transported aloft
 is likely to have a far greater impact  than surface transport on maximum
 afternoon ozone levels observed within  or downwind from cities. Thus,
 estimates of ozone aloft .are needed for control strategy development with
 OZIPM4/EKMA.  Techniques for estimating the level of ozone transported aloft
 have been the subject of two studies (Chan, et  al, 1979;  and Eaton, et al,
 1979).   Five different techniques, which  were considered to be feasible, were
 field tested in Philadelphia during the summer  of 1978 "(Chan, et al,  1979).
 The five methods are:   (1)  use of fixed ground  based stations; (2) use of
 airborne measurements in a dedicated aircraft;  (.3) use of airborne-
 measurements with a portable instrument package; (4) use  of free lift bal.loon
 soundings;  and  (5)  use of soundings by tethered balloon.   Chan,  et al (1979),
 contains a  detailed description of each of these techniques as well  as a
 discussion  of the findings  of the  study.  Of the five measurement techniques
 evaluated,  surface  measurements at fixed sites,  airborne  measurements by
 dedicated instrumented aircraft, and  soundings by ozonesonde  beneath  a free
                                       25

-------
 balloon were judged to be practical means of providing information on ozone

 transported aloft.

      During the summers of 1985 and 1986, measurements of ozone aloft were

 made over six cities using aircraft.  Cities involved in this analysis were:

 Dallas, Texas;  Tulsa,  Oklahoma; Atlanta, Georgia; Birmingham, Alabama;

 Philadelphia,  Pennsylvania;  and New York, New York.  Ninety percent of the

 ozone aloft values from this study fall  between approximately 25 to 60 ppb.

 [Baugues,  1987].   Thus,  measurements at  other sites should be near this range.

 For those  cities  located in  the ROMNET domain,  an alternative procedure is

 being developed.   Present and future aloft values for NMOC,  NOV,  CO and 0,
                                                               X          o

 will  be based upon results from the ROMNET simulations.   Exact procedures and

 data  bases  will be avail-able  in mid-FY-90.

      Recommended  Procedure:   In selecting this  recommendation,  consideration
 has been given to  such factors  as  technical  capability and  available  funding,
 and the intended  use of  the data.   Ozone measurements taken  on  the day being-
 modeled are recommended  as the  best estimate of ozone aloft.   These
 measurements should be obtained at surface  monitoring sites  upwind of the city
 during  the  first  hour after breakup of the  nocturnal  inversion.  An acoustic
 radar (sodar^can  be used to  determine the  time of  inversion  breakup  for  the
 day.   If the time  of the  breakup of the  nocturnal radiative  inversion is  not
 known,  the  1000-1200 Local Civil Time  average ozone concentration  recorded at
 the upwind  monitor should be  used  as the transport  estimate.  A major
 advantage of surface measurements  is that  it is the only method which provides
 continuous  measurements  and,  thus,  assurance that measurements exist  for  days
 or for  times of day which are later determined  to be  of interest.   The sitefs)
 should  be located  in as rural a location as  possible so as not to be
 appreciably affected bv  local sources  of precursors.  The distance such
 upwind  sites should be located  from a  city depends on the extent of urban
 development.  Because it  is desirable  not to measure  pollutants which are
 recirculated from  the citv under review,   a distance of 40 km  or more  upwind
 from  the urban core should be sufficient.  This  distance perhaps could be
 reduced for smaller cities.   Figure 3-3  depicts  orientations  for acceptable
 upwind  sites.

     Alternate Procedure:  Information on the vertical distribution of ozone
transported above the surface layer  in the early morning may  be used  directly
 if it is available.  Such information might  include aircraft  or free-lift
ozonesonde measurements.   The reader is  referred to Chan,  et  al, 1979, for a
detailed discussion of these techniques.   Use of an alternate procedure is

                                      26

-------
 subject to the approval  of the EKMA contact person in the appropriate U. S
 EPA Regional  Office.

      Missing  Data:'  In the event that an estimate of transport is not
 available for a given day being modeled, the median transport value from the
 remaining days being  modeled should be used as a default value. This procedure
 applies to all the data gathering techniques described above:  fixed site,
 aircraft,  and ozonesonde observations.

      C.   " Future Transport of Ozone                                        -

           If  control  programs are implemented in upwind areas,  ozone  "

 transported into the  city may be reduced.   However,  in most  cases,  the source

 area  and  the  level  of future controls  are  not likely to be known  to any degree

 of  certainty.

      Recommended Procedure:   Because of  the considerable uncertainty .in  the
 location  and  future control  levels  of  the  source area(s)  for ozone  transported
 into  the  urban  area,  the  relationship  depicted  in  Figure 3-4 is recommended
 for estimating  the future  ozone  transport  level  given  the level of  present
 transport.  The  solid  curve  in  Figure  3-4  was derived  on  the basis  of changes
 in VOC emissions which are projected assuming a  national  mix of source
 categories; national  estimates  of projected, growth  in  stationary  source
 emissions  and  vehicle  miles  traveled;  anticipated  impact  of  applying
 reasonably  available  control  technology  to  stationary  sources and the impact
 of the Federal Motor  Vehicle  Control Program on  mobile  sources; and
 consideration of natural background levels.   It  was assumed  that  future  ozone
 levels would not exceed the  NAAQS.  The  sol id curve  is  most  appropriate  for
 use by cities subject to  impacts from  large  upwind nonattainment  areas.  The
dashed curve  is most  appropriate for use when a  city is  isolated  and  not
 impacted by large designated  nonattainment  areas.

     Future ozone transport  levels can be-computed by  use  of the  following
equations:

     For areas with large designated nonattainment area upwind:

          03 (future)  - 0.7 * (03 (present)  - 0.04) + 0.04

     For isolated areas:

          03 (.future)  - 0.9 * (03 (present)  - 0.04) + 0.04

     Where:

          03 (future)  = future ozone transport level (ppm)

          03 (present) = present ozone transport level  (ppm)

                                      27

-------
                                                   (8)
                                                     '  STAGNATION
              Denotes upwind area
Figure 3^3   Examples  of acceptable monitoring  locations for estimating
             transported ozone.
                                      28

-------
      The coefficients "0.7" and "0.9" in the preceding expressions were  '
 obtained by reviewing OZIPM4 runs with varying conditions.  The preceding
 expressions assume an irreducible background component of 0.04 ppm.
      Without information to the contrary, future transport along the surface
 should be-assumed equal  to zero.  If significant nonzero concentrations are
 found for present ozone transport along the surface, then future ozone
 transport levels should be obtained using the relationships shown in
 Figure 3-4.               "  '             . '           . .       -
      For those cities located  in the ROMNET domain, an alternative procedure
 is  being developed.   Present and future aloft values for NMOC,  N0y, CO and 0-,
 will  be based upon results from the ROMNET simulations.   Exact  procedures and
 data  bases  will  be available in mid-FY-90.
      3.2.6   Precursor Transport
           Just as for ozone, precursor pollutants could  be transported in
 both  the surface layer and aloft.   However,  outside urban areas,  the surface
 layer is expected to  be  very shallow.   Thus,  long-range  transport of
 precursors  in the surface  layer may not  be  significant.   Transported precursor
                                                                 i
 concentrations tend to be  substantially  less  than concentrations  within  urban
 areas  (EPA,  1978).  Recent measurements  of  NMOC  aloft  over six  cities
 indicates that most NMOC aloft  values  fall  within a range'of  0-50  ppbC.   The
 overall  median value  for these  data is 30 ppbC  [Baugues,  1987].   Future  NMOC
 aloft  should  be  reduced  20 percent  from  present  levels.   Present  and future
 levels  of NOX aloft should be set to 2 ppb  (.002  ppm).   Surface levels of NMOC
 and NOX for both  present and future conditions should  be  set  to zero.
     Carbon monoxide  levels  at  the  surface  should be set  to. zero.   Concen-
 trations a-loft should  be set to  0.5 ppm.  Future  levels of  aloft CO  may  be
 reduced  20 percent from  present  levels.   For  those  cities  located  in the
ROMNET domain, an alternative procedure  is being  developed.   Present and
future aloft values for  NMOC, NOX, CO and 03 will be based upon results from
                                      29

-------
                                                                        04

                                                                        6



                                                                        to
                                                                        T~
                                                                         •
                                                                        o
\
  \
    \
      \
        \
                                                                       CD
                                                                       t—
                                                                         •
                                                                       O
                                                                    -
                                                    CO  g

                                                    3  o.
                                                    o  rsj
                                                        O

                                                    10  f-
                                                 -  o
                                                     •
                                                    o
                                                                    L q
                                                                      o


                                                                      CM
                                                                   I- o
                                                                       •
                                                                      o
CM    i-

o"    d-
             O)
             O
CO
o
q
o"
to
o
m
o
                        o
                         •
                        o
                  q
                  CD
                                                      q
                                                      o
i-    o
o.
                                                                           LJ
                                                                           tn
                                                                           LU
                                                                           o:
                                                                           Q.
                                                                            tL   §
                                                                            D.   5
                                                              en
                                                              s-
                                                              ID
                                                         Q-   5
                                                         tn   i
                                                         2   "
                                                         <   (O
                                                         2   |
                                                              O)

                                                              re

                                                             "o
                                                                                                    O
                                                                                                    a.
                                                                     ui
                                                                     c;
                                                                     o.
                                                                     O


                                                                     o
                                                                                                ce:
                                                                                                =>
                                                                                                CD
                                                                                                    D:
                                                                                                    o
                                                                                                    a.
                                                                                                    oo
                                                                                 O
                                                                                 r-j
                                                                                 o
                                  3NOZO
                                         30   •

-------
 the ROMNET simulations.   Exact procedures and data-bases will be available in
 mid-FY-90.
      Recommended Procedure:   Transported concentrations of NOY and NMOC in
 the surface layer should be  set to zero.  The recommended default NMOC aloft
 value is  30 ppbC based upon  recent data.  Present and future levels of NOX
 aloft should be set to 2 ppb.   Carbon monoxide levels aloft are recommended to
 be  set to 0.5 ppm.   Future levels aloft CO may be reduced 20 percent from
 present levels.   The reactivity-of NMOC aloft will  be discussed in-the  .
 following section.
      3.2.7   Organic Reactivity
           A.    Surface NMOC
           The fundamental  concepts underlying the treatment of organic
 reactivity  in the CB-4 mechanism were described in  Section 2.1.   As noted  in
 that  section,  the organic  reactivity  input that is  required by OZIPM4  consists
 of  specifying a  set of apportioning factors,  or as  they are more  commonly
 termed, carbon-fractions.  Specification of these fractions permits the  OZIPM4
 program to  apportion  total NMOC  concentration" into  the  individual  carbon
 groups—PAR,  ETH, OLE, ALD2, TOL,  XYL,  FORM,  ISOP,  and  NR.   (The  apportioning
 procedure is  carried  out within  the model  for  the NMOC  concentrations  that
 occur  both  initially  and as a  result  of  subsequent  post-0800  emissions.)
     Two basic approaches  are  possible for estimating the  carbon-fractions.
The recommended approach consists  of  using a set  of default fractions  that
have been derived through  analyses of available ambient organic species data,
and review of pertinent, scientific experimental  results.  The second, an
alternative approach,' requires the analysis by gas chromatography  (GC) of
 individual organic species' concentrations  in ambient air within the city
under review.  Typically, this latter approach requires a special field study.
                                      31
                                      f

-------
      The recommended approach of using a default value rather than making a
 city-specific determination arises primarily as a consequence of two factors.
 First,  the default values are estimated to be representative of typical urban
 reactivity based on an analysis of ambient sampling results conducted in a
 number  of locales (Jeffries, 1987).  While some city-to-city variations in
 organic composition are to be expected, the default recommendations should
 adequately represent most -United States cities (Jeffries, 1987).   The second
 factor  relates to the resource requirements associated with the alternative
 approach.  The cost of conducting a special  ambient sampling program can be
 substantial.
      Instead  of using default values,  carbon-fractions can be computed from
 GC  analysis of ambient  samples.   Monitoring considerations in performing GC
 sampling/analysis are discussed  by Singh  (1980)  and  EPA (1980),  and will  not
 be  repeated here.   However,  it should  be  noted  that  GC analysis  is  not an
 automated  technique,  and  is  most often done on  a special  study basis.   Thus,  a
 monitoring program of limited duration is the most pragmatic  approach  for
 developing the  information needed  to compute carbon-fractions.  While  it  is
 difficult to  prescribe  exactly the  number of samples needed,  enough  should be
 analyzed to ensure  that representative, average  carbon-fractions can be
 computed.  Since  the  carbon-fractions  will  be used to  apportion initial
 concentrations and  concentrations due  to fresh emissions*  the most appropriate
 sampling period is  one prior to  the onset of significant  photochemical
reaction. As with NMOC monitoring,  sampling during the 6-9 a.m. time period
within the area of maximum emission density (i.e., usually the center city)  is
                                      32

-------
 generally recommended.  Ambient  samples  for  GC  analysis  should  be  collected  by
 Integration over a period of 3 hours.
      The GC analysis must identify all species  up  to  and including C-12
 '(compounds containing 12 carbons).  Identification of any peak  over 0.5 parts
 per billion (ppbC) is required.  Where an  individual  peak cannot be identified
 as a specific compound, it must  be analyzed  to  determine the carbon number and
 the class (paraffin, olefin,- or  aromatic)..
      Regardless of the technique employed  in their derivation,  the carbon-
 fractions are used to apportion total concentrations  of  organic compounds
 which are based upon ambient measurements.  Of  the two organic  compound
 monitoring techniques [i.e., PDFID (preconcentration  direct flame  ionization
 detection) and GC], both rely on a flame ionization detector,that  is
 relatively inefficient in responding to many oxygenated  compounds  such as
 aldehydes and ketones (i.e., these techniques measure hydrocarbons  only).  SAI
"has estimated that, initially,  total  carbonyl compounds  (i.e., those including
 aldehyde and ketones, as well  as-some surrogate carbonyls) are about 5 percent
 of total nonmethane hydrocarbon concentrations  (Killus and Whitteri, 1983).
 Only about 1 percent of the  total carbon that is measured can be classified in
 the carbonyl  group (i.e.,  surrogate carbonyls).   The remainder of the
 carbonyls (i.e.,  5 percent of the nonmethane hydrocarbons that are measured)
                                                          I     :
 is attributable to oxygenates  that are not detected.   The carbon-fractions
 which would  sum to 1.05  (or  105 percent)  are then adjusted so that, they total
 only 1.00 or 100  percent.  If  ambient  measurements  of aldehydes  are available,
 a  city-specific determination  of the  carbonyl fraction can be made. However,
 these measurements  tend  to be  complex,  using techniques that are mostly
                                       33

-------
 conducted by research groups.  As a consequence, carrying out a special
 aldehyde monitoring program cannot be routinely recommended.
      Whenever city-specific estimates are made by the techniques discussed
 above,  special  care must be taken to ensure that the computed carbon-fractions
 represent a realistic distribution of NMOC species.
      Reactivities of NMOC in numerous cities have been.computed based upon
 KOH values-   KOH values are rate constants which give a measure of the
 reactivity of a class of compounds with  OH radicals.   The weighted sum of
 these KOH values give an estimate of the overall  reactivity of the NMOC mix.
     The  KQH value for an NMOC  mix can be determined  using the following
 equation:
KQH = PAR *  1203 +  ETH *  5824 + OLE *  20422  + ALD2  *'11833'+  TOL  *  1284  +
          XYL * 4497 + FORM  * 15000
     Where:.    KQH  is the average  KQH  value  for  the NMOC  mix,  PAR is  the
               fraction of the mix considered paraffin  (based  upon  the CB-4
               splits).
     A typical city is expected to have an average  KQH  value  that falls
between 2700 and 3600 min"1.  If the computed KQH value,  based upon a city-
specific NMOC distribution, does not fall within this range, the  process of
determining the city-specific distribution should first be redone to ,check for
errors.   If no errors are found,  the default reactivity should be utilized.
     Requests to use reactivities  other than the default must be  reviewed and
approved by the appropriate Regional  Office  in cooperation with the Model
Clearinghouse.
                                      34

-------
      Recommended Procedure.  The carbon-fractions recommended  for  use  in  an
 EKMA/CB-4 analysis are listed below:

                           PAR    =  .564
                           ETH    =  .037
                           OLE    =  .035
                           ALD2   =  .052
                           FORM   -  .021
                           TOL    =  .089
                           XYL    =  .117  .      -                            -,
                           ISOP   -  0        '-
                           NR     =  .085

 They should normally be used unless sufficient information is  available to
 derive city-specific information by the method discussed below.

      Alternate Approach.  If analyses of ambient air samples by gas
 chromatography are available for a particular city, the results can be used to
 derive carbon-fractions.  The ambient samples should be taken  in the high
 emission density area (normally the urban core) within the 6-9 a.iiu Local
 Daylight Time (LDT-) period during the ozone season. Integrated samples are
 required.  It is desirable that enough samples be analyzed to provide a
 representative average.   For supplemental-information regarding monitoring
 aspects, the reader is referred to Singh (1,980) and Rhoads (1987),  and for
 details on how carbon-fractions are computed from the sampling results, the
 reader is referred to Appendix C of this document.  Those considering this
 approach should discuss  it with the EKMA contact in the appropriate Regional
'Office.

      Caveat>  If the alternative approach  is^sed, the resultant reactivity
 must fall within the range of 2700-3600 min"1.  If it does not, it  is strongly
 recommended that the data and computations  be thoroughly checked to ensure
 that no errors have been introduced.   If the problem cannot be resolved,  use
 of the default carbon-fractions listed in  the recommended procedure above is
 preferable.

