JULY 1974
EPA-450/3-74-046-b
               AIR POLLUTION
        *
             CONSIDERATIONS
  IN RESIDENTIAL  PLANNING
                    VOLUME II:
              BACKUP REPORT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
      Office of Air and Waste Management
   Office of Air Quality Planning and Standards
   Research Triangle Park, North Carolina 27711

-------
                               EPA-450/3-74-046-b
         AIR  POLLUTION

       CONSIDERATIONS

IN  RESIDENTIAL  PLANNING

            VOLUME  II:

       BACKUP  REPORT

                 by
        T. M. Briggs, M. Overstreet,
         A. Kothari , and T. W . Devitt
      PEDCo-Environmental Specialists, Inc.
          Suite 13, Atkinson Square
          Cincinnati, Ohio 45246
           Contract No. 68-02-1089


       EPA Project Officer: John Rob son


              Prepared for

DEPARTMENT OF HOUSING AND URBAN DEVELOPMENT
          Washington, D. C. 20410

                 and

     ENVIRONMENTAL PROTECTION AGENCY
       Office of Air and Waste Management
    Office of Air Quality Planning and Standards
      Research Triangle Park, N. C. 27711

               July 1974

-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers.  Copies are
available free of charge to Federal employees, current contractors
and grantees, and nonprofit organizations - as supplies permit - from
the Air Pollution Technical Information Center, Environmental Protection
Agency. Research Triangle Park, North Carolina  27711;  or, for a
fee, from the National Technical Information Service, 5285 Port Royal
Road,  Springfield, Virginia  22161.
This report was furnished to the Environmental Protection Agency
by PEDCo-Environmental Specialists, Inc. , Cincinnati, Ohio 45246,
in fulfillment of Contract No. 68-02-1089.  The  contents of this report
are reproduced herein as received from PEDCo-Environmental
Specialists, Inc. The opinions, findings, and  conclusions expressed
are those of the author and not necessarily those of the Environmental
Protection Agency.  Mention of company or product names is not to
be considered as an endorsement by the Environmental Protection
Agency.
                   Publication No. EPA-450/3-74-046~b
                                 11

-------
                       ACKNOWLEDGEMENT





     This report was prepared for the U.S. Environmental Protec-



tion Agency and the Department of Housing and Urban Development



by PEDCo-Environmental Specialists, Inc., Cincinnati, Ohio and



Vogt, Sage and Pflum, Cincinnati, Ohio.  Timothy W. Devitt



was the PEDCo Project Manager.  Principal investigators were



Mr. Terry Briggs, Mr. Mace Overstreet and Mr. Atul Kothari.



     Mr. John Robson was Project Officer for the U.S. Environ-



mental Protection Agency and Mr. Charles Z. Szczpanski served as



Project Officer for the Department of Housing and Urban Develop-



ment.  The authors appreciate the assistance and cooperation



extended to them by members of the U.S. Environmental Protection



Agency and the Department of Housing and Urban Development.
                              111

-------
The procedures presented in the manual should not be considered
accurate estimating methods.  They represent a first attempt to
present simplified procedures for determinging the impact of air
pollutants at residential developments.  The procedures presented
in the manual have not been tested empirically to determine their
validity.
     The manual has been written for-use, primarily by residential
planners and assumes the user has little or no formal training in
air pollution and related scientific disciplines.
                               IV

-------
                      TABLE OF CONTENTS
                                                          Page

                                                          vii
LIST OF FIGURES

LIST OF TABLES

1:    INTRODUCTION

     Scope of the Project                                 ^
     Objectives of the Manual                             2
     Content of Report

2:    SELECTION OF POLLUTANTS

3:    AIR QUALITY STANDARDS

4:    SELECTION OF POLLUTANT SOURCES
                                                          19
     Outdoor Pollution Sources
     Indoor Pollution Sources

5:    TRANSPORT AND DISPERSION OF AIR POLLUTANTS           2?
                                                          27
     Dispersion                                           2?
     Identification of Worst-Case Conditions              35
     Significant Sources

6:    POLLUTANT EMISSION AND DISPERSION PROCEDURES         39

     Roadways                                             . .
     Point Sources                                        47
     Space Heating                                        52
     Parking Lots                                         54
     Airports

7:    ANALYSIS FOR SITE DESIGN                             63

     Research                                             6_
     Conclusions
                               v

-------
                                                           Page


8:   INDOOR-OUTDOOR RELATIONSHIPS                           71

     „                                                      71
     Research                                               71
     Findings                                               73
     Theory and Modelings                                   80
     Operational Assumptions                                g^
     Relationships                                          87
     Re c ommend at ions

9:   CONCLUSIONS                                            97

     Summary of Results                                     gg
     Limitations                                            ^g
     Present Usefulness of the Manual                       gg
     Recommendations for Further Study

APPENDIX                                                    101
                               VI

-------
                       LIST OF FIGURES

Figure                                                     Page


 5-1    Coordinate system showing distributions in
        the horizontal and vertical.
                                                            59
 6-1    Computerized point source model.

 8-1    Schematic representation of outdoor-
        indoor model.
                                                            81
 8-2    Typical outdoor pollution profile.

 A-l    Average speed correction factors.
                              VI1

-------
                      LIST OF TABLES
5 -1  Key to Stability Categories
                                                         Page
                                                          28
A-l  1975 FTP (Hot Operating) CO Emission Factors         101
                                                          1 0?
A-2  Light Duty Gasoline-Powered Vehicle Model
     Emission Ratios
                                IX

-------
                     1.0  INTRODUCTION






SCOPE OF THE PROJECT



     The contract for this project calls for production of two



documents:  (1) a manual for use by land-use planners, engineers,



or designers in evaluating air pollution aspects of residential



development, and (2) a backup technical report for use by profes-



sional air pollution specialists and urban planners, citing



the research materials used in preparing the manual, outlining



development of procedures and models, and presenting recoiranenda-s



tions.



     The principal objective for the project is to identify



site design practices that will reduce exposures to air pollutants



in residential environments, with resulting benefits to human



health.  Environmental impacts other than human health effects



are not considered.



OBJECTIVES OF THE MANUAL



     The manual was  developed as a practical calculation proce-



dure for the residential planner, who is assumed to have limited



background in the scientific/technological aspects of air pollu-



tion control.   We therefore omit theoretical explanations of



the procedures.  To the extent possible, procedures are presented



in an orderly, step-wise fashion to reduce confusion and the

-------
possibility of error.



     Following are some of the objectives that guided preparation



of the manual.



     1.  The methodology should enable the planner to assess



impacts of air pollution sources on any proposed residential



configuration likely to be found in this country.  The procedures



must therefore be valid for a wide range of cases, encompassing



variations in locale and type of housing.



     2.  The manual should be concise and easy for the planner



to use.



     3.  The manual should outline recommended practices for



design of building sites and for planning of  structural and



mechanical features of the buildings.  Sub-objectives are that



the practices recommended should be practical, economical,



and compatible with other characteristics ordinarily desired



by the land-use  planner or developer.



CONTENT OF THIS  REPORT



     This final  report presents the rationale that underlies



the procedures given  in the manual.   Section  2 considers the



technical basis  for selection of particulates, sulfur dioxide,



and carbon monoxide as the air pollutants to  be evaluated and



for the exclusion of  other significant air contaminants.



     Section  3 describes development  of  the pollutant standards



presented in  the manual by adapting various elements of the



National Ambient Air  Quality Standards.



     Section  4 presents the basis  for selection in evaluating



and outdoor pollutant sources to be considered in evaluating

-------
pollution impact at a residential site.
     Section 5 describes the basic dispersion model as it
is applied to the different pollutants and meteorlogical condi-
tions .
     Section 6 considers development of emission data for
the various classes of outdoor pollution sources presented in
the manual.
     Section 7 describes in detail the development of a model
for conversion of outdoor to indoor pollutant levels, citing
pertinent empirical data.
     Section 8 describes briefly the background information
on which the recommended design practices are based.  This informa-
tion is recognized to be both limited and essentially qualitative.
     Section 9 evaluates the over-all project, indicating the
perceived strengths and weaknesses, and giving recommendations
for further research efforts toward evaluation and reduction
of air pollution impact at residential sites.

-------
                2  SELECTION OF POLLUTANTS





     For this study, consideration of the impact of air pollutants



is restricted to their effects on human health.  Effects on



plants and materials are not evaluated, primarily because including



these elements would further complicate the procedures of overrid-



ing concern:  those for evaluating effects on human health.



     Pollutants selected for evaluation in the manual are particu-



lates, sulfur dioxide, and carbon monoxide.  Their effects



on human health are fairly well established, and they are widely



dispersed in the atmosphere.  Additionally, they are the only



pollutants whose atmospheric dispersion can be calculated for



local impacts and short time durations.  They are among the



few pollutants for which reliable emission data are available.



They are the only pollutants for which we can fairly reliably



determine atmospheric concentrations that can be correlated



with human health effects.



     Several significant pollutants and pollutant groups, known



to be hazardous to human health, are excluded.  In the following,



we consider these pollutants and the reasons for their exclusion.



     1)  Hydrocarbons - We evaluated in detail the procedures



for calculating hydrocarbon levels, but determined that in



the context of residences the sources for which hydrocarbon

-------
emissions could be calculated are too close to the receptor



to create significant deleterious health effects.  The national



air quality standards for hydrocarbons are based on their ability



to interact in the atmosphere to form photochemical oxidants.



The normal time required to form significant oxidant levels



is in the order of a few hours.  Procedures in the manual,



however, can adequately determine the impact only of pollutants



within a few kilometers; at this close range a hydrocarbon dis-



charge does not have sufficient time to participate in formation


                                     1 2
of oxidants to an appreciable extent. '



     2)  Nitrogen oxides -^Nitrogen oxides are strongly implicated



in acute and chronic^ respiratory disease and in systemic effects.