      B.   NMOC Aloft'                                     '   '

      OZIPM4 also requires carbon-fractions  for the NMOC aloft.   The

 recommended approach is  to' use  the  -default  value provided.   These values  are

 based upon GC speciation of aircraft  samples  taken over six cities  during the

 summers of 1985  and 1986 [Baugues,  1987].   In order to develop  a city-specific

 distribution,  a  special  field study would be  required.   Such  an analysis  is .

 not  recommended.
                                       35

-------
      For those cities located in the ROMNET domain, an alternative procedure
 is being developed.  Present and future aloft values for NMOC, NCL, CO and 0,
                                                                  x          o
 will  be based upon results from the ROMNET simulations.  Exact procedures and
 data  bases will  be available in mid-FY-90.
      Recommended Procedure.   The carbon-fractions recommended for use in
 EKMA/CB-4 for NMOC aloft are:
PAR
ETH
OLE
ALD2
FORM
TOL
XYL
I SOP
NR.
_
=
=:
=
=
=
=
=
.498
.034
.020
.037
.070
.042
.026
0
.273
     3.2.8  Temperature
          Hourly  temperature  data must  be  utilized  in  OZIPM4.   Use  of hourly
temperatures allows reaction  rates to be increased  or  decreased according  to
the hourly temperature.  If not specified, OZIPM4 uses  a default temperature
of 303°K.  The hourly surface.temperatures to be utilized,  in OZIPM4 should be
from an urban meteorological  station.   Tapes which  provide complete hourly
data are available from the National Oceanic and Atmospheric Administration
(NOAA) in Asheville, NC.
     Recommended Procedure.  Hourly surface temperatures from an urban
meteorological  station are recommended for use in OZIPM4.
                                      36

-------
       3.2.9 Water Vapor
             Recent work has shown that ozone predictions are sensitive to the
 amount of atmospheric moisture content.  A new option has been included in
 OZIPM4 which will  estimate the atmospheric moisture content given relative
 humidity values and an ambient pressure level.  Hourly values of relative
 humidity can be found on meteorological tapes available from NOAA (in
 Asheville,  NC).
       Recommended  Procedure.   Hourly relative humidity values are recommended
 for  use in  OZIPM4.  .
       3.2.10 Biogenic Emission Estimates
             OZIPM4  has recently been modified to contain an option to allow
 inclusion of biogenic emission rates.   The inputs to OZIPM4 are  emission
 estimates of the biogenics, typically broken out as:  isoprene,  a-pinene,
 monoterpenes and unknowns.  The units  for these  values are  kilograms  per
 square kilometer per  hour (kg/km2/hr).
       The biogenic  emission estimates, are sensitive  to several meteorological
 parameters:   air temperature,  wind  speed,  relative humidity and  cloud cover.
 Therefore,  biogenic emission estimates  must  be developed for each  day modeled
 and the  values  based  upon day  specific  meteorological  parameters.
       EPA will  provide  a  computer program that can be  run on an  IBM-PC (or
 compatible machine) which will  estimate  biogenic  emissions  rates on a county
 basis.   The  user would  need, to  provide  day specific  meteorological parameters.
This program should be  available by mid-1990.
      As with man-made  emissions (Section  3.2.3),  emission  rates should be
 included for the county in which the straight  line trajectory  is over for each
hour.
                                      37

-------
       OZIPM4 also requires initial values for biogenic species.  All  initial
 values should be set to 0.0001 ppm in the absence of measured concentrations.
 They should not be left.at zero, as this may cause the program to "hang up".
       3.3   Predicting Peak Ozone
             In one study,  it was found that EKMA could yield a lower control
 estimate for a case when  peak ozone is underpredicted as compared to one in
 which  peak ozone is more  accurately predicted (Jeffries, et ,al,  1981). In
 extreme  cases of underprediction,  a solution may not be possible with EKMA.
 In  a similar fashion,  a large overprediction could lead to a control estimate
 that is  higher than that  obtained  when good agreement is found.   As  part of
 the  supplemental  EKMA  guidance issued  in December of 1981,  EPA addressed this
 problem  by recommending that  predictions of peak ozone be made,  and
 appropriate adjustments or compensations be made if poor agreement  is found
 (Rhoads,  1981).   In this  section,  the  procedures for making the  predictions,
 comparing  them with observations,  and  making appropriate  adjustments are
         •
 described.
      3.3.1  Procedures  for Making  Ozone  Predictions
             In  Section  3.2, most of the  OZIPM model  inputs  that  are-  needed
 either to  predict  peak  ozone  or to estimate  VOC  emission  reductions  were
 discussed.  In  order to  make predictions  of  peak  ozone,  one  additional  set of
 model input  variables  is needed:  the  concentrations  of NMOC,  N0vy and CO that
                                                                 X
 are representative  of the  initial (i.e., 8  a.m.), urban core levels.  These
model" inputs are the most  critical for making predictions,  and should be
estimated on a case-by-case basis.  Because of the model sensitivity to these
 inputs, use of mean or median values compiled from measurements taken across a
                                      38

-------
 number of days may lead to erroneous results.  Thus, day-specific measurements
 should normally be used to make these estimates.
      As for estimating the initial concentrations, the recommended procedure
 is to make use of ambient NMOC, NOX, and CO measurements routinely taken in
 the urban core,.but which represent neighborhood scale levels.  The initial
 concentrations are intended to represent the NMOC, NOY, and CO that is
                                                      A
 initially present within the mixed layer at the start of the model simulation
 (i.e.,  8 a.m.).  While several approaches could be taken,  the recommended
 method  is to use the 6-9 a.m.  average concentration measured by collocated
 NMOC, NOX,  and CO monitors within  the urban core which represent neighborhood
 scale values.   If more than one set" of measurements are available from several
 such monitors,  then  the 6-9 a.m.  average concentration at  each monitor should
 be averaged  to obtain an overall,  urban average NMOC,  NOV,  and CO
                                                         A
 concentration.   Algebraically,  the above procedure can  be  expressed  as
 follows:
     '(NMOC)o
      (NMOC)6-9
                                             1  (3-4a)
                          N
                     and
      '(NOx)o
 n
 2
1-1
and
     '(C0)o
                1=1
'(NOx)6-9
                           N
      C(CO)6-9
                           N
(3-4b)
                                               (3-4c)
                                      39

-------
                  where
 .,„._  .,ln             (CNMOC)o,  (cNOv)o,  (CCO)o    -initial  concentrations of
 NMOC, NOX, and CO (in units of ppmCT, ppm, and ppm, respectively)  input  to
 OZIPM4 simulation

                   , .  C(^NMQC)6-9]i,  [(cNOx)6-9]i   -the 6-9  a.m.  average
                      [rCO)6-9]i           concentrations of NMOC, NO  ,  and CO
                                           (in units of ppmC, ppm, and ppm,
                                           respectively) taken in  the urban
                                           core (or high emission  density area)
                                           at site i

                                           N  =  total number of collocated

 monitors for which day-specific NMOC and NOY measurements are available.
                  '        '•  <  .              **

      As noted above,  the initial  NMOC and NOY concentrations are derived from
  i                                         i  f^

 day-specific measurements of  NMOC1 and NOX.  In some instances,  an NMOC

 measurement  may not be available  for the day being modeled.  In  such a case,

 the  initial  NMOC concentration  can be approximated  by making use of the

 median  NMOC/NOX ratio (see  Section 3.4.3)  and a  day-specific measurement of

 NOX  alone, provided it -is available.   The  initial  NMOC concentration for use

 with  the  OZIPM4 simulation  can  be computed as the  product  of the median

 NMOC/NOX  ratio  and  initial  NOX  concentration,  or
                          r    •    '  '   • r
                          (LNMOC)o =     (LNOx)o  (NMOC/NOX)           (3-5)

                where

                          (CNMOC)o =     the initial  NMOC  concentration for
                                         the OZIPM4 simulation, ppmC
                          C
                          (  N0x)o   =     the initial  NOX concentration
                       .                  calculated by equation 3-4b, ppmC

                       (NMOC/NOX)   =      the  median NMOC/NOX ratio as derived
                                         according  to  the  procedures  outlined
                                         in Section 3.4

     It should be emphasized that  this approach is an  approximation,  and the

one described in the preceding paragraph is preferable.
                                      40

-------
      With the estimates of initial  NMOC,  NOX,  and CO,  and the corresponding
 day-specific inputs listed in Table 3-1,  the CALCULATE option of OZIPM4 may be
 used" to perform a single model  simulation.   An example simulation and
 additional  information are contained in EPA, 1989.   Thus, no additional
 discussion  will  be Included here.
   .   Recommended Procedure.  The CALCULATE  option of the OZIPM4 program
 should  be used to predict peak  ozone for  comparison with the observed peak.
 The  model  inputs discussed in Section 3.2 should be used, with initial
 concentrations that have been derived according to  equations 3-4a,  3-4b,  and
 3-4c, using data that  are specific  to the .day  being modeled.  In the event -
 that day-specific NMOC measurements are unavailable,  the initial  NMOC
 concentration can be approximated by means  of  equation 3-5,  with the
 recognition that some  uncertainty may be  introduced in the analysis.   If  day-
 specific  measurements  pf NMOC,  NOX,  and CO  are not  available,  predictions of
 peak ozone  cannot be made.   In  this case, computation  of VOC control  estimates
 are  recommended,  but without  the requirement of reasonable agreement  between
 prediction  and'observation.
      3.3.2  Comparisons of Predictions With  Observations
          The principal  output  of concern obtained  with a CALCULATE
 simulation  is the predicted ozone.   By numerically  integrating the
 differential  equations  describing ozone formation processes  (i.e.,  chemical
 reaction, emissions, dilution,  etc.),  instantaneous concentrations  of ozone
 are  computed  throughout  the simulation  period.   From this computed  profile of
 instantaneous  ozone  concentrations,  the OZIPM4  program.calculates the hourly
 average concentrations  occurring during the  model simulation.   The  predicted
 ozone concentration  that  occurs, at  the  time  of  the  observed  peak  is used  in
 the  performance measure  that  is  recommended  to  evaluate  model  performance.
This performance measure  is the  relative deviation  of  the prediction  from the
observation, or
                          DEV  =   C   -  C0
                                _£	»	x  10Q             (3-6)
                                     Co
                                      41

-------
                     where

                DEV  =    deviation...of the model prediction from the
                          observation, percent

                 Cp  =    maximum 1-hour average predicted peak ozone, ppm

                 CQ,  -    observed peak ozone,  ppm

 If  the  relative deviation is found to be no more than ± 30 percent, then

 agreement  between  the prediction and  the observed peak is judged to be

 sufficient to  proceed with  control  estimate calculations.  If the deviation is

 outside the ±  30 percent  range,  a comparison between the measured peak and

 predicted  peaks 1  hour before (or after) the time of the observed peak should

 be1made.   Due  to the  uncertainty in trajectories,  it is possible for the time

 of  the  predicted peak to  be off  by an  hour.

      If the model  underpredicts  by more  than 30 percent (i.e.,  DEV < - 30

 percent) or overpredicts  by more than  30 percent (i.e.,  DEV  > + 30 percent),

 the review of,  and possible adjustment to,  key  model  inputs  according to the

 discussion  of  Section  3:3.3 below is warranted.   R  should be noted that the

 observed ozone  peak (not  the  predicted)  is  recommended  for subsequent control

 calculations.           .                         '

     Recommended Procedure.   The  relative deviation  of  the model  prediction
 from the observed peak  should be  computed according  to  equation 3-6  above.
 The model predicted peak  to be used in this  computation  is the  hourly average
 ozone concentration calculated by the  OZIPM4 program at the time of  the
 observed peak.    If the  computed "deviation is within ± 30  percent,  then the
model  results are sufficiently accurate  for  control estimate  calculations.  If
the deviation  is outside  the ± 30 percent envelope comparisons between the
measured peak and the predicted values 1 hour before (or  after) the  time of
the measured peak should  be made.   If  the ± 30 percent test is not met,  then
the procedures  discussed  in Section 3.3.3 should be applied, in an  attempt to
 improve the simulation results.
                                      42

-------
      3.3.3  Review and Adjustment to Model  Inputs
           If  inadequate agreement between  a model  prediction  and  an  observed
 peak is found, review of the model  inputs  should be conducted.  The  objective
 of this review is to investigate whether some modifications to key model
 inputs can be justified on some physical basis in  order to improve the model
 predictions.  This review should focus on those model inputs that most
 critically affect predictions of peak ozone.  Of most importance are the
 initial NMOC,  NOX, and CO concentrations; dilution; and post-0800 emissions.
 Adjustment of these inputs,  within the uncertainties associated with their
 development, is warranted if improvements in model  predictions can be made.
 Obviously, any errors that may have been made in their derivation should be
 corrected as we!1.
      While specific recommendations in  trouble-shooting poor model perform-
 ance  are difficult to make,  some general  guidelines can be made depending on
 the nature of  the  problem, be it an  underprediction or an  overprediction.  If
                                                            «              i
 some  uncertainty exists with  regard  to  the  data  from which they were  derived,
 then  the inputs can be adjusted  within  that range of uncertainty.   In general,
 initial  NMOC,'NOX,  and CO  levels may be  adjusted by ±  15 percent  and  maximum
 afternoon  mixing height by ±  200 meters  (Seila,  1986 and Rhodes  and Evans,
 1986).   Outliers in the data  may be  deleted if adequate justification is
 available.   For example, assume  that an  initial NMOC concentration had  been
 derived  from 6-9 a.m. measurements taken  at three monitoring sites.   However,
 the 6-9  a.m. concentration at one site (say site A)  deviated substantially
 from the concentrations measured at  the other two sites (say sites B  and  C).
Then,  improved agreement between"model predicted and observed ozone might be
                                      43

-------
 found if the initial NMOC concentration for the day in question was derived
 solely from the measurements taken at the two sites in agreement (i.e., sites
 A and B).   Consider the case of underproduction first.  The following steps
 should be  taken:
      1.    Check inputs for errors (especially morning and afternoon mixing
 heights  and ozone aloft).                                                 ~-
      2.    Increase the initial  NMOC,  NOX and CO concentrations by 15 percent.
      3.    Reduce afternoon mixing height by 200 meters.
      4.    Increase original  .afternoon mixing height by 200 meters (may
 improve  situations where  ozone  aloft  is  high).
      5.    Increase morning mixing height by 50  meters.
      All steps  are cumulative,  except for 3/4,  where the  step that  improves
 the situation should be  included  with Step 5.   Steps are  to  be followed  in the
 order above, and carried  out only until  the deviation  is  within the  ±  30
 percent  range.   Further adjustments should not  be  carried out to reduce  the
 deviation.
  1
  1  '  When  changes  are made to the morning  mixing height,  make sure that
 changes  are reflected in  all  options.  The following options  use the morning
 mixing height:   DILU  (Dilution),  MASS (Emissions),  CRED (CO)  and BIOG
 (Biogenics).
      Guidelines  for  correcting  a  problem of  overprediction are similar in
 concept to those for  underprediction.  The following steps should be taken:
      1.  Check  inputs for  errors  (especially morning and  afternoon mixing
 heights and ozone aloft).  Also make  sure that all   emission rates are being
                2
read  in as  kg/knr/hr  and not  in fractions of the initial  concentration.
                                      44

-------
       2.  Reduce initial NMOC, NOX and CO concentrations by 15 percent.
       3.  Increase afternoon mixing height by 200 meters.
       4.  Reduce original afternoon mixing height by 200 meters (may improve
  cases where ozone aloft is high).
       5.  Reduce morning mixing height by 50 .meters.
       The  same rules  apply that were  described earlier for the underprediction
  case.
       As noted-above,  the model  inputs that  substantially affect model
  predictions of  peak ozone  include the initial NMOC, NO  ,  and  CO
                                                       A
  concentrations, and initial, mixing height.   The possibility exists that mass
 balance techniques could be used  to evaluate the appropriatenessiof a
 particular set of initial concentrations and an initial mixing height.  For.
 example, one could test by means of a simplified box model whether or not a
 city's emissions are sufficient to generate the measured initial
 concentrations within a mixed layer corresponding to the postulated initial
 mixing height.  While such an approach is intuitively appealing,  such
 calculations may not  be able to account properly for ventilation,  and  for
 advection  of pollutants from source areas nearby the precursor monitors.
 Nevertheless,  it does  provide  one  means of assessing the  reasonableness  of the
 postulated model  inputs.
mnft  Recommended Procedure.  To  improve agreement with observed 0-3  levels
so canTcited"1 b^npv^d'^th in  reasonable  ranges  if justification for doing
l»Arn       f    sPeciflcally, the key model  inputs are  initial NMOC  NO
be aSiUSpf^hati0+nhS'  and miXlng  heights'  Fina11*' model  inputs shouldlnly
c.th i£fjed W]thl" the  ra?9e of reas°nable uncertainty, and not just selected
obtained  ^Ll^T* bet??? the mode1 Prediction and observed peak ts
should stirS IV!  ac"ptable Demerit cannot be found, control estimates
should still  be made and the procedures discussed in Section 3.5 applied.
                                      45

-------
 3.4  Computing VOC Emission Reductions



      The recommended procedure for. computing VOC emission reductions  is to



 use the EKMA option in OZIPM4.  Use of this option eliminates the need to



 generate isopleth diagrams.  OZIPM4 performs the necessary calculations and



 determines the VOC emission reduction.  Several.variables are needed to



 perform this calculation.   These include:  "initial  NMOC, initial NOY, initial
                                                                    A


 CO,  NMOC/NOX ratio,  maximum observed ozone, present and future levels of NMOC



 aloft,  NOX aloft,  03 aloft, and assumptions regarding future levels of NOX and



 CO.   Use of this  option is discussed further in EPA,  1989.



      3.4.1  Derivation of  Empirical  Data



           Two pieces of empirical  data are needed for calculating control



 requirements.   The first is the maximum 1-hour average ozone concentration



 observed at the site of interest.   The degree  of emission control  necessary to



 reduce  this "peak"  to  0.12 ppm is  to be calculated; hence,  the peak level  will



 be termed  the  daily  site-specific  ozone maxima.

                                                                              i


      The second piece  of information needed is  the  NMOC/NOV  ratio.   This
                                                          J\


 ratio is derived from  the  6-9  a.m.  concentrations of  NMOC and  NOV  within the
                                                                 X


 urban area.  The ratio  will  be  termed  the- design ratio.   The procedures for



 deriving both  the daily site-specific  ozone control values and the  design



 ratios  are  described below.



     3.4.2   Daily  Site-Specific  Ozone  Control Value



          The  daily  site-specific  ozone  maxima  is used  in conjunction  with



 the NMOC/NOX ratio for  calculating control  estimates  needed to  reduce  the day-



 specific and site-specific  observed  peak ozone to 0.12 ppm.  The daily ozone



maxima should be expressed  in ppm units rounded  to two decima-1  places.
                                      46

-------
      Recommended Procedure:  A daily site-specific ozone maxima  is obtained
 for each site which is downwind of the city, and/or within the city  in the
 case of light and variable winds on the day for which the control strategy is
 to be developed.  Surface wind data should be examined to assure that the site
 is not "upwind" of the city.  Based on the'results of field studies  and
 reviews in which ozone gradients dov/nwind from urban areas were examined, peak
 ozone concentrations should generally be observed within 15-45 km downwind of
 the central business district (EPA, 1978; Martinez and Meyer, 1976;  and EPA,
 1976).
      3.4.3 NMOC/NOV Ratios
                   A
           The prevailing 6-9 a.m.  LCT NMOC/NOX ratio measured in the urban
 core of the city is the second piece of empirical  data required.   The design
 ratio is viewed as characteristic  of the.city which would prevail during the
 remainder of the morning and early afternoon in the, absence of chemical
 reactions.   OZIPM4 expresses peak  ozone concentrations as a function of  the
 initial  concentration  of NMOC and  NOX.   Thus,  the  6-9 a.m.  LCT NMOC/NOy  ratio
 is  considered to be the appropriate ratio for use  in  OZIPM4 since this ratio.
 is  consistent with the conceptual  basis of the  model  (Dimitriades,  1977).  To
 ensure that representative ratios  are  obtained,  the NMOC and  NOY  instruments
                                                                J\
 should be  collocated in the central  core  of the urban area.   The  site(s)
 should be  located  in an area of relatively uniform emission density  and  not
 significantly influenced  by any individual  source.  More  detailed guidance on
 siting  NMOC instruments is  contained in  EPA,  1980.  Guidance  on the  operation
 of  NMOC  instruments  is  available in  EPA,  1985.
      Significant discrepancies have  been  found  between  NMOC/NOX ratios
 calculated  on  the basis of  ambient measurements and those obtained from
 emission inventory data (Drivas, 1978).  Reasons for  the lack  of  correlation
between the two  ratio calculation procedures have not been resolved.  As a
result, only ambient NMOC/NOX ratios should be used with EKMA  since these
                                      47

-------
  ratios are  consistent with the conceptual  basis of the model and the. emission
  ratios have been shown to be poor, surrogates  for these ambient ratios.
       NMOC data analyzed only with the PDFID (preconcentration direct flame '
  ionization  detection) or.GC (gas  chromatograph)  should be used with OZIPM4
  (Rhoads, 1985).. Due 'to the large uncertainty  in  low NOX values,  any day with
  a 6-9  a.m. NOX concentration  at or below 0.020 ppm should be excluded  from' the
  process of estimating NMOC/NO • ratios.
                       1       A                                 '           '
      The NMOC  data  are to  be  collected during the season  of peak  ozone
 concentrations (summer).   Because NMOC concentrations  are apt .to  be relatively
 high in central  urban  locations at those times of the  day (early morning) when',
 these measurements  are required for use in  EKMA more confidence can be placed
 in  the estimate.  Because of the variability .in individual NMOC readings, the
 NMOC/NOX ratio calculated for a single day  is  not  recommended for  use in  city-
 specific EKMA.   Considering instrument reliability and model sensitivity,  the
 following procedure  is recommended for calculating NMOC/NO  ratios.
     .                                 ..                   'A
     Recommended
          A
avail fhio F PH:cursor measurements from more than  one  urban site are
              ««"» -
     3.    The peak ozone level for all  days  with NMOC/NO  ratine -u
                                    48

-------
      4.   Once-the local ozone peaks  have  been  determined  for  all  days  with
 NMOC/NOX ratios, the highest ozone days are selected.  This consists of either
 all days with ozone peaks above the NAAQS  or the top 10 days if less than  10
 days exceed the NAAQS.

      5.   The days selected in Step 4 are  then  ranked and  a median NMOC/NtX,
 ratio determined.     "                                                   ,  x

      It is the median NMOC/NOX ratio that  is to be utilized in OZIPM4. Two
 examples of calculating a median NMOC/NOV  ratios are shown in Table 3-2.
                                         A

 3.5  Selection of the VOC Emission Reduction Target

      3.5.1 Without "Overwhelming" Transport          • '  '

         •  After all  site/day combinations  have been modeled,  the final step    '

 of the modeling analysis involves the selection of the overall  VOC emission
                                   .'  '   •      '                     ' '       '  i
 reduction  target.   In essence,  this  procedure is dictated  by  the form of the ,

 ozone NAAQS,  and is  identical  to  the method recommended  in  the  March  1981

 guidelines document  (Gipson,  1981).   In  summary, a control  target  is  selected

 for each site that permits,  on  average,  one hourly-average  concentration above

 0.12 ppm per  year.  This corresponds to  selecting  the  fourth  highest  control

 level  if 3 valid years of data  are available,  the  third  highest control  for

 2  years of data, and  the second highest  control  estimate for  only  1 year.*

 The overall control target  is then chosen  as  the h.ighest of the site-specific

 control estimates to  ensure  t.hat  the ozone  standard  is attained at  all sites.

      As noted  in the  March  1981 guidelines, an  additional factor that  could

 affect the procedure  just described  is the  consideration of model predictions
       A site is considered to have a valid year of data  if valid daily maxima
exist for at least 75 percent of the days during the ozone season and there  is
no obvious pattern of missing data during periods when maximum ozone is most
nkeiy.  A valid daily maxima exists if 75 percent of the hours In a day
report data and there is no systematic lack of data during times of day when
high ozone is most likely.  The reader is referred to EPA 1979b and FR (March
19, 1981) for further information.

                                      49

-------
 Example 1
           TABLE 3-2.  EXAMPLE CALCULATION  OF THE DESIGN NMOC/NOY RATIO
 Date

 7/1
 6/2
 7/3
 7/4
 8/5
 8/8'
 7/9
 6/10
 7/11
 9/12
 7/15
  Max
 Oo  fPDITl)

  .15
  .15
  .14,
  .14
  .14
  .14
  .13'   '
  .13
  .13
  .13
  .13
 NMOC/NOY  ratio
     Site  1

     6.9,
     7.5
 .   11.3
    14.0
     5.3
     8.7
     9.2
     6.7
     8.4
     9.5
    12.1
 Rank      NMOC/NO.. ratio
                  A
  1         '   5.3
•2     ,       6.7
  3      ;      6.9
  4            7.5 '
  5            8.4       V   :
  6'  ,  ,     ;   8.7<---Median
  7           '912
  8            9.5
  9           11.3
 10           12.1
 11           14.0
Example 2
Date

8/1
7/2
6/3
7/5
8/8
9/9
6/10
7/11
6/12
9/15
 Max
Oo fppm)

 .16
 .16
 .15
 .14
 .13
 .13
 .13
 .12
 .12
 .12
NMOC/NOY ratio
    Site 1
     5.8
     8.4
    12.3
     8.8
     7.8
     7.8
     8.3
    10.1
     9.0
     8.8
Rank      NMOC/NOV ratio
                 A
  1             5.8
  2             7,8
  3             7.8
  4             8.3
  5             8.4
  6             8.6<---Median
  7       .8.8
  8             9.0
  9            10.1
10            12.3
                                      50

-------
 versus observations.  Recall from Section 3.3.2 that a VOC emission reduction
 estimate should not be used when the model predicted peak ozone disagrees with
 the observed peak by more than ± 30 percent.  This does not apply to days
 which do not have day-specific measurement of NMOC and NOV.  However, it has
                                  i                 ,       A
 been observed that substantial underpredictions of base case, peak ozone may
 lead to control  estimates which are too low (Jeffries,  et al, 1981).
'Conversely,  significant .overpredictions of base case, peak ozone may yield
 control  estimates which are too high.  Under some circumstances,  this finding
 enables one  to use control  estimates for those days  in  which base case,  peak
 ozone is poorly predicted.   To illustrate, consider  an  example 'in. which  model
                           "            *                            •
 predictions  and  control  estimates  have been made for a  site with 3 years of
 ozone data  (see  Table  3-3).   Note  that for Day 1,  peak  ozone is  substantially
 underpredicted,  and the control  estimate is the highest,of all days.  If any
 improvements were made to predicted peak ozone,  the  control  estimate for this
 day would likely be increased even  more.  Since the  control  estimate for Day 1
 is  already higher than the  control  target (i.e.,  45  percent),  any improvements
 in  model  predictions would  not affect  the selection  of  the final  control
 target.   Thus, the results  from  Day 1  can be  used, even though the model
 significantly underpredicted  peak ozone.  The converse  situation  occurs  for
 Day 5.   In this  case,  any improvements in model  predictions  would likely
 reduce  the control  estimate for  that day,  again  having  no  bearing on the
 choice  of the final control target.  If the model  prediction  is  poor,  but
 neither  of the situations described  above occur  (i.e.,  overpredictipn  and high
 control  estimate,  or underprediction and low  control  estimate), then it  is
                                      51

-------
               TABLE 3-3.  EXAMPLE  ILLUSTRATING  EFFECT  OF  MODEL  PREDICTIONS
                               ON  SELECTION  OF  CONTROL TARGET
      Observed     Predicted    Relative^
Day   Ozone, ppm   Ozone, ppm   Deviation.
                       Control
     Deviation
observed        x 100
Rank of
                      Estimate. %   Control Estimates
1
2
3
4
5

0.27
0.22
0.20 '
0.18
. 0.15

0.18
0.20
0.22
0.18
.0.21

-33
- 9 .
+10
0
+40
predicted - observed
55 : ' '
47
51 '•
45**
42

tO \*\Jlt 1st \J
1
1
3
2
4
5

                                                                         I i
     Control  Target  =  fourth  highest control  estimate (for 3 years of data)
                                                                        1      I
                                     52

-------
 recommended that the site/day be discarded,  and replaced by the day with the

 next  lowest peak ozone concentration.

      Recommended Procedure.   To obtain the final  VOC emission reduction
 target,  site-specific control  requirements must first be determined.   In
 general,  a  candidate control  estimate  is chosen for each site Abased on the
 number of years  of data and  the statistical  form of the ozone standard (i.e.,
 fourth highest control  for 3  years,  third highest for 2 years,  and the second
 highest  for 1 year).  -Of the  candidate site-specific control  estimates,  the
 highest  one is selected as the overall  VOC emission reduction target.
 However,  all cases in which  predictions and  observations disagree by  more than
 30 percent,  should be discarded,.unless:                 ,      ,

      (1)  peak ozone is  underpredicted  and the VOC reduction estimate  is
 greater  than the candidate site-specific estimate;        '   '
                                                              I  i  •
      (2)  peak ozone is  overpredicted and the VOC  reduction  estimate is lower
 than  the  candidate site-specific  estimate.                      I     '

      In  the event that  a day  is eliminated,  the next lowest peak  at the  site
 in question  should be added for modeling. ',
                                                               i
      3.5.2   Selection of a VOC Reduction Target at  Sites Sub.lect  to
                  Overwhelming  Transport
                                                                  I.
          In Section  3.1,  .we  noted that some days initially selected  may be

 discarded (for modeling  purposes), if  shown  to  be subject to  "overwhelming
                                          i i                 I
 transport."  Nevertheless, a demonstration that the local MSA will  attain1 the
                                                           I ,  i'   •  I      '   '
 NAAQS still needs  to  be  made.   Once  a  day has been  discarded  for  local

 modeling  analysis,  a- determination needs to  be  made concerning  what MSA/CMSA

 is most likely responsible for the excluded  observed daily  maximum.
                                                              i
 Procedures  for doing  this  are  suggested by Meyer  and Baugues ,(1989).   These

 procedures  require  a  review of all surface National  Weather ^ervice (NWS)  wind

 data within 100 miles of the monitoring site  plus any special study surface

wind data collected at properly exposed sites (EPA,  1986).  In  general,  if  the

wind data suggest  an  air parcel located at the  monitor  at the time  of  the

observed maxima may have been  within an  upwind  MSA/CMSA between 8 a.m. to
                                      53

-------
  noon, that  upwind  CMSA/MSA  may  be  instrumental  in causing the observed daily
  maximum.  If the discarded  day  for the  locaj  MSA has  a daily maximum ozone
  level higher than  that for  the  fifth  highest  modeled  day for the upwind
  CMSA/MSA, the discarded day should be included  in the upwind CMSA/MSA's
                                            t
  modeling analysis.  If a discarded daily maxima  is  included  in  an  upwind
  MSA/CMSA's  analysis,  it may be  ignored  in the local attainment  demonstration. '
       In some cases, however,  it may not be possible to  identify the  upwind
  CMSA/MSA most likely responsible for  an observed  daily maximum  ozone
                                        i   i1                        '
  concentration.   If this happens, the  event is referred to as  an  "irreducible
  exceedance."  Presence .of'one or more irred^cibl^ exceedances at a .monitoring ,
•  site has. the effect of raising the local VOC control  target needed to meet the
  NAAQS at that site.  Fbr example, the site-specific control requirement at a
  site with three valid years of data would become the third (rather than the
  fourth)  highest control estimate if there were1 ope "irreducible exceedance" at
  the site.   If the particular site-specific control requirement were the
  highest  amongst all the' sites  assigned io the local MSA, the local  MSA's
                                           ' I        "   '
  overall  VOC  reduction  target would  be slimtlarly affected.
                                       54

-------
                 •  •               REFERENCES


Baugues, K. A.  (1987),  "Support Document for Selection of Default Upper
      Air  Parameters  for EKMA."

'Benkley, C. W.  and  L.  L.  Schulman  (1979),  "Estimating Hourly Mixing Depths
      From Historical  Meteorological Data,"  Journal  of Applied Meteorology.
•  V  -IS,  pp. 772-780.    '  ..   .  .

Carter, W. P. L., A.  M.  Winer  and  J.  N.  Pitts,  Jr.  (1982),  "Effects of
      Kinetic Mechanisms  and Hydrocarbon Composition on Oxidant-Precursor
      Relationships Predicted  by the  EKMA Isopleth  Technique," Atmospheric
      Environment.  Volume 16,  No.  1,  January 1982.
                 i       .   i  >'            .        •              	
Chan, M. W., D.  W. Allard and  I|. Jonibach (1979),  "Ozone and Precursor
      Transport  Intp  an Urban  Area-Evaluation of  Approaches,"  EPA-450/4-79-
      039, U.S.  Environmental  Protection Agency,  Research Triangle'Park,  North
      Carolina.              •

Decker, C. E., et al  (1977), "Ambient Monitoring  Aloft of Ozone and
      Precursors Near  and Downwind of St.  Louis," EPA-450/3-77-009,. U.S.
      Environmental Protection  Agency, Research Triangle Park,  North Carolina.

Dimitriades, B.  (1977), "An Alternative  to the Appendix J Method for
    .  Calculating Oxidant and  NOo Related  Control Requirements,"  International
      Conference on Photochemical Oxidant  Pollutant  and Its  Control:
      Proceedings. Volume II..  EPA-600/3-77-01~6, U.S.  Environmental  Protection
      Agency, Research Triangle Park,  North  Carolina.

Drivas, P. J. (1978), "Comparison of  Ambient NMHC/NOX  Ratios Calculated
      From Emission Inventories," EPA-450/3r78-026,  U.S.  Environmental
      Protection Agency,  Research Triangle Park,  North  Carolina.

Eaton, W.  C., M. L. Saegar,  Ui  D. Bach,  Jj. E. Sickles,  II  and  C.  E  Decker
      (1979), "Study of the Nature of Ozone,  Oxides  of  Nitrogen,  and
      Nonmethane Hydrocarbons  in Tulsa,  Oklahoma  - Volume  III:   Data Analysis
      and Interpretation," EPA-450/4,-79-008c, U.S. Environmental  Protection
      Agency, Research Triangle Park,  North  Carolina.

EPA (1977), "Uses, Limitations  and Technical  Basis of  Procedures  for
      Quantifying Relationships Between  Photochemical Oxidants  and
      Precursors," EPA-450/2-77-021a, U.S. Environmental Protection Agency,
      Research  Triangle Park,  North Carolina.
                                      55

-------
  EPA  (1978),  "Ozone Isopleth Plotting Package (OZIPP)," EPA-600/8-78-014b,
      U.S.  Environmental  Protect ion. Agency,  Research  Triangle Park,  North
      Carolina.

  EPA  (1979),  "Procedures  for the  Preparation of Emission Inventories for
      Volatile Organic Compounds, Volume  II:   Emission  Inventory Requirements
      for Photochemical Air  Quality Models,"  EPA-450/4-79-018,  U.S.
      Environmental Protection Agency, Research Jri angle Park,  North Carolina.

  EPA  (l979b)y "Guideline  for the  Interpretation  of Ozone Air Quality
      Standards," EPA-450/4-79-003, U.S.  Environmental  Protection Agency
      Research Triangle Park, North Carolina.  ,

  EPA  (1980),  "Guidance for Collection  of  Ambient NonMethane  Organic  Compound
      (NMOC) Data for Use in  1982 Ozone SIP Development,  and  Network  Design and
      Siting Criteria for the NMOC and NO  Monitors," EPA-450/4-80-011, U. S.
      Environmental Protection Agency, Research  Triangle  Park,  North  Carolina.

  EPA  (1981),  "Addendum 1  to  the User's Manual  for  the Kinetics  Model  and
      Ozone Isopleth Plotting Package (OZIPP)," U.S.  Environmental Protection
      Agency,  Research Triangle Park,  North Carolina.

  EPA  (1982), "PTPLU - A Single Source Gaussian Dispersion Algorithm,"
      EPA-600/8-82-014,  U.S.  Environmental Protection Agency, Research Trianqle
      Park,  North Carolina.

  EPA  (1985), "A Cryogenic Preconcentration Direct  FID (PDFID) Method  for
      Measurement of NMOC in  Ambient Air," EPA-600/4-85-063,  U.S. Environmental
      Protection  Agency,  Research  Triangle Park,  North Carolina.

.EPA (1986), "On  Site Meteorological Program Guidance for Regulatory
      Modeling Applications," EPA-450/4-87-013, U.S.  Environmental  Protection
      Agency,  Research Triangle  Park,  North Carolina.

 EPA (1988), "A  PC Based  System for Generating EKMA Input Files",
   .  EPA-450/4-88-016, U. S. Environmental  Protection Agency, Research
      Triangle Park,  North Carolina.

 EPA (1989), "User's Manual for  OZIPM-4 -  Ozone Isopleth Plotting
      With Optional  Mechanisms,".   EPA-450/4-89-009a,  U.  S. Environmental
      Protection  Agency,  Research  Triangle Park,  North  Carolina.

 Federal  Register.  March  19,  1986.

 Gery, M. W.,  J.  P.  Killus, and  G.  Z.  Whitten (1988),  EPA Report,
      "Development and Testing of  the  Carbon  Bond IV Mechanism for Urban and
      Regional Modeling,"  to  be published  (1988).
                                      56

-------
 Gipson,  6.  L.,  W.  P.  Freas,  R.  K.  Kelly and E. L. Meyer (1981), "Guideline
     for Use  of City-Specific EKMA in  Preparing Ozone SIPs," EPA-450/4-80-027,
     U.S. Environmental  Protection Agency,  Research Triangle Park,  North
     Carolina.

 Gipson,  G.  L.  (1984),  "Guideline for Using the Carbon Bond Mechanism in
     City-Specific  EKMA,"  EPA-450/4-84-005,  U.S.  Environmental  Protection
     Agency,  Research  Triangle  Park ,^ North  Carolina.

 Godowitch,  J. M.','j.  K.  Ching and  J.  F.  Clarke (1979), "Dissipation
     of  the Nocturnal  Inversion Layer  at  an  Urban and Rural  Site  in St.
     Louis,"  Fourth Symposium on Turbulence,-  Diffusion,  and  Air Pollution,
     Reno, Nevada.                      •      ,                ,

 Hewson,  E.  W. (1976),  "Meteorological  Measurements,"  Air Pollution.
     Volume I.  3rd edition, A.  C.  Stern,  ed.,  Academic Press, pp.~591-597.

 Holzworth,  G. C (1972),  "Mixing Heights,  Wind Speeds, and Potential for
     Urban Air  Pollution Throughout the Contiguous  United  States,"  AP-101,
     U.S. Environmental Protection Agency, Research Triangle Park,  North
     Carolina.

 Jeffries, H,  E., K. G. Sexton and  C. N. Salmt (1981), "Effects  of Chemistry
     and Meteorology on Ozone Control  Calculations  Using  Simple Trajectory
     Models and the EKMA Procedure," EPA-450/4-81-034,  U.S.  Environmental
     Protection Agency, Research Triangle Park, North Carolina.

 Jeffries, H.  E., K. G. Sexton and  J. R. Arnold (1987),  "Analysis of
     Hydrocarbon Composition from Ground-Level  and  Aloft Measurements for the .
     Carbon Bond and Carter, Atkinson  and Lurman  Photochemical Mechanisms,"
     Cooperative Agreement CR-813107,  U. S. Environmental  Protection Agency-,
     Research Triangle Park, North Carolina.

 Killus,  J. P. and G. Z. Whitten (1982), "A New Carbon Bond Mechanism for
     Air Quality Simulation Modeling," EPA-600/3-82-041, U.S. Environmental
     Protection Agency, Research Triangle Park, North  Carolina,

Killus,  J. P.' and G. Z. Whitten (1983), "Technical  Discussions  Relating to
     the Use of the Carbon Bond Mechanism in OZIPM/EKMA,"  EPA-450/4-84-009,
     U.S. Environmental Protection Agency, Research Triangle Park,  North
     Carolina.

Ludwig,  F. L. (1979),  "Assessment of Vertical  Distributions  of
     Photochemical  Pollutants and Meteorological Variables in .the Vicinity of
     Urban Areas,"  EPA-450/4-79-017, U.S.  Environmental Protection  Agency,
     Research Triangle Park,  North Carolina.

Lurmann, F.  W.,  W.  P. L.  Cartre and L. A. Coyner  (1987),  "A  surrogate
     Species Chemical  Reaction  Mechanism for Urban-Scale Air Quality
     Simulation  Models, (Vols.  I and II)."
                                      57

-------
 Meyer,  E.  L.  and K. A. Baugues (1989), "Consideration of Transported
      Ozone and Precursors and Their Use in EKMA," EPA-450/4-89-010, U.S.
      Environmental  Protection Agency,  Research Triangle Park, North Carolina.

 Rhoads,  R. G.  (1981),  memorandum to Director, Air and Hazardous Materials
      Division,  Regions I-X,  "Effects of Chemijstry and Meteorology on Ozone
      Control Calculations Using Simple Trajectory Models and the EKMA
      Procedure,"  December 3,  1981.