Additionally, they are precursors of photochemical oxidant forma-



tion in the atmosphere.  Unfortunately, since retraction of



the EPA method for sampling and analysis of nitrogen oxides



in June 1973, no reliable and accepted analytical method is



available.  For that reason, nitrogen oxides are not included



in the calculation procedures.  It is recommended that procedures



for determining impact of nitrogen oxides be added to the manual


                                                2

when reliable data and methods become available.



     3)  Oxidants - Although oxidants constitute a major  class



of air pollutants, they are secondary pollutants and are  thus



beyond the scope of the manual.



     4)  Particulate sulfates - It appears that the degradation



products of sulfur dioxide, namely particulate sulfates and



sulfuric acid aerosols, are more potent irritants than S02 itself.



Again, however, particulate sulfates are secondary pollutants

-------
and their concentrations cannot be determined adequately by


procedures of the manual.  The calculated levels of SO2 therefore


must be used to estimate particulate sulfate levels.  Because


of the paucity of available data, the validity of this assumption

             2
is not known.


     5)  Carcinogenic air pollutants - A .number of known carcino-


gens have been shown to occur in polluted air.  These include


polynuclear aromatics; azaheterocyclic compounds; various metals


such as nickel, chromium, and arsenic; asbestos fibers; and


radionuclides.  At present no method is available for determining

                                                              2
the levels of individual carcinogens, or of total carcinogens.


     6)  Lead - Airborne lead represents a serious hazard mainly


in urban areas.  Since airborne lead compounds result mainly


from auto emissions, the highest concentrations occur in areas


adjacent to heavily trafficked roadways.  No adequate predictive


model of atmospheric lead levels is presently available.  It


appears, furthermore, that lead in ambient air is not the major

                                                      3 4
source of the current health problems related to lead. '


      CO levels should be a fairly good guide to prediction


of relative lead levels, since emissions of both are pyedominently


from auto exhaust.


     7)  Asbestos - Inhalation of high levels of asbestos fibers


has been associated with asbestosis and cancer.  A long latent


period of 20 years or more usually occurs between the initial


exposure and the recognition of cancer; the latent period for


asbestosis is often- much shorter.  At present, the only epidemi-


ological data showing a definite health hazard from exposure

-------
to asbestos fiber relate to industrial workers with exposure

levels far higher than those to which the general public is

exposed.

     A residential development adjacent to an asbestos fiber

processing plant may receive high exposure levels, but we have

no data to verify this.  Additionally, there is no adequate

and reproducible analytical method for determining asbestos

fiber concentrations in air.  Section 6.0 of the manual, Recom-

mended Design Practices, includes no caution against the use

of asbestos insulation as a construction material.  There are

no data to indicate that the resulting potential levels of expo-

sure present any health hazard; or in fact, whether any exposure

does result.  Further, since asbestos insulation affords some

degree of fire protection and conserves thermal energy, any con-

demnation of its use in construction must be well justified.


     The manual presents a list of hazardous air contaminants

and of industries with which they are commonly associated (Appen-

dix B).  This material is included primarily as a warning of

the potential local health hazards presented by a large number

of pollutants.  The distance at which these pollutants should

be considered by the planner is designated as 2 kilometers,

with the proviso that realistic hazard evaluation can be obtained

through the local air pollution control agency.  The data in
                                     g
Appendix B were obtained from Sittig.

-------
                    REFERENCES
Air Pollution, Vol. I, Ed. Stern, A.C., 2nd Edition, Academic
Press, 1968.

Summary of Proceedings, Conference of Health Effects of Air
Pollution, National Academy of Sciences, prepared for the
Committee on Public Works United States Senate, U.S. Govern-
ment Printing Office, November 1973.

Kehoe, R. A.  Toxicological Appraisal of Lead in Relation to
the Tolerable Concentration in the Ambient Air, J. Air
Pollution Control Association.  19,690  (1969).

An American Chemical Society Symposium, Air Quality and
Lead.  Environmental Science Technology.  4,217 (1970);
4,305 (1970).

Stern, A. C. et al., Fundamentals of Air Pollution, Acedemic
Press, 1973.

Cralley, L. J.  Epicemiologic Studies of Occupational Dis-
eases, in:  Industrial Environmental Health, The Worker and
the Community (L. V. Cralley ed.), Academic Press, 1972.

Bloomfield, B. D. and Barrett, J. C.  Hazard Evaluation and
Control, ibid 6.

Sittig,  M.  Pollutant Removal Handbook, Noyes Data Corp.,
1973.

-------
                  3  AIR QUALITY STANDARDS






     The initial approach to defining air quality standards for



residential developments was to present acceptable, marginally



acceptable, and unacceptable levels for each pollutant.  The



hope was that within these ranges we could specify permitted



types of human exposure.  The air quality standards, however,



were set to protect the most susceptible segment of the population;



thus a safety factor is already built into the existing standards.



Because the available experimental exposure data are sparse and



difficult to interpret, one standard concentration for each pollu-



tant was all that could realistically be defined.



     The national air quality standards for short-term exposures



are expressed as the level not to be exceeded more than once



per year for that time interval.  Since the methods presented



in the manual are intended for general, nationwide application,



we avoided dealing with extreme local meteorological conditions.



Accounting for such extreme conditions would have complicated



presentation of the emission data and the dispersion models.



We therefore considered only moderate meteorological conditions,



defined as the worst case not to be exceeded more than 3 percent



of the time periods, .per year.  We applied the Larsen mathematical



model  to adjust the national standard to the 3 percent level.
                                11

-------
The following equation from Larsen was used to make  the trans-

formation :
where:  Cn  =  pollutant concentration of adjusted  standard
               at 3 percent level

        Cs  =  national air quality pollutant concentration

        Zq  =  Number of standard deviations between the  3
               percent level and the median

        Zs  =  number of standard deviation between the
               national standard and the median

        Sq  =  standard geometric deviation


     Use of this equation  requires knowledge of  the standard geomet-

ric deviation, which is a  function of location.  The standard geo-

metric deviation data presented by Larsen for different U.S. cities

were used to arrive at average standard geometric deviation levels

for determining the adjusted pollutant concentration  (C ) standard.

     Following are listed  the relevent national  air quality stan-

dards not to be exceeded more than once in the given time period

per year and the adjusted  standards not to be exceeded more than

3  percent of the time period per year :

                                 National air       Adjusted
                                 quality  std.       standard

   CO duration   -                   1 hr               1 hr
      level  mg/m                    40                 15
     duration   3                  8 hr
      level, mg/m                   10
   SO2 duration  3                   24 hr
      level  yg/m                  365  (primary)
     duration   3                  3 hr               3 hr
     level,  yg/m                 1300  (secondary)    450

   Particulates duration            24 hr              24 hr
     level  yg/m3                 260                210
                                12

-------
     The Preliminary Evaluation presented in Section 2 of the

manual employs annual average air quality standards for particu-

lates and S02.  Continuous Air Monitoring Program  (CAMP) data

is used for this analysis.  Annual average CAMP data is used

since this is more readily available data interpreted to gen-

erate worst case not exceeded more than once per year or 3

percent of the time period.

     A few words are appropriate here concerning the known health

effects of the three pollutants and their relation to the standards,

     The present national carbon monoxide standard is based

on the atmospheric concentration necessary to result in a 3 percent

blood carboxyhemoglobin (COHb) level.  This is the level at which

predictible angina has been found.  At the 3 percent CoHb level,

patients suffering from angina pectoris develop pains sooner

after exertion.  The following data show the relationships between

atmospheric carbon monoxide levels and the percentage of COHb
                               2
for 1-hour and 8-hour exposures :

                     1-hour exposure

       % COHb      Rest      Light activity      Exercis*e
                                                    46
                                                    73
                                                    99

2.0
3.0
4.0

90
143
196
mg/m3 CO
57
90
123
                     8-hour exposure
2.0
3.0
4.0
18
29
39
18
24
33
                                                    15
                                                    23
                                                    31
                                13

-------
     These data show that the national CO standard provides a



substantial margin of safety.



     It is known that altitude significantly affects the COHb



level and can cause serious problems when persons having coronary



artery disease go to high altitudes.  Such people can, however,



become acclimatized,, and the effect is moderated with time.



The manual does not account for this effect, since the EPA has



not modified the CO standard in this regard.  The manual does



emphasize that the procedures presented cannot be applied to



unusual terrains, extremes of climate, and the like.



     Health effects of paticulate and S02 are more complex and



not well understood.   No well-defined physiological responses



to these pollutants have been observed.  It does appear, however,



that the products of their interaction, particulate sulfates,



are significantly more hazardous.



     The S02 standard presented in the manual is a 3-hour value



derived by combining the Federal 24-hour primary standard and



the 3-hour secondary standard.  Again, we applied Larsen's statis-



tical techniques.  Although the secondary standard is not intended



to protect human health directly, it is less difficult to meet



than the 24-hour primary standard.



     The manual's paritculate standard was not adjusted to the



3 percent level by Larsen's technique, since that procedure would



yield unrealistically low levels.  The standard selected  (210



yg/m ) lies between the Federal primary standard  (260 yg/m )



and the secondary standard  (150 yg/m3).  To improve the accuracy
                               14

-------
of the pollutant dispersion models/ we tried to establish short-


term standards whenever possible.  For particulates, however,


a time period of less than 24 hours was not feasible, since no


adverse health effects could realistically be considered for

                      2
a shorter time period.
                              15

-------
                     REFERENCES
Larsen, R. I., A Mathematical Model for Relating Air
Quality Measurements to Air Quality Standard, EPA,
AP-89, 1973.

Personal communication with Dr. Love, National Environ-
mental Research Center, Human Studies Lab., Research
Triangle Park, N.C.

Vaughn, D. J. and Stanek, E. J.  Sulfur Dioxide Standards-
Primary More Restrictive than Secondary?  Journal of the
Air Pollution Control Association, Vol. 23, No. 12,
December 1973.