 Rhoads,^R. G.  (f985),  memorandum to Director, Air Management Division,
      Regions I-X,  "Partictpation  in the Summary 1986 NMOC Sampling Program,"
      December  23,  1985.

 Rhoads,  R. G.  (1987),  memorandum to Darryl Tyler, "Proposed Draft
      Federal Register  Notice  on  Use of EKMA in Post-1987 Ozone SIP's,  April
      30, 1987.                                          .

 Rhodes,  R. C.  and E. G.  Evans (1986),  "Precision and Accuracy Assessments
      for State  and  Local  Air  Monitoring Network,  1983,"  EPA-600/4-86-012,  U.S.
      Environmental  Protection Agency,  Research Triangle  Park,  'North Carolina.

 Schere,  Ki L.  and K. L.  Demerjian (1977),  "A Photochemical  Box Model for
      Urban Air  Quality Simulation,"  Proceedings.  4th  Joint  Conference  on
      Sensing of Environmental  Pollutants.  American Chemical  Society.

 Seila, R.  L. (1986), "GC  - Personal  Computer System  for  Determination  of
     Ambient Air  C-2 to C-12  Hydrocarbon Species," Presented  at  1986 EPA/APCA
     Symposium on Measurement of  Toxic  Air Pollutants, Raleigh,  North
     Carolina.

 Singh, H.  (1980),  "Guidance for the  Collection and Use of ambient
     Hydrocarbon Species Data  in  Development  of Ozone Control  Strategies,"
     EPA-450/4-80-008,  U.S. Environmental  Protection Agency,  Research  Triangle
     Park,  North Carolina.

 Slade, D.  H. (1968), "Meteorology and Atomic  Energy," NTIS  No. TID-24190,
     NTIS U.S.  Department of  Commerce,  Springfield, Virginia 22161,  ppg.33-39.

 Wanta, R.  C. and W. P.  Lowry,  "The Meteorological Setting for  Dispersal
  .   of Air Pollutants," Air  Pollution. .Volume I. 3rd Editiion, A. C.  Stern,
     ed., Academic Press, pp.  337-352.

 Whitten,  G. Z.  and M.  W. Gery  (1986), "Development of CBM-X Mechanisms for
     Urban  and  Regional AQSM's, "EPA-600/3-86-012, U.S. Environmental
     Protection Agency, Research Triangle .Park, North Carolina.

Whitten,  G. Z.  and H. Hogo (1977), "Mathematical Modeling of Simulated
     Photochemical Smog," EPA-600/3-77-011, U.S. Environmental Protection
     Agency,  Research Triangle Park, North  Carolina.
                                      58

-------
Whitten, G. Z. and H. Hogo  (1978),  "User's  Manual  for Kinetics Model  and
     Ozone Isopleth Plotting Package," EPA-600/8-78-014a,  U.S.  Environmental
     Protection Agency, Research Triangle Park, North -Carolina.

Whitten, 6. Z., H. Hogo, M. J. Meldgin, J.  P.  Killus  and
     P. J.  Bekowies (1979), "Modeljng of Simulated Photochemical Smog With
     Kinetic Mechanisms," EPA-600/3-79-011a, U.S. Environmental  Protection
     Agency, Research Triangle Park, North Carolina.

Whitten, G. Z-., H. Hogo and J, P. Killus (1980), "The  Carbon  Bond
     Mechanism:-  A Condensed Kinetic Mechanism for Photochemical Smog,"
     Environmental Science and Technology.  Volume 14, No. 6.
                                      59

-------
       APPENDIX A





LISTING OF CB-4 MECHANISM
          A-l

-------
TABLE A-l.  CB-4 MECHANISM


1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
mmm


N02
0 --
03 -f

Reaction
--> NO + 0
•>o3
• NO --> N02
0 + N02 — > NO
0 •*
0 -t
N02
°3-
03-
01D
01D
°3 +
o3 +
N03
N03
N03
N03
N205
N205
NO +
NO +

• N02 --> N03
• NO — > N02
+ 03 -> N03
-> 0 '
-> 01D
--> 0 .
+ H20 — > 20H
OH --> H02
H02 --> OH
--> 0.89 N02 +'0.89 0 + 0.11 NO
+ NO --> 2 N02
+ N02 --> N02 + NO
+ N02 --> N205
+ H20 --> 2 HN03
--> N03 + N02
NO --> 2 N02
N02 + H20 -> 2 HN02
A- 2
Rate Constant
at 298°K
Cppm"1 min"1)
1.0
4.323 x 10s
26.64
1.375 x 104
2309
2438
0.04731
0.053
1.0
4.246 x 105
3.26
100
3
33.9
4.416 x 104
0.5901
' " 1853
1.9 x 10'6
2.776
1.539 x 10"4
1.6 x 10"11

Activation
Energy
(10
0
- 1175
1370
0
- 687
-. 602
2450
• o "
0
•- 390 .
0
940
580
0
•- 250
1230
- 256
0
1.09 x 104
- 530
0


-------
                      TABLE  A-l.   CB-4 MECHANISM (CONTINUED)
               „   ,. .
               Reaction
 22.  NO + OH -> HN02
 23.  HN02 --> NO + OH
 24. "- OH + HN02 --> N02
 25.  HN02 + HN02 --> NO + N02
 26:  N02 +OH ->HN03
 27.  OH + HN03 -> N03
 28.  H02 + NO -> OH +
 29.  H02 + N02 --> PNA
 30.  PNA --> H02  -f N02
 31.  OH + PNA --> N02
 32.  H02 + H02  ->  H202
 33.  H02 + H02 +  H20  -
 34.   H202  -> 2 OH
 35.   OH +  H202 --> H02
 36.   OH +  CO  -->  H02
 37.   FORM  + OH --> H02 + CO
 38.   FORM  --> 2 H02'+ CO
 39.   FORM  --> CO
 40.  FORM + 0 --> OH + H02
 41.  FORM + N03 --> HN03 +
42.  ALD2 + 0 --> C203 + OH
43.  ALD2 + OH -> C203


"
) + N02


N02




> «202



h CO
)

i02 + CO
I + H20 + CO
• OH

Rate Constant
at 298°K
(pom min"}
9799
.1975
9770
1.5 x 10"5
1.682 x 104
217.9
1..227 x 104
2025
5.115
6833
4144
.2181
.189
2520
322
1.5 x 104
1.0
1.0 •
237
0.93
636
2.4 x 104
Activation
Energy
f'n
-. 806
0
0
0
- 713
- 1000
- 240
- 749
1.012 x 104
- 380
-. .1150
- 5800
0
187
0
0
0
0
1550
0
986
- 250
                                     A-3

-------
                     TABLE A-l.   CB-4 MECHANISM (CONTINUED)
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
Reaction
ALD2 + N03 — > C203 + HN03
ALD2 --> FORM + 2H02 + CO + X02
C203 + NO --> FORM + N02 + H02 + X02
C203 + N02 — > PAN
PAN --> C203 + N02
C203 + C203 --> 2 FORM + 2X02 + 2H02
C203 + H20 --> 0.79 FORM + 0.79 X0? +
0.79 H02 + 0.79 OH
OH --> FORM + X02 + H02
PAR + OH --> 0.87 X02 + 0.13 OX?N +
0.11 HOp + 0.11 ALD2 -
0.11 PAR + 0.76 ROR
ROR --> 0^96 X02 + 1.1 ALD2 + 0.94 H0? -
2.1 PAR + 0.04 X02N + 0.02 ROR
ROR --> H02
ROR + N02 -->
0 + OLE --> 0.63 ALD2 + 0.38 H0? +
0.28 X02 + 0.3 CO +
0.2 FORM + 0.02 XO?N +
0.22 PAR + 0.2 OH
OH -f OLE --> FORM + ALD2 - PAR + X0?
+ H02
03 + OLE --> 0.5 ALD2 + 0.74 FORM +
Rate Constant
at 298°K
7ppm min )
3^7
1.0
. 1.831 x 104
1.223 x 104
.0222
.3700
9600
21
1203
1.371 x 105
9.544 x 104
2.2 x 104
5920
4.2 x 104
.018
Activation
Energy
f°K)
0
• -0
- 250
- 5500
1.4 x 104
0
0
1710
0
8000
0
0
' 324
• - 504
2105
                  0.22 X02 + 0.1 OH +
                  0.33 CO + 0.44 H02 - PAR

59.  N03 + OLE --> 0.91 X0? + FORM +
                   0.09 XOoN + ALD2 +
                   N02 - PAR
11.35
                                     A-4

-------
                      TABLE A-1.   CB-4  MECHANISM (CONTINUED)
       ,*

                                                    Rate Constant    Activation
               04..''                      at ?98°K i       Energy
               Reaction                             fopm"1 min"1)       f°K)

 60.  0-+ ETH --> FORM + 1.7 HO? + CO  +             1080                    79?
                  0.7 X02 + 0.30 H                        '

 61.  OH + ETH --> X0? + 1.56 FORM +                1.192 x 104         -   411
                   0.22 ALD2 + H02

 62.  03 + ETH --> FORM + 0.42 CO + 0.13 H02        2.702 x 10'3           2633

 63.  TOL + OH --> 0.44 HOo + 0.8 X0? +             9150                -   322
                   0.36 CRES + 0.56 T02

 64.  T02 + NO --> 0.9 NOo .+ 0.9 HOo +              1  2  x 104                  0
                   0.9 OPEN  '   .  •*

 65.  T02 --> CRES + H02                            250                        0

 66.  OH + CRES  --> 0.4 CRO + 0.6 X0? +             6.1  x 104                  0
                    0.6 H02 + 0.3 OPEN

 67.  CRES +  N03  --> CRO + HN03                      3.25  x  104                 0

 68.  -CRO + N02  -->    '          -                   2 x.104                    0

 69.   OPEN --> C203  + H02  + CO                      8.4                        0

 70.   OPEN +  OH  ->  X02  +  2 CO +  2  HOo  H-            4.4 x 104    '              0
                    C203 +  FORM

 71.   OPEN +  03 -->  0.3 ALD2 + 0.62 C?0,  +          0.015                   500
                    0.7  FORM + 0.3  XO^  +
                    0.69 CO + 0.8 OH  +
                    0.76 H02 + 0.2  MGLY

 72.   OH + XYL --> 0.7 H09 + 0.5 XOo +  0.2 CRES     3.62 x 104           -  116
                +• 0.8 MGCY + 1.1 PAR +•
                  0.2 T02

73.   OH + MGLY --> X02 + C203  '                    2.6 x 104                  0

74.  MGLY --> C203 + H02 +. CO                      8.96                       0
                                     A-5

-------
                     TABLE A-l.   CB-4  MECHANISM (CONTINUED)
              Reaction
75.  0 + ISOP --> 0.6 H0? +  0.8 ALD2
                + 0.55 OCE +.0.5  X0?
                + 0.5 CO + 0.45 ETH
                + 0.9 PAR
76.  OH + ISOP --> X0? + FORM + 0.67 H0?   .
                 + 0.13 XO?N + ETH +
                   0.4 MGLY  + 0.2 C?0, +
                   0.2 ALD2
77. ' 03 + ISOP --> FORM + 0.4 ALD2 +
                   0,55 ETH  + 0.2 MGLY +
                   0.1 PAR + 0.06 CO +
                   0.44 H02  + 0.1 OH
78.  N03 + ISOP --> X02 N
79.  X02 + NO --> N02
80.  X02N + NO -->
81.  X0
           X02 --
                                                   Rate Constant    Activation
                                                     at 298°K ,        Energy
                                                   (ppm"1 min"*)
                                                   2.7 -x 104
                                                   1.42 x 105
                                                  ,018
82.  NR — > NR
470
1.2 x 104
1000
2000
1
                                                                       ( 10
     0
     0
     0
- 1300
     0
                                     A-6

-------
                  APPENDIX B
ESTIMATION OF MIXING HEIGHTS FOR USE IN 02IPM4
                     B-l

-------
      In OZIPM4, the rate of dilution of atmospheric pollutants  is governed by
 the diurnal change in mixing height.  The mixing height is the  top of a
 surface-based layer of air which is well-mixed due to mechanical and thermal
 turbulence-.  As described in Section 3.2.2, the input variables required for
 OZIPM4 include:  the mixing height at 0800 LOT, the maximum mixing height, the
 time at which the mixing height begins to rise if it starts to rise after 0800
 LCT, and the time at which the mixing height reaches its maximum.   The rate of
 rise is computed internally by OZIPM4.
      Three different procedures exist for determining daily morning and
 afternoon mixing heights;   The recommended procedure entails the use of
 temperature soundings  taken routinely by the National  Weather Service at
 various locations  throughout the United  States.  If more direct measurements
 are  available (e.g.,  radiosondes taken  in  the urban area or sodar  data),  they
 may  be  used instead  of NWS  data.   If neither of the above  two sets  of
 measurements  can be  used, then  the  use of  250 m for the  0800  LCT mixing  height
 and  the climatological  mean  value for the  maximum mixing height is
 recommended.  The  procedures  to be  followed  for each  approach are described
 below:

 B.I  RECOMMENDED PROCEDURE USING  NWS  RADIOSONDES
     Temperature soundings are  taken  by the  NWS at  sites throughout the
 United  States. ,  Soundings are usually taken  every 12 hours  at  1200 and 0000
 Greenwich Mean Time (GMT), corresponding to  0800 and 2000 Eastern Daylight
Time, respectively.  Therefore, to estimate daily mixing heights (1), a NWS
site must be selected which is representative of the city of  interest,
                                      B-2

                                        /

-------
  (2) appropriate sounding data and urban surface data must be obtained, and (3)
  these data must be used to compute the morning and
  maximum mixing heights.  Each of these steps is discussed below.
       B.I.I   Site Selertinn
  In  selecting  a NWS site as the  basis  for  mixing height  estimation,
  care  should be taken to ensure  that the site is meteorologically
  representative of  the city of interest. .Table  B-l contains recommended
  sites for a number of cities.   Backup sites  are  listed for those cases in
 which radiosonde data may not be available for  a given day, or if the
 site has significantly different meteorological conditions.  Examples of
 the latter are the case in which a surface front lies between the sounding
 site and the city or the city .is clear but cloudiness or, precipitation
 occurs at the  sounding  site.
      B.I.2  Selection of Day.Specific  Data
      The daily morning mixing height for the  model  is normally  estimated
 using  the  1200 GMT  (0800 EOT)  sounding, while the maximum  mixing height
 is estimated using  the 0000  GMT  (2000  EOT) sounding.   In some cases,
 these  soundings  may not  be  available or appropriate and alternate approaches
will be necessary.  Table B-2 summarizes the  order of preference'in
selecting the radiosondes for estimating the daily mixing heights.   The
actual data may be obtained from the National  Climatic Center (NCC).* '
     Please allow about 4 weeks'for NCC to fill  an order.
                                     B-3

-------
                               TABLE B-l.  NWS  RADIOSONDE  STATIONS
   CitV
Allentpwn, PA
Baltimore, MD
Boston, MA
Bridgeport, CT
Chicago, IL/IN
Cincinnati, OH/KY
Cleveland, OH
Dayton, OH
Denver, CO
Detroit, MI
:resno, CA
iartford,  CT
jouston, tX
Indianapolis,  IN
.os Angeles,  CA
.ouisville,  KY/IN
lilwaukee,  WI
iahsville,  TN
lew Haven,  CT
 ew York,  NY/NJ
shladelphia, PA/NJ
 hoenix, AZ
 ittsburg,  PA
 ortland,  OR
 rovidence, RO

 ichmond,  VA

 acramento, CA
t.  Louis, MO/IL
alt Lake City, UT
an Bernardino, CA
 an Diego, CA
an Francisco, CA
cranton, PA

aattle, WA
Dringfield, MA
"enton, NJ
         Primary
  NYC, NY; At! City, NJ
 " Dulles AP, VA
  Portland, ME
  NYC, NY; At! City, NJ
  Peoria, IL
  Dayton, OH
  Dayton, OH
 *Dayton, OH
 *Denver, CO
  Flint,  Mi-
  Oakland, CA
  Albany, NY
  Victoria,  TX
  Dayton, OH
  Vandenberg AFB,  CA
  Dayton, OH
  Green Bay,  WI
 *Nashville,  TN
  NYC,  NY;  Atl Cityj'NJ
 *NYC,  NY;  Atl City,  NJ
  NYC,  NY;  Atl City,  NJ
  Tucson,  AZ
 * Pittsburgh,  PA -
  Salem,  OR
  New York, NY

  Dulles  AP,  VA

  Oakland, CA
  Salem,  IL
*Salt  Lake City, NT
 San Diego, CA
*San Diego, CA
 Oakland, CA
 NYC, NY; Atl City, NJ

 Quilayute, WA
 Albany,  NJ
 NYC, NY; Atl City, NJ
                                                        Backups)
                           Climatological
                           Mixing Heights
                              (m AGL)

                                MAX
 Albany, NY; Dull as AP,,  VA
 Wallops Is., VA; Atl City,
 Albany, NY; Chatham, MA
 Albany, NY
 Green Bay, WI
 Huntington, WV
 Buffalo, NY
 Huntington, WV
 Grand Junction,  CO
 Dayton, OH
 Vandenberg AFB,  CA
 NYC,  NY; Atl  City, NJ
 Lake  Charles,  LA
 Peoria, IL; Salem, IL  .
 San Diego, CA
 Nashville, TN
 Peoria, IL
 Jackson, AL
 Albany,  NY
 Albany,  NY
 Dulles  AP,  VA
 Winslow, AZ
 Dayton,  OH; Dulles AP,  A
 Medford, OR;  Quilayute,  WA
 Albany,  NY; Chatham,  MA;
      Atl City,  NJ
 Greensboro, NC;
      Wallops  Is.,  VA
 Vandenberg AFB,  CA
 Peoria,  IL; Monette,  MO
 Grand Junction,  CO
 Vandenberg AFB,  CA .
 Vandenberg AFB,  CA
 Vandenberg AFB,  CA
Albany,  NY; Atl  City, NJ
     Dulles AP,  VA
Salem, OR
NYC, NY; Atl City, NJ
Dulles AP,  VA       |
    1825
NJ  1825
    1375
    1500
   1575
    1650
   1650
   1661
   3358
   1700
   2000
   1500
   1525
   1600
    603
   1700
   1575
   1845
   1450
   1512
   1700
   3250
   1794
   1575
   1350

   1725

   1600
   1625
  3673
  1200
   564-
   625
  1850

   1398
  1600
  1700
                                              B-4

-------
                           TABLE B-l (CONTINUED)

                                                    Climatological
                                                    Mixing Heights
                                                      (m A6L)

	^	   	Erimary	       Backup^               MAy


«KT^3xsrwp>«     »?SC\A   ..'      ;««
il.ington.-DE      Dulles A^VA;        Kallops Is.'; VA; -   ,       1884
               At I City, NJ         New York  NY              ,-,**'
)rcester, MA       Albany, NY           Portland;  JJE;             17°°

>ungstown, OH      Pittsburg, PA         Bu^fa™' K;              15°°
                                 . Dayton, OH          • '    1700
sr s a«s
TE:  The NYC, NY radiosonde station was replaced by AtUntic City, NJ on September 2, 1980,
                               B-5