Air Pollution, Vol. I, Ed. Stern, A.C., 2nd Edition,
Academic Press, 1968.
                          17

-------
             4  SELECTION OF POLLUTANT SOURCES





OUTDOOR POLLUTION SOURCES



     The manual presents procedures for estimating levels of



pollutants from several sources:   roadways, parking lots, point



sources, space heating, and airports.  These are the source



categories known to emit particulates, S02 > and CO in such a



way as to exert significant localized, short-term impact on nearby



receptors.  Thus their proximity to a residential development



could entail hazards to health.



     Motor vehicles represent by far the largest sources of carbon



monoxide.  Under the meteorological conditions producing high



concentrations of CO from roadways, the only other significant



local sources of CO are parking lots.  The major local sources



of particulates and SO2 are activities involving the burning



of fossil fuels.  The major emitters, such as individual industrial



plants, are listed in the NEDS point source inventory.  The smaller



but more widespread area sources  are mainly space heating units.



     Although airports constitute significant local sources of



carbon monoxide, particulates, and hydrocarbons, no simple and



accurate dispersion model allowing manual computation is available.



Thus the procedure for estimating potential significance of air-



ports was included as a safeguard against use of the manual to
                               19

-------
evaluate a residential site close to a major airport.



     Following are types of emission sources not included in



the manual and the reasons for their exclusion.



     1.  Construction and demolition.  These tend to- be short-



term projects that should not affect the pollutant levels at



a site on a continuous basis.  Also, these emissions can be



estimated only very crudely.



     2.  Shipyards.  Emissions from major shipyards  can have



a significant impact on  local receptors.  Shipyard activities,



however, would affect only a few areas in the  country.  Further,



the emission rates are poorly defined, and meteorological condi-



tions at shipyard sites  tend to deviate significantly  from the



average conditions considered in the manual.   Thus,  the manual



specifies that professional help should be sought in evaluating



sites located close to a large body of water.



     3.  Railroads.  Emissions from trains averaged  over the



time periods specified in the standards are  not  significant.



Generally,  the only significant emissions are  from the railyards



The emission rates from  railyards  are poorly defined as a rule;



large railyards  are considered, however, among the point sources



recorded  in the  NEDS  forms.



     4.   Emissions from  natural phenomena such as forest fires.



There  is  no acceptable way of measuring these  emissions.  For



short-term  standards,  their  contribution  should  be negligible.
                                20

-------
INDOOR POLLUTION SOURCES


Research


Gas Cooking


The evidence is clear that gas-fueled cooking stoves add measur-


able increments of carbon monoxide and oxides of nitrogen to


indoor air.  Yocom et al. (1969)  found that gas space heaters


did not affect indoor concentrations of CO in test homes, but gas

                                           2
cook stoves did.  W. C. Eaton et al.  (1973)  measured concentra-


tions of N02 in the vicinity of the kitchen gas ranges with


the cooktop or the oven in use.  High concentrations occurred


near the stove even with a hood-type exhaust vent in operation.


This study also noted a positive correlation between usage of


gas stoves and the incidence of lower respiratory infections


among 146 Long Island families.  The sample was small, and the


time period covered only one season.   Without further evidence


from more definitive studies one cannot state categorically that


gas-fueled cooking is a causative factor in lower respiratory


illness.





Particulate Generation


Significant particulate emissions indoors are due to cooking


and smoking.  Particulates that settle are regenerated and kept


in suspension by activities of the people inside the house.


These relationships are adequately documented by Benson et al.,


and are not discussed further here.
                              21

-------
SO2 Generation


No study indicates significant indoor generation of S02 that


is not directly traceable to faulty heating equipment.


Recommended Measures


Calculations presented in the manual account for only gas-fired


cooking units and attached garages.  These calculations entail


tack-on factors for attached garages or gas cookstoves.  In for-


mulating the recommended design procedures, we strongly considered


a flat recommendation against gas cooking appliances, but decided


that present data are not yet strong enough to warrant such a


recommendation.  If further significant data are gathered in


support of the Eaton study, banning of gas cooking appliances


appears in order.  For the moment the manual recommends inclusion


of outdoor-vented hood fans.


     A number of other indoor pollutant generators, some of which


we believe warrant consideration, were not included in our


analysis:


     1.  Housecleaning, smoking of tobacco, and turbulence of

                                                    4
movement are shown to generate particulates indoors,  but these


activities are not controllable by the builder and at present


are not well quantified.


     2.  Emissions from furnaces, fireplaces, hot water heaters,


or other generators attached to a flue are considered insignifi-


cant, provided the flue connections are in proper order.  '


No data were found for evaluating the impact of fireplaces.


     3.  Gas clothes dryers also are excluded because of the
                               22

-------
 lack of dependable data.  With adequate venting to outdoors,



 the dryer is theoretically an insignificant source.



     4.  Various aerosol sprays are reported to be potential



 health hazards.  Again, use of these sprays is not controllable



 by the builder.








 Garages



     Attached - The manual recommends a positive sealing door



 between an attached garage and living space.  "Vapor barrier"



 materials in the walls between garage and living space are also



 beneficial.



     Underground - The manual does not consider the impact of



 underground garages since we found no adequate model for or data



 on the infiltration of emissions.   Only qualitative measures



 to reduce infiltration through the garage ceiling and elevator



 shafts are discussed.  The primary impact should be from the



 vent exhaust as an outside generator.  A number of design variables



 makes this a complex relationship that warrants further work.








 Future Research



Validation and quanitification of the findings on CO generation



 in kitchens and garages deserve highest priority because relation-



 ships have been demonstrated and the expected results should



 be maximum for the dollars spent.   Such research could determine



 the-effectiveness of various hood vent configurations, and of



 connecting the oven to a flue.  The research effort should include
                               23

-------
a thoroughgoing analysis of routes of pollutant travel from garage



to dwelling structure and of economical means for reducing such



movements.
                                24

-------
                      REFERENCES


1.   Yocom,  J.  E.,  W.  A.  Cote,  and W.  L.  Clink.   Summary Report
    of a Study of  Indoor-Outdoor Air  Pollution  Relationships to
    the National Air  Pollution Control Administration.   Contract
    No. CPA-22-69-14.  The Travelers  Research Corp.,  Hartford,
    Conn.  1969.

2.   W. C. Eaton, Carl M. Shy,  John F. Finklea,  James  N. Howard,
    Robert M.  Burton, George H. Ward, and Ferris B. Benson.
    "Exposure  to Indoor Nitrogen Dioxide from Gas Stoves".
    Human Studies  Laboratory,  National Environmental  Research
    Center, Environmental Protection  Agency,  Research Triangle
    Park, North Carolina.  Revised January 1973.

3.   Ferris B.  Benson, John J.  Henderson, J. E.  Caldwell.
    "Indoor-Outdoor Air Pollution Relationships:  A Literature
    Review".  Environmental Protection Agency,  National Environ-
    mental Research Center, Research  Triangle Park, North
    Carolina.   August 1972.

4.   Lefcoe, N. M.   and I. I. Inculet. Particulates in Domestic
    Premises;  I. Ambient Levels and Central Air Filtration.
    Arch. Environ. Health.  22:230-238,  February 1971.

5.   Biersteker, K., et al.  Indoor Air Pollution in Ratterdam
    Homes,  Int. J. Air Water Pollution,  1965.
                               25

-------
        5  TRANSPORT AND DISPERSION OF AIR POLLUTANTS


     The pollutants emitted at a source disperse in the surrounding

atmosphere in a manner that depends on the meteorological state

of the local atmosphere.  The dispersion process and the disper-

sion equation used in this manual are reviewed briefly here.

References 1 and 2 provide a detailed presentation.

DISPERSION

                                  1 2
Major Factors Affecting Dispersion '

     Four parameters characterize the atmospheric dispersion

process:

     1.  Wind Speed - determines "ventilation" rate.

     2.  Wind Direction - determines path of direct
         transport of pollutants.

     3.  Mixing Height - determines the depth of the
         atmosphere available for vertical spread of
         pollutants.

     4.  Atmospheric Stability - a measure of turbulence
         in the atmosphere.  A stability classification
         method, based on wind speed and solar radiation
         or cloud cover, proposed by Pasquill, is pre-
         sented in Table 5.1.
                               27

-------
     Table 5.1  KEY TO STABILITY CATEGORIES  (AFTER TURNER)2
 Surface Wind             Day                     Night
Speed  (at 10m)  Inc°ming Solar Radiation  Thinly Overcast

   m sec'1                                       or        3/8
                Strong  Moderate  Slight  4/8 Low Cloud  Cloud
2
2-3
3-5
5-6
6
A
A-B
B
C
C
A-B
B
B-C
C-D
D
B
C
C
D
D

E
D
D
D

F
E
D
D
The neutral class, D, should be assumed for overcast conditions
durina dav or night.
     Class A  is  the most  unstable  class/  and  Class F the most

 stable  one.

     High wind speed,  unstable  atmosphere,  and  "unlimited" mixing

 height  enhance the dispersion process  and therefore are favorable

 conditions for dispersal  of  pollutants.   Low  wind speed, stable

 atmosphere, and  limited mixing  height  lead to "buildup" of pollu-

 tants in the  atmosphere.   An air pollution episode could occur

 if these conditions prevail  over several  days in which significant

 amounts of pollutants  are emitted.

     The topography of a  region affects the air movement over

 the surface,  and hence affects  the  dispersion process.

 Dispersion Equation

     There are two basic  approaches to mathematical description

 of dispersion processes:   1) statistical  modeling; and  2) model-

 ing by  conservation equations.  Several variations of these  two


                                28

-------
                         3 4
approaches are available. '   The steady-state Pasquill-Gifford

dispersion equation CGaussian distribution) is amenable to hand

calculations when only a few sources must be considered and

its precision is comparable to that of more sophisticated equations

Calculations presented in the manual are based on this equation.

     The effluent from a stack normally continues upward movement

for a while before it begins downward motion.  Stack parameters,

such as gas flow rate and temperature, and meteorological condi-

tions determine extent of plume rise.  This amount of plume rise

determines the effective height of emission of pollutants, and

consequently affects the dispersion process.  A method for esti-

mating plume rise is described later.  When a source emits pollu-

tants near ground level, e.g. an automobile, the emission height

is usually taken as zero.

     The ground-level concentration, C, at a point (X, Y, 0)

due to a continuous source with an effective emission height,

H, is given by the equation below.
exp
/ 2\
H |
I 2Qzj
exp
f 2\
2a2
I yy
\l2if G v2ir G
z y
   C (X,Y,0;H) =
                        \l£-n  O          \l4Tf  Q

C = concentration of a pollutant, usually expressed as  (yg/m  )
    or (mg/m )

X, Y, Z = are coordinates of the point  (receptor) at
          which concentration is estimated, meter

H = effective emission height, meter

Q = source strength  (pollutant emission rate), gm/second

IT = a constant, 3.14
                                29

-------
a  and QZ = lateral and vertical  dispersion  coefficients,
            meter.  These depend  on  stability,  surface
            roughness, wind  speed, and distance between
            source and receptor,  a   and  a   increase with
            distance between source^and receptor.
            These also depend on  concentration  averaging
            time; and their  values are available  for
            averaging times  of a  few minutes.

u = mean wind speed, m/sec


     This equation applies over a relatively smooth terrain.

Figure 5.1 illustrates a source-receptor  system.  The X-axis

is usually oriented along the direction of wind;  and it is con-

venient to consider the source as the origin for  the coordinate

system.

     When the concentration  is to be calculated along X-axis

(i.e., along the direction of wind,  Y=O), the equation simplifies

to:

     C (X,Q,0;H) = „  Q  „
If the source discharges essentially at ground-level, then:


     C (X,0,O:O) =
                   ay az U
Estimation of Plume Rise  '  '

     A number of equations  are available for estimating plume

rise.   It is difficult to  choose among these formulas.  Reference

8 provides detailed discussion of plume rise formulas.  The

Briggs equations for neutral conditions (D stability) is used

in the manual for estimation of plume rise, AH:


     AH = 1>6 pl/3  (3'5 x *}

                                30

-------
Figure 5.1  Coordinate system showing Gaussian distribu-
          tions in the horizontal and vertical.
                           31

-------
where:



      F = 3.14 Vf



      T = ambient air temperature,  °K



      Ts =  stack gas exit  temperature, °K




      u  =  average wind  speed at stack level,  m/sec



      V^ =  stack gas flow rate, m /sec



      F  -  buoyancy  flux  parameter,  m /sec

      *

      x  =?  distance  at which atmospheric turbulence begins

           to dominate entrainment, m


      *        5/8
      x  =  14 F /    For  F less than 55



      *        2/5
      x  =  34 F      For  F greater than or equal  to 55
Figure 4-7 in the manual was prepared with this equation.


For a group of stacks, the average stack parameters are used



in calculating the plume rise.  The effective height of emissions



is obtained by adding plume rise to physical stack height.



     H = h + AH, meter



IDENTIFICATION OF WORST CASE CONDITIONS2'7'9



     Temporal and spatial emission patterns, emission rates,



and the state of the atmosphere determine the level of pollution



at a site.  As mentioned previously, an air pollution episode



could occur if conditions of low wind, stable atmosphere, and



limited mixing height prevail over several days.  The manual



is not intended, however, for use in estimating concentrations



that could occur about 3 percent of the. time.


     The concentration averaging time for a pollutant, specified
                                 32

-------
in air quality standards, is based on health effects.  In estimat-



ing worst-case concentrations for a pollutant, one must consider



temporal and spatial emission patterns, averaging time for the



pollutant, and meteorological conditions simultaneously.  Because



particulate and SO2 have similar emission characteristics, they



are treated together in this analysis; CO is considered separately.



Carbon Monoxide (CO)



     The major portion of CO is emitted essentially at ground



level by roadway vehicles and aircraft.  For ground-level sources



(consider equation 5.3), higher concentrations occur with low wind



speed and stable atmosphere.  The CO emission rate on a road is



usually highest during morning peak of traffic volume.  The concen-



tration averaging time for CO is 1 hour, and during the morning



hours stable atmosphere could prevail.  Consequently, stability



Class F, 1 m/sec wind speed, and morning peak hour are designated



as worst-case conditions for CO.



Particulate and
     The major portion of these pollutants is emitted by elevated



stationary sources (e.g. power plants, space heaters) .  For



elevated sources  (consider equation 5.2), higher concentrations



over a time period of a few minutes occur with unstable atmospheric



conditions.  However, as wind directions  fluctuate widely  during



unstable conditions, these concentrations also fluctuate consider-



ably.  These high concentrations occur near the source  (from



1 to 5 stack heights downwind) , and the concentrations decrease



rapidly downwind with increasing distance.



     Under stable conditions, the maximum concentrations occurring





                                 33

-------
for a time period of a few minutes are lower than those occurring
under unstable conditions.  Concentrations averaged over a time
period of a few hours, however, could be higher during stable
conditions because of narrow flucuations in wind direction.
Further, these maximum concentrations occur at greater distances,
and consequently, significant concentrations could occur over
large areas.
     Stable atmospheric conditions  (classes E and F) occur during
evening and early morning, and during day time neutral stability
(class D) could prevail.  Industrial activities, and hence particu-
late and S02 emission rates, are at their peak during the day.
The concentration averaging times for particulate and SO^ are
24 hours and 3 hours, respectively.  Thus, in estimating concen-
trations, the temporal variations of emission rates and stability
should be weighted over the averaging time period.  Variations
in emission rates are difficult to estimate.  The combined con-
sideration of stability, emission rates, and concentration aver-
aging times led to selection of D stability as the worst-case
condition for particulate and S02.
     For ground-level sources, the ground-level concentration
increases as the wind speed decreases.  This is not so for elevated
sources over relatively short time periods.  The plume rise is
inversely proportional to the wind speed, and the ground-level
concentration decreases exponentially as emission height increases.
Thus, maximum concentration occurs at some intermediate wind
speed.  The combined consideration of average emission character-
istics and D stability led to selection of 4.5 m/sec wind speed
                               34

-------
as the worst-case wind speed for particulate and
     For each given receptor and set of sources there is a unique
set of meteorological conditions that yield maximum concentrations
Generally, however, it is very difficult to determine the atmos-
sheric conditions of wind direction, wind speed, and stability
that will result in the maximum combined concentration from mul-
tiple sources.  Thus, D stability and a wind speed of 4.5 m/sec
are general conditions likely to result in high concentrations.
SIGNIFICANT SOURCES
     The pollutant concentration at a site due to a source depends
on:  1) rate of pollutant emissions;  2) location of the source
in relation to the site (source-receptor geometry); and  3) mete-
orological conditions.
     The pollutant concentrations due to a ground-level source
decrease as the distance between the source and the site increases,
Pollutant concentrations from an elevated source pass through
a maximum before they start decreasing with distance.  Thus,
a source located far from the site may have only marginal impact
on the site.  Further, since pollutant travel time increases
with increasing distance from source to site, if emission rates
vary markedly and the concentration averaging time is short,
the source may become insignificant.  The ambient air quality
standard (AAQS) is, of course, an important factor in determining
significance of a source.   When the concentration at a site due
to a source is less than certain percentage  (say, 5%) of the
standard, the source may be considered insignificant and need
not be accounted for in calculating total pollutant concentration
                               35

-------
at the site.  To minimize the number of required calculations, the
manual gives criteria for identifying significant sources in each
source category.  The methodologies for estimating emission rates
and concentrations are discussed separately under each source
category in the following section.
                                36

-------
                        REFERENCES
1.  Air Pollution,  Vol.  I,  Ed.  Stern,  A.C.,  2nd Edition,  Academic
    Press, 1968.

2.  Turner, D. B.  Workbook of  Atomspheric Dispersion Estimates.
    U. S. Dept. HEW,  PHS Pub. No.  999-AP-26,  1969.

3.  Darling, E. M., Jr.   Computer  Modeling of Transportation
    Generated Air Pollution. Report No.  DOT-TSC-OST-72-20,
    U. S. Dept. of  Transportation, June 1972.

4.  Rote, D. M. et  al.   Studies of the Argonne Integrated -
    Puff Model.  ANL/ES  - 9, Argonne National Laboratory,
    October 1971.

5.  Carson, J. E. and H. Moses.  "The  Validity of Several Plume
    Rise Formulas", J.Air Poll. Control Assoc., 19:862, 1969.

6.  Briggs, G. A.  Plume Rise.   TID-25075, U. S., AEC, 1969.

7-  Personal communication  with Paul Morganstern, Walden  Research,
    Cambridge, Mass.

8.  Stern, A.C. et  al.   Fundamentals of Air  Pollution, Academic
    Press, 1973.

9.  Rote, D. M. and J. W. Gudenas.  "A Steady State Dispersion
    Model Suitable  for Air  Pollution Episodes", Paper presented
    at 64th APCA meeting, June-July 1971.
                               37

-------
         6  POLLUTANT EMISSIONS AND CONCENTRATIONS


     This section describes development of techniques for estimat-

ing pollutant emissions and concentrations at the site resulting

from each major category of sources considered in the manual..

ROADWAYS

Emission Factors '  '  '  '

     The emission rate of CO on a road depends on three major

factors*:  1) number  of vehicles on a unit segment of the road;

2) types of vehicles; and  3)  mode of operation of these vehicles -

starting, accelerating, decelerating, idling, or cruising at

steady speed.  Vehicles of different size, type, and age  emit

CO at different rates.   Populations of these vehicles vary on

different types of roads in different areas.  For simplification,

we used data on national vehicle population in developing weighted

average emission rates for different modes of operation.  These

average emission rates are known as emission factors.