-------
             TABLE  B-2.   PREFERENTIAL  ORDER OF DATA SELECTION
 Horning  Mixing  Height Estimate
 1.   1200 GMT Sounding at Primary Site
 2.   0600 GMT Sounding at Primary Site*    -    ,
 3.   1200 GMT Sounding at Backup  Site
 4.   0600 GMT Sounding at Backup  Site*
Maximum Mixing Height  Estimate
1.  0000 GMT Sounding  at Primary  Site
2.  1800 GMT Sounding  at Primary  Site*
3.  1200 GMT Sounding  at Primary  Site
4.  0000 GMT Sounding  at Backup Site
5.  1800 GMT Sounding  at Backup. Site*
6.  1200 GMT Sounding  at Backup Site
*Soundings are not normally taken at these times,  but  may be available in
 some instances.
                                   B-6

-------
      In addition to the sounding data, surface temperature  and pressure data
 are also needed for each day modeled.  The urban surface temperature at 0800
 LCT (or the average temperature between 0800-0900 LOT) and  the maximum
 temperature occurring prior to 1800 LCT are needed to estimate the morning and
 maximum-mixing height, respectively..  The surface temperature data should be,
 measured to the nearest 0.1° C-at a well  ventilated site (EPA, 1986).  The
 site should be located near the center of the urban area.  Surface atmospheric
 pressure measurements are needed at the same time and location of the urban
 surface temperature measurements,  if at all  possible.  If these measurement?
 are not available,  a local  NWS or-Federal  Aviation  Administration weather
 reporting  station's barometer reading may be used.
      If the elevation of the pressure reading and the urban  temperature site
 are different,  an adjustment should be made  to the  pressure  measurement using
 equation  (3)                                  '       .
                Psfc = pobs'+ [.ilmb/m x (Zobs - Zsfc)]                (3)
     where
             '   zobs  =  the elevation,  in meters  above  sea  level  (mASL),
                       of the pressure  measurement
                zsfc  =  tne elevation  (mASL) of the urban temperature
                       measurement
                pobs  =  the Pressure,  in  millibars, at  ZQbs
                Psfc  =  the pressure,  in  millibars, at  the urban
                       temperature site
NOTE:     Zobs w111  be eclual to-zero meters ASL when a pressure reduced at
          sea level  is used.
                                     B-7

-------
 The value of Pgfc from equation  (3)  is  an  approximate  value  and  can  be  rounded
 to the nearest whole millibar.
      B.I.3  Mixing Height Estimation
      The procedures for estimating the  0800  LCT mixing height  and the maximum
 mixing height are outlined in Table Br3.   The procedures  in  Table B-3 are
 designed for use with the worksheet displayed in Table B-4.  Figure B-l
 contains a flow diagram of the process.  The procedures use  the mandatory and
 significant pressure levels reported for each sounding (Table  B-5).  The steps
 lead to determination of the height at which the.adiabatic lapse rate
 (extended from the surface temperature and pressure) intersects the vertical
 temperature profile). -(For background information, the reader is referred to
 Wanta and Lowry,  1976; Hewson,•1976;  and Slade,  1968).  An example problem is
 presented in Section B.4.  •
      In some instances,  the mixing heights estimated by this procedure may
 not be representative.  If the 0800 LCT morning mixing height  is estimated to
 be less than 250  meters,  then a value of 250 meters should be used.   This
 assumed minimum value for  the 0800 LCT mixing height accounts for the effects
 of mixing due to  mechanical  turbulence caused by increased surface roughness
 in the urban area (Godowitch,  et al,  1979;  and Bentley and Schulman,  1979).
 Similarly,  if the city's maximum mixing height is  greater than twice  the
 climatological  maximum value  (e.g.,  see Table B-l), the surface temperature
 and pressure used and the  choice of sounding site  should  be checked  for
 representativeness  using the  guidelines in  B.I.I and B.I.2 above.   If no
 backup  data  are available, twice the  climatological value should be  used as
•the maximum.  Also,  a maximum  mixing  height less than or  equal  to the morning
                                      B-8

-------
              TABLE B-3.  PROCEDURES FOR ESTIMATING MIXING HEIGHTS


      Step 1 -- For reference,  the information at the top of Table B-4 should
 rfri,1,? J Ve;i!''n2Se,V?1ty'*etc'}-   If the morn™9 mixing height is to be
 calculated  the 0800 LCT surface data are used.   If the maximum mixing height
 is  to be calculated,  the data  corresponding to the time of maximum temperature
 (i.e.,  between 800-1800 LCT.) are used.   In the row Tabled URBAN SURFACE DATA
 enter the following information:  (1) the elevation of the urban temperature'
•site  in meters above  sea level;  (2)  the surface  pressure in millibars (this
 value is Psfc);  and (3)  the surface  temperature  in degrees Celsius (°C).

      Convert  the surface temperature in column four to degrees  Kelvin (°K)  bv
 adding  273.2,  and  enter  the result  in column  five.   This value  is T$fc(  K).

      Use Equation  1 below and  the values  just entered  to calculate the
 potential  temperature at  the surface.(Of  in Y, to the  nearest  O.l'K)  and
 enter this  value under column  six "(  K)?*          .


                        .    '   -0.286
                sfc  (in  K) = Tsfc (in  °K)  £sfc...(1n mb)                 (1)
                                            ~™0  mb

i   iSt!E 2 Iu Using the  temperature sounding data,  find the highest  pressure
level  other than the founding's surface value that  is less than the. pressure
at the urban surface.     From this pressure level on the sounding, enter the
5?J?«   (1£ i,lstndi' Pressure>  and temperature (in °C) into the row marked
 (Z)  on Table B-4.
     **
       For example,  if  the  urban  surface  pressure  is  985  mb,  and  the  soundinq
     pressures  are:   1005,  1000,  963,  850 mb,  etc., 963 mb  is the "highest
     pressure level  that  is  less  than  the pressure at the urban surface." 850
     mb  is the  "next lowest  pressure level"  needed in Step  4.

                                     B-9

-------
                             TABLE B-3 (CONTINUED)

     Step 3  -- Convert the temperature at  this  level  to  the  Kelvin  scale  and
 enter  in column  5.   Compute  the potential  temperature (D)  to the nearest  0.1°K
 using  the pressure  (P,  in mb)  and temperature (TD in  °K|) at  this level  in
 Equation 2 below:                                v
                                            -0.286
           „ (in  K) = T_ (in  °K)   P Mn  mb)            (2)
           v     ,       H            1000 mb

 Enter  the value  of  n found from Equation (2)  into the same row  under  the
 column labeled ir(K)7"

     Step 4  -- If the potential  temperature of  the last  row  that
 was entered  is greater than  the potential  temperature sfc, and  this is  the
 first  level  above the surface,  then 250  meters  should 5e used as  the  mixing
 height (if given),  pressure  and temperature of  the next  lowest  pressure level
 found  on the sounding into the  next row  of Table B-4  and return to  Step 3.

     Step 5  -- The mixing height  is between the  last  two levels entered into
 Table  B-4.   If height values are given for both  of these levels,  the  elevation
 of the mixing height can be  found  using  Step  6.   If one  of the  levels does not
 have a height value, use linear interpolation to find the  pressure  value  for
 the potential temperature value sfc + 0.1°K.  Enter this pressure into  the row
 marked "MIXING HEIGHT" at the bottom of  Table B-4 under  the  column  "PRESSURE
 in mb."  Proceed to Step 7.

     Step 6 — From the two levels where height  is given on  the sounding  •
 surrounding  the mixing height level, use linear  interpolation to find the
 height On meters ASL) at the value _fc  +  0.1°K  (i.e., the potential
 temperature  at the mixing height).  Enter  the value found  by linear
 interpolation into the row labeled "MIXING HEIGHT" under the column "HEIGHT
 (mASL)" and proceed to'Step 8.,

     Step 7 -- Use linear interpolation to find the height above sea level of
 the mixing height using the pressure at  the mixing height  (found  in Step  5)
 and the pressure .levels on the  sounding  above and below  the  mixing  height
 pressure that have both pressure and height values.   Enter the  height value
 found  into the row "MIXING HEIGHT" under the column marked "HEIGHT, (mASL)"
 and proceed to Step 8.

     Step 8 -- Subtract  the  elevation of the urban site (mASL) from the height
 (mASL) of the mixing height,   the result is the  height of the mixing height in
meters above the surface of the city (mAGL).  Enter this value  into Table B-4.

NOTE:        Despite the fact that pressure and height, and potential
            temperature and height, are not linearly  related, linear
            interpolation does not produce significant errors over the limited
            ranges used above.
                                     B-10

-------
Date:
              TABLE B-4.  WORKSHEET FOR COMPUTING MIXING HEIGHTS
                     Time of Mixing Height for Input Into Model:
                                  Sounding Method:
Time of Sounding:                 LOT, Surface Elevation:
Location of Sounding:  ,         '   '     •                   -
LOCATION OF URBAN SURFACE AREA (IF DIFFERENT THAN ABOVE)
     1     '       .  2           3          *          56
                                                                       mASL
LEVEL ' r .
Urban (
Surface r
Data ' (1)
(2)






HEIGHT
(mASL)
i

•





PRESSURE
' (mb)
i i .







TEMP.
Co








TEMP.
(*K)




•'

•• .

(°K)








REMARKS
<"sfc








MIXING
HEIGHT

(Lfr + 0.1 K
C(°K)



PRESSURE



HEIGHT
(mASL)



Height
(mAGL)



HEIGHT USED IN
MODEL (mAGL)



                                    B-ll

-------
                            - ENTER URBAN SURFACE DATA
                            INTO TABLE A-«f,
                            - CONVERT SURFACE TEMPERATURE
                            TO •*,
                            -FIND 6 ,  USING EQUATION  (1).
                       (5)- on TABLEAU LIST DATA TROM
                       W SOUNDING FROM THE FIRST PRESSURE
                        LEVEL ABOVE THE URBAN  SURFACE LEVEL.
                           ; - CONVERT  TEMPERATURE TO *K AND
                            USE EQUATION .(2)  TO COMPUTE PO-
                            TENTIAL  TEMPERATURE (6p) FOR
                            THE LEVEL  JUST ENTERED.
                                                         - ENTER NEXT PRESSURE
                                                           LEVEL INTO TABLE A -M-.
                                  - fll I.ISG HV.I(.HT IS
                                  AT e
                                      sfe
       - ARE THE
  "HEIGHT VALUES GIVEN
KIR THE LAST TWO ROWS
      ENTERED INTO
       JTABLE A-4"

               NO
                                                           TES
                             - USE LTREAK INTERPOLATION
                             TO FIND THE PRESSURE AT
                               - FIND THE HEIGHT (aASL) OF
                               OF THE MTITHG HEIGHT BY
                               LINEAR INTERPOLATION FROM
                               TEE PRESSURE AT THE MIXING
                               HEIGHT.
                                                                     D- USE LUilEAS.  INTER-.
                                                                      POLATIOH TO FIND  THE
                                                                      HEIGHT' (BIASD AT
                                                                      a
                                                                      ymixing height.
                                                V
                            fi)- FIND TEE HEIGHT- ABOVE THE
                            URBAN AREA  (mAGL)  OF THE MIS-
                            ING HEIGHT -(THIS  GIVES THE
                            ANSWER.} .                  	
Figure,B-l.  Flow Chart for TableB-3.  Numoers  in  circles are step  numbers  in  Table pi-3.

                                         B-12

-------
                      TABLE  B-5.   SURFACE  AND  SOUNDING  DATA

 Hour Starting at. LCT          Temperature °C   .       Pressure, mb







1
,
-

1200
Pressure
(mb)
S 1015
M 1000
S 967
M 850
'S 827
S 817
M 700
S 680
S 661
S 608
M 500
S 491
S 453
S 438
M 400
S 388
S 349
S 324
M 300
S 267
M 250
M 200
M 150
S 148
8
9
10
11
12
13
14 ,
15
16
17
18
GMT Sound ina
Height
(m,ASL)
1 . 8* i
139
—
1550
_'__
___
3168
—
—
—
5860
> —
—
—
7560
—
—
—
"9640
—
10890
12370
14190
—












Temp.
( °Q)
. 23.0
23.0
24.4
16.2
14.2
13.6
4.6
5.6
5.6
0.4
- 8.3
- 9.3
-12.7
-13.9
-18. 7
-20.1
-26.3
-29.7
-33.7
-39.5
-47.7
-51.7
-60.9
-61.5
       23.2
       23.9
       25.8
       27.3
       28.7
       29.3
       3Q.1 ,
       30.4
       30.8
       31.4
       31.2

Sounding Data
                                                              1010.3
                                                              1010.7
                                                              1Q10.8
                                                              1010.6
                                                              1010.3
                                                              1010.0
                                                              1009.
                                                              1009.
                                                              1008.8
                                                              1008.6
                                                              1008.5
                                                           0000 GMT Sounding
Pressure
(mb)
S 1012
M 1000
M 850
S 831
S 791
S 778
S 760
M 700
S 628
S 560
M 500
M 400
S 371 -
M 300
S 265
M 250
S 205
M 200
M 150
S 127
S 120
M 100
M 70
M 50
M 30
M 20
S 15
Height
(m ASL)
8*
114
1537
—
—
—
___
3164
• , —
—
5860
7560
—
9650
--.-
10900 "
—
12370
14190
—
—
16690
18900
21040
24350
27030
—
Temp
( °Q)
31.0
30.6
16.4
15.4
13.2
11.8
11.2
7.0
1.6
- 1.5
- 7.3
-18.9
-21.7
.-33.1
-39.9
-42.9
-52.9
-53.3
-61.1
-64.9
-61.7
-63.3
-58.5
-54.5
-49.9
-44.7
-42.1
NOTE:  M = Mandatory Levels and S = Significant Levels
            If NWS data are used,  both  the mandatory and  significant  levels
            are needed.
            The 0000 GMT sounding  is  the following  day in GMT.
           *The lowest level of the sounding should not be used in the mixing
            height calculations.
                                     B-13

-------
mixing height, or less than one-third the climatological maximum mixing height
value is suspect. Using data from a backup site may provide a more realistic
value.  However, if the low afternoon mixing height is due to the existence of
a surfacebased stable layer, an adjustment to. the procedures putlined in.Table
B-3 can be employed.  Replace the "Urban Surface Data" with the following data
from the sounding site:  (1) the maximum temperature, (2) the estimated or
observed surface pressure at the time of maximum temperature, and (3) the
height of the sounding surface level.  Then compute the mixing height
according to'the procedure in Table B-3.  If this problem occurs on a majority
of modeling days, then an alternative, more representative site should be used
for all the modeling days.

B.2  USE OF ALTERNATE DATA
     Other, more direct measurements of mixing height may be used to increase
the representativeness of the estimated values.  These methods include direct
urban temperature sounding and sodar data.  The measurements should be taken
over the urban area near the center of the city at 0800 LCT, and close to the
time of the climatological maximum surface temperature. It is not recommended
that these measurements be taken specifically for the OZIPM4/EKMA techniques;
however, they may be employed if available. Examples are discussed below.
     1.  Local Urban Radiosonde --'The methods described in Section,8-1.3 can
be used to find the mixing height from radiosondes taken within the urban area
as opposed to NWS sites,  the radiosonde surface temperature and pressure
should be used  in place of the URBAN SURFACE DATA.
                                     B-14

-------
      2-   Urban Helicopter Soundings -- Similarly, vertical temperature
 profiles obtained from helicopter soundings can be used in place of the NWS
 soundings.   The urban site surface temperature and pressure should be used as
 the URBAN SURFACE DATA.           •              .
      3.   Sodar -- (also  known as Acoustic Radar)  the mixing height found by
 sodar (in mAGL) can  be used directly in  the model.
 NOTE:  Regardless of the procedure applied,  the limitations concerning the
 morning  and  maximum  mixing heights that  were described  in  Section  B.I.3 should
 be  observed.

 B.3   USE OF  CLIMATOLOGICAL MEANS
      If  radiosonde data  are not  available,  250  m  should  be used  for  the 0800
 LCT mixing height and  the  city-specific  climatological mean  value  may  be used
 for the  maximum,mixing height.   Table  B-l  lists representative values  for
 several  cities, and  Holzworth, 1972, contains  information  for the  contiguous
 United States.  If Holzworth  is  used,  values for  summer, nonprecipitation days
 should be used.   The appropriate starting  and ending tiems of the mixing
 height rise  in  the model  are 0800 LCT  and the time of the maximum temperature.
 If the latter  is  unknown,  1400 1ST (1500 LOT) may be assumed.
B.4  EXAMPLE PROBLEM
     To  illustrate the procedure described in Section B.I.3, an example
problem  is included for reference.  Table B-5 shows relevant data typically
available.  Note that both the 1200 GMT and the 0000 GMT soundings are used in
                                     B-1.5

-------
the calculations, the former for the morning mixing height and the latter for
the maximum mixing height.  Table B-6 shows the individual computational  steps
for the morning mixing height calculation, while Table B-7 shows the same for
the maximum mixing height.
                                     B-16

-------
                 TABLE B-6.  MORNING MIXING HEIGHT DETERMINATION

Example from Table B-5:
     08 LCT temperature = 23.2°C
     Maximum temperature after 08 LCT = 3.4°C at 17 LCT
     08 LCT pressure = 1010.3 mb            -        -   -
     Pressure at time of maximum temperature (1700 LCT)
     morning mixing height = 0800 LCT
     Time of maximum mixing height = 1700 LCT
                                     1008.6 mb%T1nfe of
Problem:
     Find the 0800 LCT mixing height using data from the sounding shown in
     Table B-5 (i.e.,  the 1200 GMT sounding).  A worksheet is shown as Table
     B-6A.  The elevation of the urban surface site is 62 mASL.
Solution:

     STEP 1
           Enter 62.,  1010.3,  and 23.2 into row (1)  of Table B-6A
                    (URBAN SURFACE DATA)
           Temp (  C)   =  23.2
           Converted to °K =   23.2  + 273.2 - 296.4°K
           Enter 296.4 into row (1)  of Table B-6A under "TEMP(°K)"  Using
           Equation  (l).on the Urban  Surface Data:
                                                       -0.286
                                         1010.3  mb

                                         1000 mb
Osfc = 296.4°K
                     Osfc = 295.5'K
    STEP  2  -  Enter  139.,  1000.,  and  23.0  into  row  (2)  of  Table  B-6A
                                     B-17

-------
STEP 3 - 23.0 + 273.2 =296.2 K

            Using Equation (2):
                Op = 296.2 K
                                                  -0.286
1000  mb

1000 mb
                Op = 296.2°K (enter  this  value  into  Table  B-6A)
                                                                   i

STEP 4 - Op (29612eK) is greater than Osfc (295.5°K).       ,


            Since 0_ is from the first level  above the surface,  the 250 m
            defaulrvalue should be used for the 0800 LCT mixing height.
                                B-18

-------
                      TABLE B-6A.  EXAMPLE  (HYPOTHETICAL DATA)
Date:  Date of Modeling   Time of Mixing Height for Input Into Model:  0800 EOT
City:  City to be Modeled   Sounding Method:  NWS, Urban Radiosonde or Helicopter
Time of Sounding:  0800     LCT. Surface Elevation: (of sounding)  mASL
Location of Sounding:  Name'of Sounding. Site                  "
LOCATION OF URBAN SURFACE DATE (IF DIFFERENT THAN ABOVE) - Street Address,
                    Building or Park,  etc.

    1          2           3          4          R'        «      •      7
LEVEL ,
Urban
Surface
Data (1)
(2)



HEIGHT
(mASL)
62
139



PRESSURE
(mb)
1010.3
1000.