     Because driving  patterns  (based on mode of operation) are

distinctly different  on local roads  (collector streets, arteries)

and on freeways, the  manual provides emission factors for both

types of roadway.
* Data for estimating effects of ambient temperature are not
  available.
+ The emission control device  (if any) deteriorates with age,
  leading to higher emissions.


                               39

-------
Local Roads

     The driving cycle used by EPA in emission rate test procedures

(1975 Federal Test Procedure, FTP) is based on normal driving

patterns on local roads.  Emission factors given in the manual

for local roads are based on this cycle.  Although the driving

patterns on different local roads are similar, the average running

speeds and consequently emission factors, are different.  Therefore,

speed correction factors are used for estimating emission rates

at different speeds.  The average running speed for 1975 FTP

is 19.6 mph, and the emission rate of this test is used as a

reference.

     Measurements of average traffic speed on a local road during

rush hour are usually not available.  The posted limits on these

roads give some indication of average running speed.  Following

are the speeds used in preparing Table  4.1 for application to

local roads:

     Posted speed limit, mph   45   40   35   30,  25   15

     Average peak-hour
     running speed in          30   25   25   20   20   10
     Table 4.1, mph


     Emission factors for different years and different speeds

are calculated as follows:

     Emission factor for _ 1975 FTP emission factor for that
     a particular speed    year x speed correction factor


The data are taken from Reference 1, which gives a detailed pre-

sentation.  The 1975 FTP emission factors and speed correction

graph are in the Appendix.
                                40

-------
Freeways

     On freeways, the car travels essentially at constant speed.

The speed, however, varies with traffic volume.  The emission

ratio, presented in Reference 1 and defined below, was used in

calculating emission factors for different steady speeds.

     Emission Ratio  =  Emission rate at the steady speed
                             1975 FTP emission rate


Because this ratio is essentially constant for different years,

only one emission ratio is given for each steady speed.  The

emission ratios are tabulated in the Appendix.

Estimating Emission Density

     The volume of peak-hour traffic on a road is usually available

from the local traffic authority-  If this is not available,

the Annual Average of Daily Traffic (AADT) can be used to estimate

peak-hour traffic, which is about 8 to 15 percent of the AADT.

Chapter 3, Reference 6, gives details on estimation procedures.

Peak-hour traffic speed on a highway, if not available, can be

estimated by the procedure in the manual.  Reference 6 describes

the method in detail.

     When traffic volume and speed on a road are known, emissions

on the road, usually expressed as emission density, can be calcu-

lated as follows:

     Q_ = K x V x E,
                    CO
where:
                                 gm
     Qco ~ C0 emission density, sec_m

     V   - peak hour traffic volume, vehicles per hour
                               41

-------
         - emission factor, mile

     K   - conversion factor, calculated as below.

     K = f hour .  , ,mile>     ,
         ^'        >               = 1.73 x 1(T7
Estimating Concentrations7 ' 8 ' 9 ' 10

     The dispersion equation 5.1 can be used to estimate CO concen-

tration due to a road when the road is at grade-level and the

surrounding terrain is smooth.  The elevation  (or depression)

of a road affects the air movement pattern and, consequently,

the dispersion process.

     Some experimental dispersion data for at-grade, elevated,

and depressed roads are given in Reference 8 .  Because it is

difficult to account for all factors affecting dispersion in

an equation, we applied these data in preparing the manual.

Figure 4.3 in the manual is taken from Reference 8; and the

values are valid when the angle between the wind direction and

the road, 0, is greater than 22.5°.  When 0 is less than 22.5°,

it should be considered to be equal to 22.5°.  Further, values

in this figure apply only if the road is  "infinitely" long.

No correction factors are introduced, however, to simplify cal-

culation procedures .

Calculating Total CO Concentration from Roads

     The CO concentration at a site usually varies with wind

direction.  Since identification of the wind direction that results

in maximum CO concentrations is often difficult, CO concentrations

at the site are calculated for eight wind directions to determine

maximum CO concentration.  As it is difficult to estimate CO


                                 42

-------
concentration when the angle between the road and the wind direc-



tion is less than 22,5°, only eight wind directions are considered,



Significant Roads



     The standard used in the manual for CO is 15 mg/m .  If



the worst-case CO concentration at the site due to emissions



on a road is less than 20 percent of that standard (i.e., 0.2



x 15 = 3 mg/m ),  the road may be considered insignificant.  The



significance criteria for local roads and freeways are discussed



below.



     Local Roads



     Because traffic volume on the feeder (residential) streets



is very low, these streets are not considered in calculating



total CO concentration at the site.  For "collector streets",



the maximum traffic volume is estimated at 3000 vph.   At 25 mph



traffic speed, the CO emission density  is 2.5 x 10~  -3^—  for
                                     J                sec-m


the year 1974.  The distance at which the worst-case CO concentra-



tion due to this  road falls below 3 mg/m  is estimated to be



400 meters.  The  angle between the road and the wind direction



is taken to be 22.5°.  Thus, collector streets within 400 m radius



of the site should be considered in calculating total CO concentra-



tion.



     Freeways



     Most freeways outside of a city are four-lane roads, and



near or within the city they become six-lane roads.  The rush-



hour traffic volume on a six-lane freeway could be 12,000 vph.



At 45 mph traffic speed, the emission density on this freeway


           -2
is 4.5 x 10   (gin/sec-meter) for the year 1974.  The distance





                                43

-------
at which the CO concentration falls below 3 mg/m  is estimated



to be 2000 meters.  Thus, freeways within 2000 meters radius



of the site should be considered in calculating total CO concentra-



tion at the site.  At 1 m/sec wind speed, the pollutant travel



time for this distance is more than half an hour.  The concentra-



tion averaging time for CO is 1 hour, and the CO emissions on



a road drop considerably after rush hour.  Consequently, the



maximum distance for significant freeways is limited to 1000



meters.



POINT SOURCES



Emission Rates



     Emission rates for particulate and SO2 from point sources



are obtained from the National Emission Data System  (NEDS) prepared



by the EPA.11



     A single plant may have more than one stack emitting pollu-



tants.  Each NEDS form is considered here to represent a stack



at the source.  Ideally, each stack at the source should be evalu-



ated separately.  To minimize calculations, however, stacks



that discharge pollutants at approximately equal effective heights



are grouped together.  Stack parameters, such as stack height



and exit gas flow rate, and meteorological conditions determine



effective emission height.  Selection of the parameters and



their ranges for grouping stacks is somewhat arbitrary; it is



a compromise between amount of computation and precision of results.



The emission rates are converted from tons per year to grams



per second by use of operating schedule data.
                               44

-------
Estimating Concentrations



     The procedure for calculating effective emission heights



for elevated sources is described in Section 5.0, Estimation



of Plume Rise.



     The ground-level concentration along the wind direction



(i.e, downwind concentration) due to a point source can then



be calculated by the equation :
                 C (X,0,0:H) =
                                  "
exp
                                              H
As values of dispersion coefficients are available for a time



period of a few minutes only, this concentration value is valid



for that time period.  Because of meandering of wind, the downwind



concentrations become  lower over longer averaging times .  The



concentration averaging times for particulate and SO2 are 24



hours and 3 hours, respectively.  Hence, it is necessary to adjust



the concentration that is calculated with the equation.



     The relationship between longer and shorter averaging times



depends on a number of factors.  '    A power law relation  is



often used, but is not well established.  For 24-hour periods,



the downwind concentrations were multiplied by 0.6;  and for 3-



hour periods, the calculated value was used.



     The ground level concentration at a point not along the



wind direction can be calculated from the equation:




     C  (X,Y,0:H)  =







The expression exp
Q
0y az u
2 "
Y
L 2 °*2J
exp
is
2
Y
L 2oy2J
termed th
exp
e Coi
r H2
2a*2_
•rection Fa
ctor .
                                45

-------
     As before, this concentration is valid for a time period



of a few minutes.  But, unlike the downwind concentration, this



concentration is greater for longer averaging times because of



meandering of wind.  No definite relationship is established



for calculating concentrations over longer averaging times.



So, concentrations for longer averaging times were calculated



by adjusting the correction factor to a higher value.  The lateral



coefficient  a  increases as the atmosphere becomes less stable.



A 3-hour concentration for SO2 is taken to be the same as the



concentration for a period of a few minutes; the correction factors



were calculated by using o  for D stability.  For particulate,



correction facotrs were calculated by using a  for C stability.



     The dispersion coefficients used in the manual are taken



from Reference 7, and are valid for rural or suburban areas when



the terrain is smooth.



Total Pollutant Concentration Due to Point Sources



     When the proposed site is affected by a number of point



sources, identification of wind direction that will result in



maximum particulate and SO,, concentration is often difficult.



The ground-level concentration due to point source decreases



rapidly as lateral distance from the wind direction increases.



The procedure used for calculating maximum CO concentration there-



fore is not suitable here.  In the manual, then, concentrations



are calculated for four source-to-receptor wind directions associ-



ated with the four major point sources.  One of these wind direc-



tions could be expected to give the maximum likely concentration.



The procedure is the same for calculating particulate and S02
                               46

-------
concentrations.


Significant Point Sources


     The air quality standard given in the manual for particulate


is 210 yg/m  for 24-hour averaging time; for SO2 it is 450 yg/ra


for 3-hour averaging time.  If the pollutant concentration due


to a point source is less than 10 percent of the standard, the


source may be considered insignificant.  A downwind concentration

          3
of 40 yg/m  for a period of a few minutes was selected as the


criterion for determining significance of a source of particulate


or SO-.  Since 24-hour particulate concentration is obtained


by multiplying short-term concentration by 0.6, 40 yg/m  is about


11 percent of the air quality standard for particulate.  The


3-hour SC>2 concentration is taken to be the same as the concentra-


tion for a period of a few minutes; thus, 40 mg/m  is about 9


percent of the standard for SO- •


     In preparing the plot of distance versus source strength


(Figure 4.5 in the manual), we designated the effective emission


height as 10 meters, then obtained the normalized concentration


(C/Q) for different distances and calculated source strength


Q for concentration C = 40 yg/m .