TEMP.
(•c)
23.2
23.0



TEMP.
(°K)
296.4
296.2



(°K)
295.5
296.2



REMARKS
<"sfc
level is
higher than
sfc +0-1 °K





MIXING
HEIGHT

°sfc,t 0.1'K
( K)
295.6


PRESSURE
(mb)



. HEIGHT
(mASL)
0


Height
(mAGL)
0


HEIGHT USED IN
MODEL (mAGL)
250


                                       B-19

-------
                TABLE  B-7.   MAXIMUM  MIXING  HEIGHT DETERMINATION
Example from Table B-5:

     08 LCT temperature = 23.2°C
     Maximum temperature = 31.4  C at 17 LCT
     08 LCT pressure = 1010.3  mb
     17 LCT pressure = 10Q8..6  mb  •              "                 ,     -
     Time of morning mixing height = 0800 LCT
     Time of maximum mixing height = 1700 LCT

Problem:

     Find the maximum afternoon  mixing height using data from the sounding
     shown in Table B-5 (i.e., the 0000 GMT sounding).  A worksheet is shown
     as Table B-7A.  The elevation of the urban site is 62 mASL.

Solution:

     STEP 1                              .

           Enter 62, 1008.6, and 31.4 into row (1) of Table B-7A (Urban
                 Surface Data)

                 Temp (°C) = 31.4°C
                 Converted to  °K - 31.4 + 273.2 - 304.6°K
                 Enter 304.6°K into row (1) of Table B-7A under "Temp ( K)"
                 Using Equation  (1)  on the Urban Surface Data
                                                       -0.286
                     sfc
                         = 304.6 K
1000  mb

1000 mb
                     sfc
                           303.9
     STEP 2
           Enter 114., 1000.,  and 30.6 into Table B-7A.
                                     B-20

-------
                   TABLE'B-7 (CONTINUED)
 STEP 3
       30.6°C + 273.2 = 303.8°K
       Using Equation (2):

           n = 303".8°K         1000.  mb
                               1000 mb
               -0.286
           p = 303.4°K
STEP  4  -  303.8°K  is  less  than 303.9°K
           Therefore, enter 1537., 850., and 16.4  in Table B-7A and
           return to STEP 3.
STEP 3  -  16.4'C =273.2 = 289.6°K
           Using Equation (2)
              289.6°K
            = 303.4°K
                                             -0.286
 850  mb
                              1000 mb
STEP 4 - 303.4°K is less than 303.9°K
           Therefore,  enter 831.  and 15.4 into Table B-7A (note
           that there  is no height value for this pressure level)
           and return  to STEP 3.
STEP 3 - 15.4°C = 273.2 = 288.6°K
           Using Equation (2):
                                            -0.286
            = 296.2°K
              304.3CK
 831  mb
1000 mb
                          B-21

-------
                  TABLE  B-7  (CONTINUED)

STEP 4 - 304.3°K is greater than 303.9°K

STEP 5              ...
           sfc
               + 0.1  K = 303.9  K = 0.1  K =  304.0  K
           Using linear interpolation  from temperature  (0)  to
           pressure since a highest value  is  not xjiven  for  the 831
           mb pressure level -            -"           .       .
                         0 ( K)

                         303.4
                         304.0
                         304.3
Pressure (mb)

   850
   P mixing height
   831
pmixing height = 831 mb - (850 mb - 831 mbU304.0°K -  304.3°K)
                                     304.3 K - 303.4 K

                  - 831  -  (19 mbU-0.3'tO
                               0.9 K

                  = 837.3  mb

      The pressure at the mixing height (rounded to the  nearest
     •whole millibar) is 837 mb.
STEP 7
           Use linear interpolation  to  find  the  height above sea
           level  of the mixing  height.   Enter  3164.  and 700. into
           Table  B-7A.

                          Pressure  (mb)Height (mASL)

                               8501537.
                               837Z mixing height
                               7003164.

      Z mixing height = 1537 m + (3164  m - 1537  mH837 mb -  850 mb)
                                            700  mb - 850  mb

                      = 1537 m  + (1627  mU-13  mb)
                               -150 mb

      Z mixing height = 1678 m
                          B-22

-------
                  TABLE B-7 (CONTINUED)


STEP 8
             1678 mASL - height of mixing height
            -	62 mASL - elevation of urban surface site
             1616 mAGL = mixing height in meters above the urban
                         area

             1616 m is the height of the maximum mixing height to
             be used in the model with the time of 1700 LCT
                         B-23

-------
                       TABLE B-7A  EXAMPLE (HYPOTHETICAL DATA)
'Date:
 City:
 Time of Sounding:
Time of Mixing Height for Input Into Model:  1700 EOT
Sounding Method:  NWS
LCT. Surface Elevation:       8.  mASL
 Location of Sounding:    Name of Sounding Site
 LOCATION OF URBAN SURFACE AREA (IF DIFFERENT THAN ABOVE)
      1               2            3           45
LEVEL
Urban
Surface
Data (1)
(2)






HEIGHT
(mASL)
62
114
1537
—
3164



PRESSURE
(mb)
1008.6
1000.
850.
831.
700



TEMP.
(°C)
31.4 .
30.6
16.4
15.4
—



TEMP.
(°K)
304.6
303.8
289.6
288.6
—



(°K)
303.9.
303.8
303.4
304.3
—

1

REMARKS,
<"-sfc
Mixing height
is between
- these two
levels at
304 0°K

Needed to
•provide upper
height value
•for interpola-
tion



MIXING
HEIGHT

O-f. + OU'K-
SfC(°K)

304.0

PRESSURE
(mb)

837.

HEIGHT
(mASL)

1678.

Height
(mAGL)

1616.

HEIGHT USED IN
MODEL (mAGL)
•
1616.

                                         B-24

-------
                                   APPENDIX C
               COMPUTATION OF CARBON BOND  FRACTIONS FROM GC DATA
      In this appendix,  the computation of carbon-fractions from results of
 gas  chromatographic (GC)  analysis is  discussed.   As  noted  earlier,  GC analysis
 actually measures  the concentrations  of individual organic species,  which must
 then be grouped  according to the CB-4  organic reactivity classes.   In order to
 keep the computations relatively simple for illustrative purposes,
 hypothetical  examples are discussed.   For  more detailed discussion,  the reader
 is referred  to EPA,  1989.
      At the  heart  of the  computational  procedure  is the definition of how
 individual species  should  be categorized according to carbon bond type.
 Definitions  for  numerous  individual species  and are listed  in Table  C-l.
 These definitions,  or species profiles, give the number of bond types  found  in
 each  CB-4 category.   Using this  information,  along with the carbon numbers
 shown in Table 2-1,  it  is possible to compute  concentrations of individual
 carbon  bond  classes,  and then determine percentages of carbon in each  class.
 These computations will be illustrated by the  examples in Tables C-2 and C-3,
 respectively.
                                         i
     Table C-2 presents example calculations for a hypothetical  example.  The
 individual species that might be detected by GC analysis are shown in the left
 hand column,  and their associated concentrations,  in  units  of ppbC and ppb,   •
 are shown in  the next two columns.  The remaining  columns  are associated with
the carbon bond computations.
                                     C-l

-------
 TABLE  C-l      SPECIES  PROFILES-BY  BOND  GROUPS FOR  CB-4

  COMPOUND HAKE                   PAR'  OLE   ETH    TOL   XYL  FORH  ALD2  I50P   NR

1,1,1-TRICHLORQETHAHE                                                            2.0
1,1,2-IRICKLQROETHftHE                                                            2,0
1,2,3,4-TETRAHETHYLBEHZENE         2.0      .              1.0
1,2,3,5-TETRAHEIHYLBEHZEHE         2.0                    1.0
1,2,3-TRIMETHYtBEHZENE             1.0                    1.0
1,2,4,5-TETRAHETHYLBEHZENE         Z.fl                    1.0
1,2,4-TRIHETHYLBEHZENE             1.0          '.         1.0
1,2-DIETHYLBEHZENE                2.0                    1.0
V-HIKETHYL-3-ETHYIBEHZEHE        2.0-            '   '  '  1.0
1,2-BIHETHYM-ETHYLBENZENE        2.0           '         1.0
1,3,5-TRIHETHYLBEHZENE             1.0       .             1.0
1,3-BOTAHEHE                           2.0
1,3-BIETHYlBENZENE                2.0             ,       1.0
1,4-BUTAHEBIOL                    4.0
1,4-BIETHYLBEHZENE                2.0                    1.0
1-BUTENE                          2.0   1.0
1-BBTYHE                          3.0                                            1.0
1-CHlflROBOTANE                    4.0
HECEHE                          8.0 '  1.0
l-ETHOXY-2-PROPAHOL                3.0                                1.0
1-HEPIEHE                  ,       5.0   1.0
1-HEXEHE                       '   4.0   1.0
1-KETHYtCYCLOHEXENE                5.0   1.0
HETHYL-2-ETHYLBEHZEHE            1.0                    1.0
l-HETHYL-3-EIHYLBENZENE            1.0                    1.0
l-KETHYL-3-ISOPROPYLBESJZEHE        2.0                    1.0
HETHYL-3-H-PROPYLBENZEHE         2.0       '             1.0
l-KETHYL-4-ISOPROPYLBENZENE        2.0          '          1.0
1-HOHEHE                          7.0   1.0
1-OCTENE                          6.0   1.0
1-PEHTEHE                         3.0   1.0
HHBECEHE                        ?.0   1.0
2,2,3-TRIHETHYLBOTANE     '         7.0
2,2,3-TRIKETHYLPE«TAHE             8.0
2,2,4-TRIIiETHYLPEHTftNE             8.0
2,2,5-TRIHETHYLHEXMlE              ?.0
2,2-BICHLOROKITROANILINE           1.0                                            5.0
2,2-BIHETHYLBOTAHE                6.0
2,2-JIHEIHYLHEXAHE                8.0
2,2-IIHETHYLPROPANE                5.0
2,3,3-TRIKETliYLPEHTAHE             8.0
2,3j3-TRI«EIHYl-l-BUTEKE           6.0                          1.0
2,3,4-TRIHET«YLPEHTftKE             8.0
2,3,5-IRIIiETHYLHEXANE              9.0
2,3-BI«ETHYLBUTAW£     '           6.0
2,3-DIHETHYLHEPTftNE                9.0
2,3-BIHETHYLHEXANE                8.0
2,3-!I«EIHYLOCTANE               lO.fl
2,3-H«ETHYLPEKIANE                7.0
2,3-BIHETHYL-l-BDTENE              5.0      '                   1.0
2,4,4-TRIHETHYL-l-PEHTEHE          7.0                          1.0
2,4,5-TRIHETHYLHEPTAHE            10.0
2,4-DIMETHYLHEPTAHE                9.0
2,4-JIKETKTLHEXANE                fi.O
                                     C-2

-------
    iwmm                    m   GLE   ETH   TDL   m  FORM   AIM   ISOP   HR

  2,4-JIUEIHYLflCTANE                10.0
  2,4-BltSETHYlPENTANE                7.0
  2,5-BIHETHriHEPTAHE                9.0
  2,5-DIHETHYLHEXANE                 8.0        •     •
  2,6-JIKETHYLflCTANE                10,0
  2,H>MTHYLSTYRENE                      1,0               1.0
  2-BUIYLTETRAHYIROFORftN             6.0                                1.0
  2-BUTYNE                           3.0                                            1.0
  2-ETHYLHEXANOL                     8.0           -
  2-ETHYL-HUTENE                   5.0                          1.0
  2-Eim-l-HEXAHflL         '8.0
  2-FBRFHRAL                         1.0   1,0                          1.0
  2-HEXEHE                 •          2.0                                2.0
  2-HETHYOEMNE                    11.0
  2-HETHYLHEPT(\NE                    8.0
  2-HETHYLHEXftWE                     7.0
  2-HETHYL8CT/SHE                     9.0
  2-lfETHYLPEHIAHE                   4,0
  2-HETHYLPROPftHE                   4.0
 2-KETHYLPROFENE                   2.0  1.0
 2-liETHYLPROPENE                   2.0  1.0
 2-«ETKYL-l,3-B«TASIENE                                                       1.5
 2-RETHYL-l-BUIEHE                 4.0                      '     1.0
 2-BETHYL-l-PEHTENE          '       5.0                           1.0
 2-HETHYL-2-BOTENE                 3.0   1.8
 2-KETHYL-2-PEHTEHE                 4.0   1.0
 2-8ETHYL-3-HEXMONE                7.0
 2-(2-BOTOXYETHQXYHTMffil         4.0                                2,0
 3,3-IIHETHYLPEKTftHE                7.0
 35H!lfiE!(lYLOCTAHE                10.0
 3,5,5-TRIHETHYLHEXftHE              9.0
 3,5-5IHETHYiHEPT(\HE                9.0
 3-HEPTENE                  .        3.0                                2.0
 3-NETHYlHEPTftNE                    8.0
 3-«EIHYLHEXftNE -                   7.0
 3-»ETHTlO£TflKE                     9.0
 3-KETHYLPEHTftNE    '                6.0
 3-BETHYL-i-IOTEffi                  3.0   1.0
 3-!!EIHYL-l-PEHIENE                 4.0   1.0
 3-(fETHYL-CIS-2-PEHTENE    '         4.0   1.0
 3-HEIHYL-TRAN8-2-PENTEHE           4.0   1.0
 3-(CHLORQHETHYL)-S)EPTftHE           8,0
 4,HETHYLENE SlftNILINE             1.0              u                           5 0
 4-lfETHYLANILINE                                     i.o
 4-HETHYLHEPTANE                    8.0
 4-HETHYLHDHME                   JO.O
 4-HETHYLOCTAHE                     9.0
 4-«ETHYL-l-PEHTESE                 4.0    1.0
 4-BETHYL-CIS-2-PENTENE              2.0                                 z.O
4-HETHYL-TRflHS-2-PENTEHE           2.0                                 2*0
4-PHENYL-l-BUTEHE                   1.0   1.0         1.0
ACEHftPHTHENE                      1.0                     i.o                     3 fl
ACEHAPHTHYLEKE                          l.fl               }.o                     2'0
ACETALIEHYSE                                                "          1 0         *"
                                        C-3    i

-------
  COMPOUND HADE                    PAR   OLE   ETH   TOL   XYL  FORK  ALD2  ISOP   NR

ACETIC ACID                        1.0                                            1.0
ACETIC ANHYDRIDE                   2.0                                            2.0
ACETONE                            3.0
ACETYLENE                          1.0                                            1.0
ACROLEIH (PROPEHAL)                       1.0                     1.0
ACRYLIC ACID                             1.0                                    ,   1.0
ACRYLONITRILE                      1.0   1.0
ADIPICACID                        4.0                     .                       2.0'
ALIPHATICS (per carbon)       •     1.0
ALKENE KETONE                      2.0   1.0
AHINOAHTHRA8UINQNE                 2.0'                                          12.0
ANILINE                            1.0                                            5.0
AHTHANTHRENE                             1.0         1.0   1.0                     7.0
ANTHRACENE                         1.0                    1.0    '                 5.0
AHTHRAfiUIHOHE                      2.0                                           12.0
ft-PINEHE                           8.0   0.5                           1.5
BEHZALIEHYBE                                                          1.0         5.0
BENZENE                            1.0                                            5.0
1EHZOIC ACIB                       1,0                                            4.0
BENZOPYRENES                             1.0         1.0   1.0                     3.0
1EHZOTHIAZOLE                      2.0                                            5.9
BEHZ8(a)AKTHRACEH£                       .           1.0   1.0   .                  3.0
lEHZBUJPYRENE                           1.0         1.0   1.0                     3.0
BENZO(b)FLllflRAHTHEK                                1.0   1.0                     5.0
lENZfl(c)PHENA«THRENE                                1.0   1.0.                    3.0
BEHZOUJPYRENE                           1.0         1.0   1.0                     3.0
BEHZO(g,h,i)FLUORAHTHENE                            1.0   1.0                     3.0
UHZO(9,l,i)PERYLENE               1.0              1.0   1.0    .                 4.0
BEHZO(k)FLUORANTHEHE                                1.0   I.fl          '   .        5.0.
BENZYLCHLORIEE                                      1.0'
BIPHENYL                                            i.O       .                    5.0
BIPHENYLOL                                          1.0                           5.0
BROMOD1HITROANILINE                1.0                                            5.0
BROHODIHITROBENZEKE                1.0            '                                5.0
BUTEHE          '             .      2.0   1.0
BUTOXYBUTEHE     '                  4.0   1.0                           1.0
BUTOXYETHOXYETHANOL                4.0                                2.0
BUTOXYETHOXYETHAHOL ACETATE        5.0                                2.0         1.0
BUTYL CARBITOL                     4.0                                2.0
BUTYL CELLOSOLVE                   4.0                                1.0
BOIYLACRYLATE                      4.0   1.0                                       1.0
BliTYLBEHZENE                       3.0              1.0
ByTYLBENZOATE                      5.3                                            4.0"
BOTYLBENZYLPHTHALATE               5.0              1.0                           7.0
BUTYLCYCLOHEXANE                  10.0
BBHLISOPROPYLPHTHALATE            8.0                                            7.0
BUTYRALIEHYBE                      2.0                                1.0
B-PHELLANDRENE                     4.0   2.0
B-PINEHE                 ,          8.0   1.0
Cl  COHPQUNBS (DIESEL EXHAUST)     0.01                                           0.79
CIO AROMATIC                       3.0              1.0
CIO COMPOUNDS  (DIESEL  EXHAUST)    5.48 0.189        0.289  0.245
CIO GLEFIHS                        8.0   1.0
CIO PARAFFIHS                     10.0
                                       C-4

-------
  CONPOOW NAtiE                   ' PAR   OLE   ETH   TOL   XYL  FORH  AID2  ISOP   fffi

C10H12'                             2.0                     U
mm                            10.0
C10H168                            8,0   1.0
Cll CQHPQUNBS (DIESEL EXHAUST!    10.7  0.15         t
Cll GLEFINS                        ?.0   1.0
Cll PARAFFIN                      11.0
C11B10                             3.0                     1.9
C11H140                            2.0                .     1,0                     1.0
C12 COMPOUNDS (DIESEL EXHAUST)     5.0               1.0
C12 OLEFINS                       10.0   1.0
.£12 PARAFFIN            ,          12,0
C12H22                            12.0
C13 CQHPQUNDS (DIESEL EXHAUST)     4.0               1.0
C13 PARAFFIN                     , 13.0
C14 COmmBS (BIESEL EXHAUST)     7.0               1.0
CIS COHPOUNDS (DIESEL EXHAUST)     8.0               1.0
C16 BRANCHES ALKANE               16.0
Mi COHPUONSS (SIESEL EXHAUST]     9.0               1.0
C17 COKFOUNDS (SIESEL EXHAUST)    10.0               1.0
CIS WOUNDS (DIESEL EXHAUST)    11.0       ,        1.0
C19 COHPflONSS.(DIESEL EXHAUST)    12.0               1.0                 '
C2 ALKYLANTHRACENES                3.0                     1.0                     5.0 .
C2 ALMENZANTHRACENE         -    2.0               1.0   1.0                     3.0
C2 ALKYLBEHZOPHEHAHTHREHE          2.0               1.0   1.0                     3.0
C2 ALKYLCHRYSEHES                  2.0               1.0   1.0                     3.0
C2 ALHLCYCLOHEXANE                8.0
C2 ALCYLINSAN                      3.0                     1.0
C2 ALKYLNAPTHALEHE                 4.0                     1.0
C2 ALXYLPHEHAHTHRENES              3.0-                     1.0                     5,0
C2 COHPOOHSS (SIESEL EXHAUST)     0.77       0.115   '                 .             1.0
C20 COfiPOUHSS (IIESEL EXHAUST)    13.S               i.O
C21 CBMPflUMSS (DIESEL EXHAUST)    14.0               i.O
C22 CQ8POUH8S (DIESEL EXHAUST)    15.0               1.0
C23 WOUNDS (SIESEL EXHAUST)    16.0               1.0
C24 COMPOUNDS (DIESEL EXHAUST)    17.0               1.0
C25 COMPOUNDS (DIESEL EXHAUST)    18.0               1.0
C-26 COMPOUNDS (DIESEL EXHAUST)    19.0               1.0
C27 COaPQUNBS (DIESEL EXHAUST)    20.0               1.0
C28 COMPOUNDS (DIESEL EXHAUST)    21.0               1.0
C2? COMPOUNDS (DIESEL EXHAUST)    22.0               1.0
C3 ALKYLCYCLOHEXANE                9.0
C3 ALmSTYRENE                    2.0   1.0         1.0
C3 CGKPOUHSS (DIESEL EXHAUST)     1.07 0.904                                     0.122
C3 PARAFFIN                        3.0
C30 COBPOUNDS (DIESEL EXHAUST!    23.0               1.0
C31 WOUNDS (DIESEL EXHAUST)    24.0               1.0
C32 COfiPflUNDS (DIESEL EXHAUST)    25.0               1.0
C33 COfiPflUNDS (DIESEL EXHAUST)    26.0               1.0
C34 COHPOUNDS (DIESEL EXHAUST)    27.0               1.0
C35 COHPOUNDS (DIESEL EXHAUST)    28.0               1.0
C36 CORPOUNDS (DIESEL EXHAUST)    29.0               1.0
C37 CORPOUNDS (DIESEL EXHAUST)    30.0               1.0
C38 COHPflUHDS (DIESEL EXHAUST)    31.0               1.0
C39 COHPOUNDS (DIESEL EXHAUST)    32.0               1.0
                                   C-5