SPACE HEATING


Introduction


     A precise analysis of pollution loads due to space heating


would involve an inventory of the spaces heated in every enclosed


structure in a prescribed area surrounding the potential site,


together with knowledge of the fuel used to heat each structure,


the unit heat load required for each space  (which would vary
                               47

-------
with structure and insulation), and the efficiency and condition
of each space-heating unit.  These data, coupled with a valid
dispersion model for the worst-case condition, would yield the
optimum result we can presently conceive.
     The major pollutants from space-heating emissions are particu-
late and SO2.  Point sources,  however, usually, contribute more
than 75 percent of these pollutant concentrations at a site. Of the
remaining 25 percent, emissions from area sources in the immediate
vicinity of  the site contribute most of the pollutant  (due to
area sources) at the site.   Since proximity is significant,
we designated a square with  a  1-kilometer side, centered on the
wind direction vector associated with maximum pollutant concentra-
tion from point sources, to  represent the area sources.  Refer-
ence 15 gives a detailed presentation.
     The procedure presented in the manual falls far short of
the ideal, chiefly because of  the overwhelming cost of collecting
the data needed to provide precise, quantitative values.  The
only method  now available for  collecting the needed data would
be a door-to-door survey of  a  2 kilometer-diameter area around the
site.  Recognizing that the  manual-user must do an analysis with
currently available data, we devised a patchwork procedure based
on a number  of information sources and involving a number of
generalizing assumptions.  The following section outlines the
procedure for analyzing the  impact of space-heating sources.
Estimating Emissions
     Values  for the amount of  residential heated space are based
                                48

-------
     on data from aerial  photographs  and "Census  Tracts",  augmented



     with  data  from field interviews  or building  permits,  or both.



     A pro-rata apportionment is  made for tracts  partially included



     in the  area influencing the  site.   A representative floor area



     was selected for each size dwelling in the census tract data.




          Data  on commercial,  industrial and institutional space  heating



     are much harder to acquire.  We  suggest a  combination of references



     to city directories  and state  "directories of manufacturers",



     and interviews in the field  as the basic data sources.



          The percentages of residences heating with  each  fuel type



     are tabulated for cities  over  10,000,  counties,  and SMSA's in



     the Census Bureau's  "Detailed  Housing Characteristics".   These



     percentages were applied to  all  building types and uses, multiply-



     ing the percentage value by  total  floor area.



          Data  on average coal and  oil  usage were taken from Reference



     11, and emission data were taken from Reference  2.  These data



     are tabulated below.
Fuel
                   Amount
           (dwelling)  (Degree-day)
                              Pollutant Emissions
                             Particulate
                 S02
Coal
0.0012 ton
20 Ib/ton
57 Ib/ton
Fuel
0.18 gallon
10 lb/103gal
43.2 lb/103 gal
          We -selected an area of 10  square meters for estimating



     emission density.  Using 1250 square feet as the average floor
                                     49

-------
area per dwelling unit, we expressed the emission data as follows:
    Pollutant
            gms x 10                           Coal        oil
    Tsec-m2)  (degree) 103 sq. ft.

    Particulate                                10.1        0.8

    Sulfur dioxide  (S02)                       28.7        3.3


Estimating Concentrations14 ' 15 ' 16

     Space heating emission sources are too numerous to be treated

individually.  Particulate and SO- concentrations due to space

heating emissions could be calculated by   (1) representing the

source with a single virtual point source, or by  (2) considering

the emissions to be uniformly distributed over the area.  The

second approach gives better results and is commonly used.  Several

methods based on this approach are available for estimating pollu-

tant concentrations.

     The method proposed by Hanna-Gifford is simple and gives

results comparable to those obtained with more complex methods.

References 15 and 16 provide detailed information.

     The pollutant concentration C is given by:
          C   =   K
                  U
     where :
          C   -  pollutant concentration,
                                    gm
          Q  - emission density  Second.m2—

          U   -  wind speed , m/sec

          K   -  function of stability, source distribution, and
                pollutant

                                50

-------
     Although this model is generally used to calculate annual



mean concentrations of a pollutant, it is also suitable for calcu-



lating short-term averages.  The values of K, suggested by Hanna-



Gifford, are 225 for particulate and 50 for SO2.   Figure 4.11



in the manual is based on these values of K.



Findings and Conclusions



     If it were not that emissions from space heating are usually



minor in relation to emissions from industries and motor vehicles,



the procedure just outlined would probably be judged to entail



too great a potential margin of error.  We include it in the



manual because it is the only method by which one can deal with



space-heating emission sources, short of the grossly expensive



door-to-door canvass method.



Re c omme n da t i o n s



     The procedure outlined could be made less cumbersome and



more accurate if the Bureau of the Census could gather and print



data by urban block giving number of rooms per dwelling and heating



fuel used, and also publish values for heated floor areas and



fuel types used in industrial and business establishments at



the tract level in the "Census of Business".  The analysis could



be strengthened also if local building inspection departments



could provide data tabulated by street address or by coordinates.



Instituting these measures would increase both the ease and the



accuracy of the space-heating analysis.  As it stands, the proce-



dure is one of the weakest in the manual.
                                51

-------
PARKING LOTS


                    18 19
Estimating Emissions   '



     The magnitude of CO emissions on a parking lot depends on



the number of cars operating on the lot at the given time and



the average time  (termed "residence time") a car operates on



the parking lot.  Residence time is a function of size of the



parking lot, the number of gates, and traffic on the adjacent



streets..  For the manual, the average residence time is taken



to be 1.2 minutes.



     Mode of operation of vehicles on a parking lot varies from



idle to about 15 mph.  Since emissions from a car at low speed



do not differ significantly from those at idle, we calculated



emission density on the parking lot on the basis of a car at



idle.  Individual parking areas range from 150 to 300 square



feet; an average is 200 square feet.  For calculating maximum



emission density on a parking lot, we assumed that 70 percent



of the cars are started in the morning rush-hour period.  Emission



density, Q, can be calculated as follows:





           Q  • sfoo  x  T  x  E  x tnnc x °-70



      where:



           N  -  number of parking spaces



           T  -  average residence time,  minute



           E  -  emission factor, 251s.
                                  mm



           A  -  average area per parking space, square meter
                                52

-------
     As the parking lot is assumed to become 70 percent empty
in a fixed time period (1 hour) , the maximum emission density
Q is independent of parking lot size.  The emission factor E
varies with years; a reference value of 15 25L_ is used in the
                                           mm
calculation of Q below.

           1.2 (min)  x   15 gms  x  10.8 (sg. ft/m2)
    g      3600 (sec )        min     200  (sq. ft   )
                    -4  gm
           2.7  x 10    - o
Estimating Concentrations  '   '
     CO concentrations at  the site due to parking lots depend
on the distance from the lot to the site, geometry of the  lot,
and meteorological conditions .  The worst-ease meteorological
conditions for CO are F stability and 1 m/sec wind speed.
     Emissions on the parking lot may be considered  to be
occuring at some point on the parking lot,  and the CO concentra-
tion can then be calculated by use of the dispersion equation.
Alternatively, the parking lot may be divided into a number of
small elements, and each element treated as a point source.  If
the parking lot is assumed to be of "infinite" length in direc-
tions perpendicular to the wind direction,  the lateral component of
dispersion equation can be taken as unity (1.0) .  The parking  lot
can then be divided into inf initesimally narrow strips , oriented
perpendicularly to the wind direction, and the concentration can
then be calculated by integrating the dispersion equation  over the
depth of parking lot.  To simplify calculation procedures, we used
                               53

-------
this approach in the manual.  The integrated form of the disper-

sion equation is presented below.

                              "a.

     C(X 0 0-H)  -  2Q   1   [1          H2
     C(X,0,0,H)  -  __  -^  | -^  exp  (_ ^^  dx
                           *,
     Q - the emission density is assumed to be constant over the
         parking lot.



     Figure 4.4 in the manual is based on this equation.

Dispersion coefficients az were taken from Reference 7, and

H is assumed to be 2 m to account for initial dispersion in the

parking lot.

AIRPORTS

     CO emissions associated with a commercial airport may be

grouped into two categories:  (1) emissions from aircraft and

ground-service vehicles at the airport; and  (2) emissions from

access vehicles in the area surrounding the airport.

     The major portion of CO comes from aircraft operations,

and the relative strengths of the sources depend on the nature

of the airport; for example, an airport with a large number of

transfer passengers may have relatively small access-traffic

volume.  CO emissions from access vehicles could be as much as

50 percent of aircraft emissions; CO emissions from ground-service

vehicles could be as much as 25 percent of aircraft emissions.

In evaluating the CO impact of an airport, the spatial and temporal

CO emission patterns should be considered.

     CO emissions due to access and ground-service vehicles occur

over a large area at and around the airport, and it is difficult

to estimate the impact without details of airport operation and


                               54

-------
without use of a computer.  As a conservative approach one could



assume that all CO emissions from aircraft occur over the runway.



The runway could then be treated as a road, and the distance



from the runway at which CO concentrations fall below a certain



level could be calculated.  Beyond this distance the airport



does not have significant impact.



     Emissions from an aircraft depend on the number of engines



and the type of aircraft.  Emissions at the airport are a function



of the number of landing and take-off operations (LTO) of these



aircraft.  Listed below are some average distributions of aircraft



types and LTO's.



                                                           18
    Distribution of operations at major commercial airports



                             Percent of LTO at airport


     Long-range jet                      38


     Medium-range jet                    49


     Turboprop                           13




     The weighted average CO emission for this distribution was


                                   2
calculated to be 41.52 kilogram/LTO .  Note that these are average



data; actual values vary widely among airports.