-------
COMPOUND HAKE
                                   PAR   OLE   ETH   IOL   XYL  FORK  ALD2  ISOP   KR
C3/C4/C5 ALCTLBENZENES             3.8               1.0
C4 ALKYLPHEHOLS                    3.0               1.0
C4 ALKYLSTYRENES                   3.0   1.0         1.0
C4 COMPOUNDS (DIESEL EXHAUST)      3.7 0.03?
C4 OLEFIH                          2.0   1.0
C4 PARAFFIN                        4.0
M'SUBSTITUTED CYCLOHEXANE        10.0
C4 SUBSTITUTED CYCLOHEXANONE      .10.0
C40 COMPOUNDS (DIESEL EXHAUST)    33.0               1.0
C41 COMPOUNDS (DIESEL EXHAUST)    34.0               1.0
C42 COMPOUNDS (DIESEL EXHAUST).   35.0               1.0  .    .
C43 COMPOUNDS (DIESEL EXHAUST)"   36.0               1.0
C5 ALIYL CYCLOHEXANE              11.0
C5 ALKYLBENZENES                   4.0               1.0
C5 AUYLBENZEHES (UNSATURATED)     2.0   1.0         1.0
C5 ALKYLPHENOLS                    4.0               1.0
C5 COMPOUNDS (DIESEL EXHAUST)      4.6 0.045
C5 ESTER                           6.0
C5 OLEFIH                          3.0   1.0
C5 PARAFFIN                       5.0
C5 PARAFFIN/OLEFIN                 4.0   0.5     .    •
C5 SUBSTITUTED CYCLOHEXAKE        11.0
C5H100     -                       5.0
C6 ALKYLBENZENE                    5.0             •  1.0
C6 COMPOUNDS (DIESEL EXHAUST)      4.5 0.218
C6 OLEFINS                        4.0   1.0
C6 PARAFFIN                       6.0
C6 SUBSTITUTED CYCLOHEXANE        12.0
C6H1803SI3       •                 6.0
C7.ALKYLBENZENE                  • 6.0               1.0
C7 COMPOUNDS  (DIESEL  EXHAUST)      1.4               0.8
C7 CYCLOPARAFFINS                  7.0
 C7 OLEFINS                        5.0   1.0
 C7 PARAFFINS                       7.0      r
 C7H12                             5.0   1.0
 C7H120                            5.0   1.0
 C7-C16                            11.0
 C8 COMPOUNDS (DIESEL EXHAUST)     4.?   0.21              0.335
 CB CYCLOPARAFFINS                  8.0
 C8 OLEFINS                        6.0    1.0
 C8 PARAFFIN                       8.0
 C8 PHENOLS                         1.0               1.0
 C8H14                              6.0   1.0
 C8H2404SI4                         8.0.
 C? COMPOUNDS (DIESEL EXHAUST)     3.62             0.056 0.608
 C? CYCLOPARAFFINS                  9.0
 C? OLEFINS                        7.0   1.0
 C9 PARAFFIN                        9.0
 C? PHENOLS                        2.0.               1.0
 CAMPHENE                           8.0   1.0
 CAPROLACTAM                        5.0
 CARBITOL                           2.0
  CARBON BISULFIDE                   1.0
  CARBOH TETRACHLORIDE
                                                                   0.111
                                                                   0.155
                                                                                 1.0
                                                                    0.065       0.?34
                                                                    0.062
                                                                      2.0
                                                                                  1.0
                                                                                   1.0
                                   C-6

-------
      COMPOUND MAKE                    PAS   OLE   ETH   Tfll    XTL   FORH  .ALD2   !SOP    MR

   SARBOHYL SULFIDE                                        ' *                          t £l
   CARYOFHYLLENE                      9.0   3.0
   CELLOSOLVE                         2.0                                i.o
   CELLOSOLVE ACETATE                 3.6           ,,                    jj          1 0
   CHLflROBENZENE                      j.Q        -                         '           5'0
   CHLORflDIFLOQROHETHANE                                                              J'0
   CHLflRQFflRH                                                                         ^
   CHLOROPEHTAFLUOROETHftNE                                                           z'ft
   CHLOROPREHE   •                          2.0
   CHLOROTRIFLIIQROHETHANE                                                             I 0
   CHRYSENE                '                            U   u                     3i'e
   CIS-I,4-BIBETHYLCYCLOH£XAHE        8.0  '''
   CIS-2-BOTENE                                                          2  0
   CIS-2-HEPIENE                     3.0                                2.'fl
   CIS-2-HEXENE                       2.0                                2.0
   CIS-2-OCTEHE                       4.0                                2*0
   CIS-2-PEHTENE                     1.0                                2!'0
   CIS-3-HEXEHE                       2.0                                \ 9
   CflRONEE                                             LC  i.o         '          ?.0
'   CREOSOTE    ,                       u                   .1.0                    20
   C8ESQL     •                                         u
   CROTOMBEHYBE                           1.0                          i.o
   COHEffi USOPROPYL B£«ZEE)         2.0               1.0
   CYCL8HEPTAKE                       7.0
   CYCLOHEXiiHE                        6.0
   CYCLOHEXftKBL                       6.0
   EYCLOHEXAHONE             ,         6.0
   CYCL08EXEIC                        2.0                                20
  CYCLOPEHTftfiHTHRftCEHES   •           3.0                    1.0          '           50
  CYCLOPENTAHE                       5.0
  CYCLDPEKTftPflESAHTHRENES             3.0                    l  o                     55
  CYCLOPEHTft(c,d!PYl!EE              2.0              1.0   1.0                     To
  CYCIQPENTENE                       1.0      •                           20
' CYCLOPEHTYLCYCLOPEHTftHE           10.0
  BECALIHS                          10.«
  BEHATURAHT                        1.0
  HACETflKE ALCOHOL                  4.0       :                          i o
  BIBE8ZAHTHRSCEHES                  1.0               }.0  1.5         '           6  0
  HBENZOPY8EHES                    1.0'              i.o  1.0                     8*0
  JIBESZOfijiJAKTHRACESE             1.0               1.0  l.'o                     <'o   i
  SIBEHZPHEWKREffiS                l.fl               i.o  i  o                     6°0   '
  JIBUTYL ETHER                      6.0                                 u
  HBIITYLPHTHALATE                   ?.o                                             7  0   !
  5ICHLOROBEHZEHES                   1.0                                             5"0   i
  HCHLOROBIFLUOROIIETHAHE               '                                            j'j   '
  BICHLflRflHETHAE                                                                    j'0   |
  BICHOLROTETRAFLflflROETHANE                                                         2V   '
  BIETHYLCYCLOHEXAHE                10.0                                             '    ;
  BIETHYLEHE fiLYCOL                 2.0                                1 0                '
  BIETHYLHETHYLCYCLDHEXANE           11.0
  BIHYBROHAPTHALENE                 2.0                    1.0
  BIHYBRQXYHAPTHALEUEBIONE           2.0                    H
  BIISOPROPYLBEHZEKE                4.0                    }  o
\BIHETHYL flLKYL  ADINES             3.0
                                        C-7

-------
  COMPOUND MANE                    PAR   OLE   ETH   TOL   XT',  FQRH  AU2   ISOP    NR

SIKETHYLBENZriALCOHOL              1.0                     1.0
HHETHYLBUTANE                    4.0
BIKETHYL8UTAHEBIOATE               4.8                                             2.0
BIHETHYLBUTENE                    4.0   1.0            .                  •
BIHETHYLBUTYLCYCLOHEXANE          12.0
BIHETHYLCYCLOBUTANOHE              f.O
BIRETHYLCYCLOHEXAHE               8.0
BIHETHYLCYCLOPENTANE               7.0
BIKETHYLCYCL0PENTEHES              5.0   1.0
BIHETHYLBECAHE                   12.0
DIHETHYLETHER           '      .    2.0                        -                 .     •
BIHETHYLETHYLBEHZOIC ACIB          2.0                     1.0                     1.0
SIKETHYLETHYLCYCLOHEXAKE          10.0
BIHETHYLFORHAKIBE                 2.0                                             1.9
DIKETHYtHEPTftNES                  9.0             ,
DieHYLHEPTANOL                  9.0
BIHETHYtHEXABIENE                 2.0   1.0                           2.0
IIHeHYLHEXAHEHOATE               i.O                                             2.0
SIKETHYLHEXA«ES                    8.0
DIKETHYLHEXEME                    4.0   1.0
DIKETHYLINIAHS                    3.0                     1.0
BIKETHYLINDENE                    1.0   1.0               1.0
BIKEIHYLNAPHTHYRIBINE              3.0               1.0
51KETHYLKAPIHALEHE                U                     1.0
BIHEIHYLHflHAHES                   11.0
BIliETHYLOCTANES                   10.0
BIKEIHYLOCTAKflL                  10.0
BIKETHYLOCTENES                    8.0   1.0
BIHETHYLOCHNE                    ?.0                                     •        1.0
WHETHYLPEHTAHE                   7.0
8IKEIHYIPEHTAHEBIOATE             5.0                                             2.0
BIETHYLPEHTAHQL                  7.0
BIKETHYLPEKTEE        ' '         5.0   1.0
BIKETHYIPHTHALATE                 3.0       -                                      7.0
1IHETHYLTEREPHTHALATE             3.0      -                                 .      7.0
BINETHYLUHBECAE           .      13.0
BIPHEKYLETHAHE             '                         2.0
B1PROPYLEHE 6LYCOL                4.0                                 1.0
BIPROPYLPHTHALATE                 7.0                                             7.0
BIVIHYUEHZEHE                          1.0               1.0
BKETHYLPHEHYUETHAHE             4.0               2.0
B1-C8  ALWL PHTHALATE             17.0                                             7.0
BOBECEHE                         10.0   1.0
D-LIHOHEE                       4.0   1.0                           2.0
EICUSAHE                         20.0
EPICHLOROHYBRIH                   3.0
ETHAE                           0.4                                             1.6
ETHAHOLAHIKE                      0.4                                     '        U
ETHYL  ACETATE                    3.0                                             1.0
ETHYL  ACRYIATE                    2.0   1.0                                       1.0
ETHYL  ALCOHOL                     0.4                                             1.6
 ETHYL  CHLORIBE                                                                    2.0
 ETHYL  ETHER                        2.0                                 1.0
 ETHYLAMIHE                        0.4                    '                         1.6
                                          C-8

-------
    CONPOUM MADE                    FftR   OLE   ETH   TOL   XTL  FQRtl  M.D2  ISOP   NR

  ETHYLBEHZE8E                       i.O               f.o
  ETHYLBICYCLGHEPTAHE               14. 0
  ETHYLCYCLBHEXANE                   8.0
  ETHYLCYCLflPENTftNE.                  7.0
  ETHYLCYCLflPEHTENE                  5.0   1.0    '
  ETHYLDIIiETHYLBENZENE                2.0                     1.0
  ETHYLBMTHYLCYCLOHEXANE           10,0
  ETHYLBIHETHYLOCTANE           •    12.0
  ETHYLDIIIETHYLPEMTAH                ?.0
  ETHYLBIfiETHYLPHENOL                2.0                     i 0
  ETHYLEHE                -                     j.o
  ETHYLEHE H1ROHI8E                         '        '                                zo
  ETBYLEIff 8ICHLORHE                       .                                        2'o
  ETHYLENE GLYCflL                    6,4                                             ^
  ETHYLEHE OXIIE                     U                                         '    j'o
  ETHYLENEAKIHES                     (J.4                .                             ^
  ETHYLFURAS        '                 2.0   2.0
  ETHYLHEPTAKE                       ?.0
  ETHYLHEPTEHE                       7.0   1.0
  EIHYLHEXANE                        8.0   '
  ETHYLHEXAHflftTE                     7.0                       '                      t  0
  EIHYLIMAN                         3.0                   -u
  ETHYLISOPROPYL ETHER               3.0                                 I.O
  ETHYLHERCAPHK                     2.8
  ETHYLKETHYLCYCLOHEXAHE             9.0
  ETHYLHETHYLCYCLOPEHTANE             8.0
  EWIHETHYIHEXAHE                  ?.0
 ETHYLHETHYLflCTfiffi                u.o
 ETHYLOCTAHE                       10.0.
 ETHYlflCTENE                         g.o   ik«
 ETHYLPEHTEHE                        5.0   i.«
 ETHYLPHEHYlPaEKYLETHAE             1.0               1.0   1.0
 ETHYLPRSPYLCTCLflHEXAHE            11.0
 ETHYLSTYREffi                        l.o   l.fl         }.{
 ETHYIT.QLOEHE                        1.0                "    It0
 ETHYL-T-BOTYL ETHER                4.0                       '          i {
 FLUORAHTHEHE                                         1.0   1 0          '          10
 FLLIOREHE                                                   u                     ,*,
 FORHALIEHYSE                                   .                  j 5
 FflRHIC ACIB                                                       '                j 0
 FORFURYL  ALCOHOL                   1.0   2.0
 6LYCEROL                           1.5                                             j 5
 BLYCOL                             o.l                                             i I
 6LYCOL ETHER                       0.8
 6LYOXAL                           l.o                           lfl                " '   ;
 HEHEICOSAfff                       21.0                                                   !
 HEPTABIENftL                       1.0   l.fl                          2  0                '
 HEPTANE                           7.0
 HEPTAHOHE                         7.0                                    .              i
 HEPTENE                           5.0   1.0                                             '
HEXABECAHE                        16.0
HEXA8ECAHOIC ACIP                  15.0                                            i 0   i
HEXABIENftL                               1.0                           2 0           *    I
HEXAFLUOROETHAHE                                                        '
                                        C-9

-------
  COMPOUND NAHE
pftR   GLE   ETH   TOL    XYL  FORM  ALS2  ISBP
HEXAHETHYIEHEIIMKIHE
HEXAHAL
HEXANE
HEXENE
KEXYLENE GLYCOL
HEXYHE
INSANE
mm    .
IHKHOaI2I3-cd)PrREffi
•ISOAKYL ALCOHOL
ISOAHYLBENZEHE
ISOBUTANE
ISOiUTYL ALCOHOL
ISOBUTYLACETATE
ISOBOTYLACRYLATE
ISOBOTYLBEHZEHE
ISOBBTYLENE
ISOBUTYLISOBUTYRATE
ISOBHTYRAUiEHYDE
ISOREBS OF 8UTEKE
ISOHERS OF BliTYLBEHZEHE
ISOHERS OF C10H10
ISORERS OF C10H18
ISOHERS OF C11H20
ISOHERS OF C5H16
ISOHERS OF DECAHE
ISOHERS OF DIETHYLEEHZEKE
ISOHERS OF 10IECAHE
ISORERS OF ETHYLTOLUEHE
 ISOHERS OF HEPTADECAHE
 ISOHERS OF HEPTAE
 ISOHERS OF HEXANE
 ISOKERS OF NQHftHE
 ISOHERS OF OCTADECAffi
 ISOKERS OF OCTANE
 ISOHERS OF PEHTADECAHE
 ISOKERS OF PEKTAHE
 ISOHERS OF PEHTENE
 ISOKERS OF PROPYLBENZENE .
 ISOHERS OF TETRWECANE
 ISOHERS OF TRIJECAME
 ISOHERS OF KCAKE
 ISOHERS OF XYLEHE
 ISOOCTANE
 ISOPEHTANE
 ISOPREKE
 ISOPROFYL ALCOHOL
 ISOPROPYLACETATE
 ISOPRQPYLBEHZENE
 ISOPROPYLCYCLOHEXANE
 ISOPROPYLCYCLOPEHTANE
 ISOPROPYLHETHYLCYCLOHEXAHE
 ISQVALERAL1EHYBE
 LACTOL SPIRITS
6.0
4.0
6.0
4.0 1.0
6.0
5.0
1.0
0.5
1.0
5.0
4.0
4.0
4.0
$.0
4.0 1.0
3.0
' 2.0 1.0
7.0
' 2.0
2.0 1.0 ,
3.0
. 1.0
8.0 1.0
11.0
7.0 1.0
10.0
2.0
12.0
1.0
17.0
7.0
6.0
9.0
18.0
8.0.
15..0
5.0
3.0 1.0
2.0
14.0
13.0
11.0

8.0
5.0
1.5
4.0
2.0
9.0
8.0
10.0
3.0
8.0
1


• %


1.0
1.0
a$s i.o i.o
a
1.0




1.0




1.0
1.0




1.0

1.0.