     Peak-hour aircraft traffic usually occurs in the morning



and constitutes about 10 percent of daily LTO.  A runway could



handle a -maximum of 60 landing or take-off operations an hour.



The average active runway length is about 1600 meters/ and the



emissions from aircraft occur at about 6 meters above ground.



The worst-case meteorological conditions for CO in morning hours



are F stability and 1 m/sec wind speed.  CO concentration of



4 mg/m  may be considered a low concentration for assessing
                               55

-------
impact of airports.  Assuming this value, we then calculated

the distance at which the CO concentration due to a runway, with

wind direction perpendicular to the runway, falls below 4 mg/m3.



         Yearly LTO            Minimum distance  between
                               the  outer boundary of  the
                               airport and  the site at
                               which  airport has insigni-
                               ficant impact,  kilometer

         Less  than 36500              1.0

         Less  than 54750              1.5

         73000 or  more                3.0
     Beyond  3 kilometer distance,  the  travel time  for a pollutant

at  1 m/sec wind  speed  is more  than 1 hour.  Thus,  beyond  3 kilo-

meter distance the  airport  does  not have  significant CO impact

for 1 hour averaging time.

RECOMMENDED  IMPROVEMENTS TO POINT  SOURCE  CALCULATION PROCEDURE

     The calculation of point  source emissions by  the present

Manual procedure  is by far  the most involved and time-consuming

procedure presented.   Simplier procedures were considered, such

as  developing industrial pollution indices  for each Standard

Industrial Classification  (SIC)  codes, as presented by Epstein,
      ? o
et  al.    However,  in  metropolitan areas  most of the particulate

and SO- pollution can  be attributed to stationary  sources.  To

develop any  confidence in the  emission rates from  the point source

emitters, the best  available data  should  be employed.  The effects

of  different pollution control equipment  can result in orders-of-
                               56

-------
magnitude differences in emission rates.  At precent the only
up-to-date point source emission data readily available on a
national basis are from the EPA National Emission Data System
(NEDS).
     Although they present all the necessary information on point
sources, the NEDS forms were not developed for general use.
Acquiring the needed data to follow the procedure presented in
the Manual may present difficulties in certain cases.   Zimmer
and Armentrout   found that state agencies had little experience
with NEDS reports and were not familiar with the other available
NEDS output formats.  Of particular interest to this project is
the available report titled "Plant Emission Summary".  This sum-
mary lists the total pollutant emissions from all sources within a
facility and could be used for the Point Source Significance
Test.  Use of the totalled emission data would eliminate the
summing of all emission points within a facility.  For larger
facilities with more than 100 emission points, the calculation
procedure can be a very tedious task.  Thus, as a minimum step
it is recommended that local air pollution agencies or HUD A-95
offices maintain complete files of NEDS forms and emission sum-
maries for their region, updated semi-annually.
     The procedure in the Manual for grouping stacks and deter-
mining effective stack height, resulting downwind concentration,
and pollutant impact from different wind directions is a lengthy
one.  If many point sources are involved, each containing mul-
tiple emission points, this procedure becomes very lengthy.  The
problem is compounded in calculations for medium and large low-
                               57

-------
density developments, since multiple cases must be run to evalu-

ate the impacts at different sight locations.

     It is recommended that a computer program be developed to

handle NEDS data to allow the user to feed in site location(s)

and receive a printout of total pollutant concentrations.  This

would eliminate the entire point source calculation procedure

in the Manual and allow the user to obtain values for cases by

simply submitting the locational coordinates of each site posi-

tion of interest.

     Because the NEDS was developed primarily as an enforcement

tool for surveillance of stationary sources, such a computer

program is not planned.  It could, however, be applied in assess-

ing the feasibility not only of residential developments but of

proposed industrial developments.  Wider use of NEDS forms and

data would then justify the maintenance of an updated data file.

     A computer model would be similar to the procedure used in

the Manual.  The overall logic flow is shown in Figure 6-1.

Advantages of the computer system over the Manual procedure other

than ease of use are:

     1.  The significance test can be developed with lower
         cut-off limits to consider more sources.

     2.  The emission point data need not be grouped, and
         estimation of plume rise is thus more accurate.

     3.  Actual meteorological data can be used to allow
         calculation of the worst case, not to be exceeded
         more than once per year.  The Manual simplifies
         by considering the worst 3% level and a wind speed
         of 4.5 m/sec.
                                58

-------
                          Input:

                      Site Locational
                        Coordinates
   NEDS
 Emission
   Data
Significance
    Test
   Input:
Meteorological
   Data
                               Significant
                                 Sources
 Pollutant
 Dispersion
   Model
                            i
        Downwind
        Pollutant
        Concentration
                        Determine
                        Worst Wind
                        Direction
                        Highest
                       Pollution
                        Impact
             Figure 6-1  Computerized point source model
                             59

-------
     4.   The wind direction resulting in the highest pollution
         level can be determined with greater accuracy.
         The method used in the Manual considered only wind
         directions corresponding to vectors drawn between
         the major point sources and the site.  A simple
         iterative technique can be used to determine the
         worst-case wind direction.

     For such a computer program to be functional it should be

available on a local basis, perferably at HUD A-95 offices.  The

turn-around time required for use of a central federal facility

computer would likely be prohibitive for local planning  applica-

tions .
                                 60

-------
                          REFERENCES


 1.   Cirillo,  R.R.  and T.D.  Wolsko.   Handbook of Air Pollutant
     Emissions From Transportation Systems,  Argonne National
     Laboratory,  ANL/ES-28,  December 1973.

 2.   Compilation  of Air Pollutant Emission Factors, Second Edition
     U.S.  EPA/ AP-42,  April  1973.

 3.   A Study of Emissions  from Light Duty Vehicles in Six Cities/
     U.S.  EPA, APTD-1497,  March 1973.

 4.   Kircher,  D.  S.  and D. P.  Armstrong.  An Interim Report on
     Motor Vehicle  Emission  Estimation/ U.S.  EPA,  Pub.  No.  EPA-
     450/2-73-003/  October 1973.

 5.   Unpublished  data  provided by G.  W. Taylor/  Mobile  Sources
     Division, Environmental Protection Service,  Ottawa,  Ont.
     K1AOH3, Canada.

 6.   Highway Procedures Workbook, Report 133,  Dept.  of  Transpor-
     tation, 1971.

 7.   Turner, D. B.   Workbook of Atmospheric  Dispersion  Estimates.
     U.S.  Dep. HEW,  PHS Pub. No.  999-AP-26,  1969.

 8.   Air Quality  Manual, Vol.  5,  Appendix to Vol.  5,  Federal
     Highway Administration, Office  of Research,  Washington, D.C.
     20590, April 1972. Available from NTIS,  Report No.  FHWA-
     RD-72-37.

 9.   Sklarew,  R.  C., Modeling  Transportation Impact on  Air Quality,
     EPA,  National  Environmental Research Center,  Div.  of Meteo-
     rology, Research  Triangle Park,  N.C. 27711.

10.   Johnson,  W.  B.   "An Urban Diffusion Simulation Model for
     Carbon Monoxide", J.A.P.C.A., 23_:490, June  1973.

11.   Guide for Compiling a Comprehensive Emission Inventory, U.S.
     EPA,.  APTD-1135, June  1972.

12.   Singer,  I. A.   "The Relationship Between Peak and  Mean Con-
     centrations",  J.A.P.C.A., 13^:336, July  1961.

13.   Montgomery,  T. L. et  al./ The Relationship  Between Peak and
     Mean SO^  Concentrations.   Tennessee Valley  Authority/ Muscle
     Shoals, Alabama.

14.   Clark, J. F.  "A  Simple Diffusion Model for Calculating Point
     Concentrations from Multiple Sources",  J.A.P.C.A., 14;347,
     September 1964.


                                  61

-------
15.  Hanna, S.R.  "A Simple Method of Calculating Dispersion from
     Urban Area Sources", J.A.P.C.A., 2_1:774, December, 1971.

16.  User's Guide for the Climatological Dispersion Model.  EPA,
     National Environmental Research Center,- Office of Research
     and Monitoring, Research Triangle Park, N.C. 27711, EPA-
     R4-73-024, June 1973.

17.  Norco, J. E. et al., An Air Pollution Impact Methodology for
     Airports - Phase I, U.S. EPA, APTD-1470, January 1973.

18.  Geomet, Inc.  Vehicle Behavior in and Around Complex Sources
     and Related Complex Source Characteristics-Sub-Task-2-
     Airports, EPA, Research Triangle Park, N.C. 27711.

19.  Ibid-Sub-Task-4-Parking Facilities.

20.  U'.S. EPA, Federal Register, 121:0329, Appendix 0, 1973.

21.  Personal communication with Edwind L. Meyer, Jr., Environ-
     mental Protection Agency, Office of Air Quality Planning
     and Standards, Research Triangle Park, N.C. 27711.

22.  Epstein, A.H., Leary, C.A., McCandless, S.T.  A Guide for
     Considering Air Quality in Urban Planning,  EPA-450/3-74-020

23.  Zimmer, C.E., Armentrout, D., Establishment of a Non-EPA
     User  System for State Implementation  Plans, EPA Contract
     No. 68-02-1001, Task 4.
                               62

-------
                  7  ANALYSIS FOR SITE DESIGN






     Analysis for site design entails a set of calculations



and procedures required to optimize the choice of alternative



site layouts for the various elements of design, such as



buildings, landscaping, parking areas, and outdoor recre-



ation.  At the beginning of the study our goal was to



quantify and formulate the variables involved with trans-



porting, mixing, concentrating, or stagnating polluted air



on or across a site and to use the resulting formulae to



develop rational design procedures.  We were greatly dis-



appointed with the results achievable at this time; quan-



titative data are scarce and the problem is complex.



RESEARCH



     The data search was particularly frustrating because



much research has been done in this area, and many opinions



stated, however subjectively, as to the air pollution con-




trol benefits of various configurations; yet almost none of



this work is quantifiably reducible to general rules.