1.0



1.0




1.0





                                    1.0
                                                1.0
                                                4.0
                                                1,0
                                                1.0
                                                1.0
                                    1.0
                                           1.0
                                                 1.5
                                                 1.0
                                     1.0
                                     C-10

-------
CflMPOiJNJ NAHE
PAR   OLE   £TH   TOL   XTL  FORK  M.I2  ISO?
   LIMflNEHE                           4.0    l.fl
   HALEIC ANHY8RISE                          2.0
   HETHANE
   HETHOXYETHOXYBUTAHOHE               5.0
   HETHQXYETHOXYETHANQL                3.0
   HETHOXYNAPHTHALENE                  3.6
   HETHYL ALCOHOL                      1.9
   METHYL Cll ESTER                  !2.0
   HETHYL C12 ESTER                  13.0
   METHYL C13 ESTER                  14.0
   HETHYL C14-ESTER                  15.9
   HETHYL CIS ESTER                  14.0
   METHYL  C19 ESTER                  20.0
   METHYL C20 ESTER                  21.0
  HETHYLACETATE                      2.9
  HETHYLACETOPHEHO.NE                 1.9
  METHYLACETYLEHE (PRQPYNE)           2.9
  HETHYLACRYLATE                     1.9   i.o
  HETHYLAL                           3.0
  HETHYLALLENE                        1.9   1.5
  HETHYLAHYL JEETONE                  7.6
  BETHYLAHTHRACEHES                  2.0
  HETHYLBE8ZANTHRACENES    '          1.9
  HETHYLBE8ZPHENANTHREE             l.fl
  HETHYLBIPHEMYL
  HETHYLBUTABIEffi                    i.o   2.9
  HETHYLBUTEHE                       3.0   l.fl
  METHYLBOTYL IETONE                 U
  HETHYLCARBITQL                     3.9
  HETHYLCELLOSOLVE                   1,5
  «ETHYLCKLflRIi!E
  METHYLCHRYSEHES                    1.8
  HETHYLCYCLOHEXASIEffi               1.5   i.o
 KETHYLCYCLOHEXANE   '               7.0
 METHYLCYCLflHEXEHE     '             5.0   1.0
 HETHYLCYCLeflCTAHE                  9.0
 METHYLCYCLOPEHTABIEE                     i.o
 HETHYLCYCLOPEHTANE                  U
 HETHYLCYCLOPEHTEHE                  4.0   l.fl
 HETHYLDECALIHS                    n.fl
 METHYL5ECAHES                     H.5
 HETHYL1ECENE                        ?.o   1.0
 HETHYLBIHYBRQNAPHT8ALE              3.6
 SETHYLDOSECANE-                   13.0
 HETHYL508ECAHQATE                 12.0
 HETHYLEHE SROHIIE
 HETflYLEHE CHLflRISE
 HETHYLEHEB1S(C4H4NCO)              1.0
 HETHYLEHEfbM-PHEKYLISOCYAHATE
 METHYLETHYL KETflNE                 4.fl
 HETHYLETHYLHEPTANE                lO.fl
 METHYLETHYLPENTAHOATE              7.0
METHYLFLUQRANTHEHES                 1.0
HETHYLFflRHATE                      i.o
                                                                     2.0
                                                                    i.O
                                                                    1.0
                                                        1.0
                                                  l.fl
                                                 1.0
                                                 1.9
                       1.6
                       l.S
                       1.0
                       1,5
                                                 1.0   1.0
                                                                   1.0-
                                                                   1.0
                                                                   2.0
                                                                  2.0
                                                      l.fl
                                                l.fl
                                                1.0
                                                1.0    LO
                                               1.0
                                               1.0
                                               1.0
                                               l.fl
                                               1.0
                                               i.O
                                               1,0
                                               1.0
                                               1.0
                                               1.0
5.0
3.0
3.0
5.0
                                              1.3
                                              3.0
                                             1.0
                                             1.0
                                             1.0
                                             7.0
                                             8.0
                                             LO
                                             1.0
                                             1.0
                              C-ll

-------
  COKPOUHD KftBE                    PAR   OLE   ETH   ICL   XYL  FORK  ALD2  1SOP   KR

 IIETHYLGLYOXAL                                           '      -   1.0   1.0
 HETHYIHEPTAHE                      8.0
 HETHYLHEPTANflL                     8.0
 liETHYLHEPTEME                      6.0   1.0
 HETHYLHEPTYNE                      7.0                                             1.0
 KETHYLHEXAOIENE                    1.0   1.0                           2.0
 KETHYLHEXAHAL                      5.0                                 1.0
 KETHYLHEXANE                       7.0
 HETHYLHEXEHES                      5.0   1.0
 HEIHYLIHIAHS                       2.0               "     1.0
 HETHYUHBENE  '                 ,          1.0               1.0
 KETHYLISOBUTYL KEIOHE              6.0
 NETHYUSGPROPYLCYCLOHEXAHE        10.0
 HETHYUIETHACRYLATE           '      2.0   1,0                                       1.0
 HETHYLHETHYLPROPENOATE             2.0   1.0                                       1.0,
 HETHYLKYRISTATE                   14.0  '                                           1.0
 KETHYtHAPHTHALEHES                 3.0                     1.0
 KETHYLHQHAKE                      10.0
 HETHYLKKE                       8.0   1.0
 liETHYLOCTANES                      9.0
 HETHYIPALIIITATE       .           16.0                                             1.0
 HETBYLPENTAHE                      6.0            .
 HETHYLPENTENES                     4.0   1.0
 HEIHYLPHEMIHREKES             .   2.0                     1.0                     5.0
 HETHYLPROPYLCYCLOHEXANE           10.0
•eHYLPROPYLdaKAHE                13.0
 HEIHYLSTEARATE                    18.fi                       .                      1.0
 KTHYLSTYREKE                   ,         1.0         1.0
 HETHYLUH8ECAHE                   .12.0
 «ETHYL-T-BUTYL ETHER               3.0                                 1.0
 HIKERAL SPIRITS                    6.0                                 1.0
 HYRCESE        -                   4.0   3.0
 K-BICHIOROBEHZEKE                  1.0                                             5.0
 H-IIETHYLBEHZEHE                   2.0                     i.O
 K-ETHYLTflLUEKE                     1.0  .                   1.0
 H-XYLENE            '                     .1.0
 H-XYLEffi AHB P-XYLEKE                                      1.0
 HAPHTHA                            8.0
 HAPTHAIEHE                         2.0                     1.0
 NITROBEHZEKE  .                 •    1.0                                             5.0
 KGHASEGAHE                        17.0
 KOKA6IEKE                          3.0   1.0                           2.0
 !«KE                             9.0
 mm                             7.0   i.o
 NOHEKOE                           7.0   1.0
•NOHYIPHEHOL                        8.0               1.0
 H-AHYLJENZEKE                      4.0               1.0
 H-BUTAE                           4.0
 H-BUTYL ALCOHOL                    4.0
 H-BUTYLACETATE                     5.0                                             i.O
 K-iECAKE                          10.0
 H-EQSECAHE                        12.0
 H-HEPTAJECAffi                     17.0
 K-HEXYLBEHZEHE                     5.0               1.0
                                       C-12

-------
COHPOONB  h'AHE                   PAR   OLE   ETH   TCL    XTL  FORK  AL52  ISOP   MR
N-PENTft&ECANE
M-PEHTftHE
H-fENTENE
H-PEHTYLCYCLOHEXANE
N-PHEHYLAMILINE
N-PROPYL ALCOHOL
M-PRQPYLACETATE
N-PROPYLBENZENE
N-TETRftBECANE
N-TRIBECANE
N-UNBECAHE
QWHYBROINDENES
OCTAHETHYLCYCLQTETRASILOXAHE
OCTANE
OCTAKOL
OCTATRIENE
flCTENE
OXY6EHATES
0-JICKLOROBENZENE
0-ETHYLTOLUEH,E
8-XYLENE
PALMITIC ACIB
PARAFFINS (C1H34J
PARAFFINS (C2-C7)
PARAFFINS/OLEFISS {C12-C16}
PENTABIEHE
FENTAKOL
PENTEHYHE
PEKTYLBEKZEHE
PENTYLCYCLOHEXA8E
POJTYLIDEffiCYCLOHEXfiHE
PENTYHE
PERCHL8ROETHYLENE
prnvi ryr
f LA I UCHC
PHENANTHREKE
PHENOL
PfiENYLISOCYAHAIE
PHEHYLNAPHTHALENES
FHTHALIC AHHY5RIBE
PIPERYLEKE
FflLYETHYLEHE 8LYCOL
PROPABIENE
PROPANE
PROPEHE
PROPENYLCYCLOHEXftNE
PROPIOHALBEHYBE
PROFIONIC ACID
PROPYLBENZEHE
FROPYLCYCLOHEXANE
PROPYLENE 5ICHLORIBE
PftOPYLENE fiLYCOL
PROPYLEKE OXIBE
PROPYLHEPTENES
15.0
5.0
3.0
11.0

1.5
4.0
2.0
14.0
13.0
.11.0-
9.0
8.0
8.0
8.0

6.0
4.0
1.0
1.0

15.0
25.0
4.5
13.0
1.0
5.0
2.0
4.0
11.0
10.0
4.0

1.0-
1.0
1.0
1.0
1.0
1.0
2.0

1.5
1.0
7.0
f.O
2.0
2.0
9.0
1.5
1.5
2.0
8.0


1.0












2.0
1.0







0.5
2.0

1.0


1,0







2.0

1.5

1.8
1.0







1.0
                                                 1.0                           5.0
                                                                              1.5
                                                                              1,0
                                                 1.0
                                                                 2.0
                                                                             5.0
                                                      l.o
                                                      1.0
                                                1.5

                                                     0.0
                                                                             1.0
                                                                             2 0
                                                l.o   1.0                     5.'o
                                                     1.0                     5 0
                                                                           .j.'o
                                                                "            4.0
                                                1.0   1.0
                                                                             7.0
                                                                            1.5


                                                                 1.0

                                               1.0

                                                                            1.5
                                                                            1.5
                                                                            1.0

                                               1.0   1.0                     1.0



                                C-13

-------
  COMPOUND IMISE                    PAR   OLE   ETH   ICL   XYL  FORK, ftLDZ   ISOP    NR

P-BICHLQROeENZEk'E                  1.0                       •                       5.0
P-ETHYLTOLIIEHE                     1.0                     1.0
P-TOLIIALBEHYSE                                                         2.0          4.0
P-XYLENE                                                   1.0
SEC:BUTYL ALCOHOL                  4.0
SEC-8UTYLBENZEHE                   3.0               1.0
SILOXANE
STYRENE                                  0.5         1.0
SUBSTITUTED C9 ESTER (C12)        12.0                                              1.0
TEREPHTHALIC ACID                  i.O                                              7.0
TERPEHES                         • 8.0   1.0
TETRACHLOROBEHZEKES                1.0                          -.                  5.0
TETRAFLUOROHETHAHE                                                                  1.0
TETRAHETHYLiEHZEKE                 2.0                     1.0
TETRAKETHYLCYCLOBDTENE             6.0   1.0
TEIRAHETHYLCYCLOPEKTAHE            9.0
TETRAMETHYLHEXANE                 10.0
TETRAKETHYLPENTANONE      •         9.0
TETRAHETHYLSIIAKE                  4.0
IETRAHETHYLTHIOUREA                4.0                                         '     1.0
TOLUEKE                                              1.0
TOLUEffi IIISOCYANATE                                 -1.0                          '2.0
TOLUENE ISOCYAKATE                        •           1.0                            1.0
TOTAL AROKATIC AlilKES              1.0                                              5.0
TOTAL C2-C5 ALDEHYDES              1.5                                 1.0
TRAHS-1-PHEHYLBOTENE               1.0   1.0.         1.0
TRAHS-2-BOTENE                                                         2.0
TRftHS-2-HEPTEHE                    3.0                                 2.0
TRA«S-2-HEXEH£                     2.0                                 2.0
TRftHS-2-KQHEN£                     5.0                                 2.0
TRANS-2-PEHTEffi                    1.0                                 2.0
TRAKS-3-HEXENE                     2.0                                 2.0
TRICHLOROBEHZEKES                  1.5                                              5.0
TRICHLOROFLUQROhETHANE                •                   _         .                 1.0
TRICHLDROTRIFLUOROETHftHE                                 '                          2.0
TRICHOLROETHYLEXE                              1.0
TRIETHYLEHE 6LYCOL                 2.0                       .          2'.0
TRIFLUORONETHA«£                                                                    1.0
TRIKETHYLAHIHE                     3.0
TRIHETHYLBEIJZE)£                   1.0                     1.0
TRMETHYLCYCLOHEXANES              9.0
TRIKETHYLCYCLflHEXAOOL              9.0
TRIKETHYLCYCLOPEHTAHE              8.0
TRIXETHYLCYCLOPEHTAHONE            8.0
TRIHEIHYLDECANE                  • 13.0
TRIHETHYLBECEK                   11.0   1.0
TRIIiETHYLFLUQROSILAHE              2.0                                              1.0
TRHETHYLHEPTANES                 10.0
TRIKETHYLHEXANES                   9.0
TRIMETHYLHEXEHE                    7.0   1.0
TRIHETHYIINDAH                     4.0                     1.0
TR1HETHYLKONEKE                   10.0   1.0
TRIHETHYLQCTAKES                  11.0
TRIHETHYLPEHTADIEKE                4.0   2.0
                                      C-14

-------
                                 PAR   OLE   ETH    TOL   XTL  FORK  ALD2  ISOP   NR

TRIETMLPEHTANE               .  8.0
T-EUIYL ALCOHOL                   4.0
T-BUTYLBENZENE                    2.0               1.0                          1.0
UNIDENTIFIED                      6.0
VINYL ACETATE                     1.0   1.0                      .              1.0
VIHYL CHLORIBE                 .               1.0
XYLENE BASE ACIDS                  '                     1.0
                                      C-15   -'

-------
                            TABLE C-2.  EXAMPLE PROBLEM   -  PART I

                           COMPUTATION OF CARBON BOND CONCENTRATIONS
leasured Compound
Carbon Bond Concentration
Species
Ethyl ene
Propene
n-Butane
T-2-Butane
2, 3-Dimethyl butane
Toluene
M-xylene
Benzene
TOTAL
ppbC
20
30
;170
10
100
70
40
60
500
PPb ,
10
10
42.5
i
2.5
16.7
10
5
10
106.7
OLE

10






10
PAR
i

10
170

100


10
290
TOL





10


10
XYL






5

5
FORM








0
ALD2



5




5 |
ETH
10
:






10
NR

•





50
50
I SOP








0
     This is a hypothetical problem, and is not necessarily intended to be indicative of the
MOC composition of ambient air.
                                             C-16

-------
                         TABLE C-3.   EXAMPLE PROBLEM  -  PART 2
                             COMPUTATION OF CARBON-FRACTIONS
5-4 Class

OLE
PAR
TOL

XYL
FORM
ALD2
ETH
I SOP
NR
FAL
Concentration (ppb)1
10
290 ,
10
i
5
,0
5
10 .
0
50

Concentration (ppbr.)2
20
290
70

40'
0
10
20
0
50
500
Initial Carbon
Fraction**
"6.04
0,58
0.14

0.08
o.oo :
0.02
0.04
0.00
0.10
1.00
Final 'Carbo
Fraction4
0.04'
0.55
, 0.13
'
1
• 0.08
< i
. i 0.02
0.05
0.04
0.00
0.09
1.00
      Table C-2
2
 Computed by taking concentration in ppb times carbon numbers from Table 2-1.
Computed by dividing concentrations in ppbC by the total  NMOC in ppbC  (e.e.,  500 ppbC)
Unmeasured aldehydes added (.02 to FORM AND .03. to ALD2)  and total  readjusted to 1.00.
                                         C-17

-------
     The  individual concentrations  in  the  nine  entries  in  the right hand
column are obtained by multiplying  the concentration  of each  species times  the
number of bonds for that species found in  Table C-l.  For  example,  propene  has
a concentration of 10 ppb  in the example problem.  Table C-l  shows  that
propene has one olefin bond and one paraffin bond.  Thus,  10  ppb  are put in
each category (OLE and PAR).
     After all of'the species concentrations have been  apportioned  to the
carbon bond groups, then each column is totalled.  These concentrations  are in
ppb.  To convert to ppbC, the assumed  carbon number for each  carbon  bond class
are utilized.  These are found in Table 2-1.  When the  concentrations  in ppb
                                                         i
are multiplied by these carbon numbers, concentrations  in  ppbC are obtained.
     Each total for a carbon bond class is then divided by the total  NMOC (in
ppbC) to obtain initial  carbon-fractions.  The final step  is  to add  (3.02 to
the FORM fraction and 0.03 to the ALD2 fraction.  These  adjustments  are to
account for aldehydes which are not detected by the sampling/analytical
procedure.  The fractions are then adjusted so that they total up to  100
percent.
                                     C-18

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
*•
Procedures for Applying City-Specific EKMA

a. f-tiRFORMING ORGANIZATION NAME AND ADDRESS
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
EPA-450/4-89-012
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
... |
EPA Contact: Keith Baugues ,
FvnrM,1? Vodel :wll1ch allows the user to estimate the volatile organf'c compound
qua°?lt"sSndardqfore:zSLf0r ' ""* 1n °rder *° ich1av8 the national •?*'-«« ^
S^nSSp^vJ^Mf!  thS 1flPUtS rSqUired t0  rUn °ZIPM-4 and 6XPla1nS  h°W *> de-V^0P
  the  needed  variables.
 17.
                                   KEY WORDS AND DOCUMENT ANALYSIS
                    DESCRIPTORS
           ozone
           VOC control strategies
           photochemical modeling
 13. DISTRIBUTION STATEMENT
           Unlimited
L
EPA Form 2220-1 (3-73)
                                                  b.lDENTIFIERS/OPEN ENDED TERMS  )c.  COSATI Field/Group
                                                    - SECURITY CLASS (ThisReport)
                                                     Unclassified
                                                10. SECURITY CLASS (Thispage)
                                                   Unclassified
                                                                             21. NO. OF PAGES
                                                                                     118
                                                                               22. PRICE

-------
                                                      INSTRUCTIONS

 1.   REPORT NUMBER
      Insert the EPA report number as it appears on the cover of the publication.

 2.   LEAVE BLANK

 3.   RECIPIENTS ACCESSION NUMBER
      Reserved for use by each report recipient.

 4.   TITLE AND SUBTITLE
      Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently.  Set subtitle, if used, in smaller
      type or otherwise subordinate it to main title. When a report is prepared in more than one volume,, repeat the primary title, add volume
      number and include subtitle for the specific title.                                                            .             '

 5.   REPORT DATE
      Each report shall carry a date indicating at least month and year.  Indicate the basis on which it was selected (e.g., date of issue date of
     approval, date of preparation, etc.).              ...'.-.'                      '           .

 6.   PERFORMING  ORGANIZATION CODE
      Leave blank.

 7.  AUTHOR(S)
     Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.).  List author's affiliation if it differs from the performing organi-
     zation.                                          •                           ,                                          .

 8.  PERFORMING ORGANIZATION REPORT NUMBER                                              i                 I    (l
     Insert if performing organization wishes to assign this number.
                                                                                                                        I '
 9.  PERFORMING ORGANIZATION NAME AND ADDRESS
     Give name, street, citjf, state, and ZIP code.' List no more than two levels of an organizational hirearchy. :         ;            '

 10.  PROGRAM ELEMENT NUMBER
     Use the program element number under which the report.was prepared. Subordinate numbers may be included in parentheses.

 11.  CONTRACT/GRANT NUMBER
     Insert contract or grant number under which report was prepared.    '                         ' (

 12.  SPONSORING AGENCY NAME AND ADDRESS                                                                          '
     Include ZIP code.

 13;  TYPE OF REPORT AND PERIOD COVERED
     Indicate interim final, etc., and if applicable, dates covered.

 14.  SPONSORING AGENCY CODE
     Leave blank.                                                                                                        .

 15.  SUPPLEMENTARY NOTES                                                                                        I
     Enter information not included elsewhere but useful, such as:  Prepared in cooperation with, Translation of. Presented at conference of
     To be published in, Supersedes, Supplements,  etc.

 16.  ABSTRACT                                                                                                       '   |
     Include a brief (200 words or less}  factual summary of the most significant information contained in the report. If the report contains a
     significant bibliography or literature survey, mention it here.                                                             I
                 •                                                                                                   i
 17.  KEY WORDS AND DOCUMENT ANALYSIS
     (a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
     concept of the research and are sufficiently specific and precise to be used  as index entries for cataloging.         •              i

     (b) IDENTIFIERS AND'OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
     ended terms written in descriptor form for those subjects for which no descriptor exists.

     (c) COSATI FIELD GROUP - Field and group assignments are  to be taken from the 1965 COS ATI Subject Category List. Since the ma-
     jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
     endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will fol ow
     the primary  posting(s).                            .                                                    ,            ]

 18.  DISTRIBUTION STATEMENT
     Denote releasability to the public or limitation for reasons other than security for example  "Release Unlimited." Cite any availability to
     the public, with address and price.  /

 19.&20. SECURITY CLASSIFICATION                                                             !
     DO NOT submit  classified reports to the National Technical Information service.

21.  NUMBER OF PAGES
     Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.

22.  PRICE
     Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
   EPA Form 2220-1 (9-73) (Reverse)

-------

-------
I    }

-------