     We discovered little in the way of conclusive new data



in this area since the excellent review by Rydell and Schwarz




in 1968 , with a few exceptions that will be noted.  We
                                63

-------
document here those design practices for which there is a

preponderance of reinforcing opinion, regardless of the

degree of precision provided by presently available data.

Effects of Building Shape and Arrangement

     Almost all the deductions concerning these effects are

"rules of thumb."  Having investigated in depth, we believe

that these relationships are much too complex to be con-

densed into a simple series of graphs to be readily manip-

ulated, by the mathematically unskilled.

  a. Urban "canyons" created by long, smooth-faced building

     walls set in parallel rows have the capacity to greatly
                                                         2  3
     increase the velocity of windflows along the faces.  '

           "Streets, like buildings,  alter microvlinate by
     changing topography and creating new land shapes.
     Canyon-like rows of tall buildings along narrow streets
     create a funnel effect, frequently doubling the wind
     speed, or, if the wind enters at a 45-degree angle,
     accelerating the velocity on the windward side and
     creating slower currents on the leeward."2

  b. Wind eddies should be considered in site design, since

     they can concentrate pollutants in the eddy areas.

          "The orientation of a building with respect to
     winds also has an important influence on the impact of
     air pollution.  Various building configurations with
     respect to winds create different sized eddies around
     the structure.  An eddy, which  is a slowly revolving
     stationary mass of air, can trap pollution, increasing
     its concentration many times.   The larger the eddies
     around a building, the smaller  the volume of the wind
     that passes by the building to  sweep the pollutants
     away."4   "As the pitch of the roof, the thinness, and
     the height and width of a building or a building block
     increases, the size of the eddies around the building
     increases.  A row of uneven roofs creating rough sur-
     faces can slow the wind, holding pollution in the area
     lonqer.

-------
  c.  Arrangement of Structures.  Arrangement of structures

     in such a manner as to block through movements of

     prevailing winds tends to trap,  pool, and eddy air.

     Therefore, long linear blocks of structures without

     breaks should be avoided if at all possible.

          "Not only is the impact of air pollution on a
     building affected by how the building changes winds and
     eddies, but by the kind of climate the edifice itself
     creates.  Placed on a slope, a building or mining
     debris can act like'an artifical hill, creating a new
     slope climate.""
   *

          "The building can block cold air from spreading
     downhill, holding the air stagnant to gather increasing
     concentrations of pollution.  In some southern cli-
     mates, houses on stilts allow hot ground air to "roll"
     under rather than through the buildings, avoiding the
     heat and any pollution carried in the wind."l

Effects of Site Grading

  a.  Sumps.  Site grading that creates low sump areas should

     be avoided.  During cold weather, these sumps collect a

     stratified body of air in which pollutants are trapped.

     As stated before, a building "courtyard" can also act

     as an artifical sump.

          "In the natural environment hills or uneven slopes
     can block up pools of cool air.   When streets or rail-
     road beds are constructed that cut through these cold
     air dams, they may create cold air floods.  If pol-
     lution is involved, air drainage may have serious
     consequences for the health of people in the valley.  A
     new highway can also create a new alley for cold air
     and pollution to settle in.  Anyone who drives knows of
     the efficiency of open-cut highways for trapping auto-
     mobile exhausts.  This principle also works in reverse:
     where there was once free drainage a railroad embank-
     ment or an artifically level highway can dam up pools
     of cold air and highly concentrated pollution."^

  b.  Road Grading.  The General Electric studies in New

     York  pointed up the fact that a road at the grade of
                               65

-------
          the surrounding topography, or somewhat higher,

          allows better dissipation of traffic pollutants

          than does a road in a cut.

Setbacks

     As a general rule, pollutant concentrations decrease

with distance from a high-traffic street or intersection.

There are enough disturbing anomalies in data, however,

(probably due to turbulence and eddying) that dependable

general relationships are not yet possible.

          "Traditionally, planners have used open spaces as
     a major tool to improve the quality of life in the
     city.  Today, we have even more reason to use this
     technique because open space, especially planted open
     space is not only aesthetically desirable, but acts to
     diminish the impact of air pollution in several ways.

          Greenery absorbs moisture and cools by evapo-
     ration, creating a cooler, more humid climate than
     stone and exposed soil.  Temperatures over grassy
     surfaces on sunny summer days are 10 to 14 degrees
     cooler than over exposed soil, and there can be as much
     as 1500 BTU per square foot less heat per season over
     grassy surfaces."®

     The buffer areas, which can be related to prevailing

winds, provide an opportunity for pollutants to be diluted
                           9
or dispersed.  Hilberseimer  considers this subject in

detail.  Others have studied wind and temperature changes

over green areas compared with built-up urban areas, the

implication being that planted strips may aid in generating

air currents that will carry away pollutants.

Landscaping

     Small-scale landscaping has shown no significant effect

in reduction of pollutants in the air; it does tend to
                                66

-------
increase air turbulence increases mixing, which results in a



lower net pollutant concentration at a given downwind point.



     Kalyuzhnyi et. al.   found that concentrations of pol-



lutants decreased by about half with 500 meters of open



space, and by two-thirds to five-sixths with 500 meters of



planted land.  He suggestes strips of green space to aid in



wind formation to carry pollution away.  He measured a 75



percent reduction in dust particle count over a 600-foot-



wide strip in Leipzig.   Wainwright and Wilson  , however,



found over a London park that the decrease in concentration



of sulfur dioxide with distance in the direction of the wind



was not related to wind speed but instead correlated closely



with variation of temperatures with height above the ground.



Parking



     Large masses of parking space should, if possible, be



avoided in favor of a more dispersed parking scheme.  Such a



scheme tends to reduce the peak pollution load on any given



structure by simple disperson, although it also tends to



increase the average exposure throughout the development.



Setbacks of buildings from parking should prove beneficial.



CONCLUSION



     The only contribution this study and the resultant



manual can make to present residential design practice is to



make the planner aware, in very general terms, of those



variables he can manipulate that are likely to decrease



pollution levels at a given site.  The only present-day
                                 67

-------
alternative to this approach is for the planner to schedule



a series of scale-model wind tunnel tests for his project.



The implications for future research are clear, since we now



have no reliable quantitative relationships on which to base



onsite or near-site analysis for residential planning.
                                68

-------
                        REFERENCES

 1.   Rydell,  C.  Peter,  and Gretchen Schwartz.   "Air Pollution
     and Urban Forms:   A Review of Current Literature,"
     Journal  of the American Institute of Planners, Vol.  34,
     No. 2 (March 1968).

 2.   Graham,  W.E.  "The ^Influence of Microclimate on Plan-
     ning," Planning Outlook,  Spring 1949

 3.   Landsberg,  H.E. "Microclimatology, "Architectural
     Forum,"  Vol. LXXXVI, March 1947.

 4.   Kuhn, Eric, "Air Flow Around Buildings,"  Architectural
     Forum, Vol. 107,  September 1967-

 5.   Kuhn, Eric.  "Planning the City's Climate," Landscape,
     Vol. 8,  Spring 1961.

 6.   Lawrence, E.N.  "Microclimatology and Town Planning,"
     Weather, Vol. 9,  August 1954.

 7.   Final Report on Study of Air Pollution Aspects of
     Various  Roadway Configurations.  Submitted to New  York
     City Department of Air Resources.  By The General
     Electric Company,  Philadelphia, Pennsylvania.  September
     1, 1971.

 8.   Olgyay,  Victor.  Design with Climate - Bioclimatic
     Approach to Architectural Regionalism, Princeton,  N.J.:
     Princeton University Press, 1963.

 9.   Hilberseimer, L.   The New City.  Chicago: Paul Theobold,
     1944.

10.   Kalyuzhnyi, D.N.,  Kostovetskii, T.J., Devydov, S.A.,
     Akselrod, M.B.  Effectiveness of Sanitary Clearance
     Zones Between Industrial Enterprizes and Residential
     Quarters, Gigiena i Sanitariya, 1962, p.  9-12, Tranky
     B. Levine.

11.   Wainwright, C.W.K., Wilson, M.J.G.  Atmospheric Pollution
     in a London Park,  Air and Water Pollution International
     Journal, Vol. 6,  1962, pp. 337-347.
                                 69

-------
          8  INDOOR-OUTDOOR POLLUTANT RELATIONSHIPS





RESEARCH



     Of great value in our search of available reference



material were the literature review by Benson, Henderson,



and Caldwell of EPA  and certain materials provided by Mr.



Henderson.  A comprehensive list of references is included



with this section.



     One of the first attempts to model the inflow-outflow



parameters of buildings was presented by Holcombe and


      2
Kalika , in a report sponsored by ASHRAE.  This report



summarizes the effects of air-conditioning devices on intake



air pollutant concentrations, and more importantly develops



formulae expressing theoretical indoor-outdoor relationships



for a number of typical air conditioning systems under



steady-state conditions.



     Also important to the methodology and theory developed


                                                      3 4
in our study were papers by Frederick H. Shair et al., '



not yet published at the time of our data search.  Also



useful were several issues of the "Proceedings"  of ASHRAE,



together with their publication "Handbook of Fundamentals."



FINDINGS



     The findings of most of the earlier studies were dis-



appointingly inconclusive.  Although measurements of indoor-
                              71

-------
outdoor concentrations of many pollutants have been taken at



a great number of locations, the usual result is a series of



concurrent readings, from which the reader must sort out



relationships or inferences.  Sulfur dioxide is the only



commonly monitored pollutant for which reasonably consistent



data can be plotted on graphs of indoor-outdoor versus


                             789 "LO "LI
outside concentration ratios.   '''



     Many tests indicate the effectiveness of filters in


                                                        14 32
removing pollen and other particulates from the indoors.- '



A number of less general findings are listed below:



  1. Gas cookstoves and attached garages contribute no-


                                         12
     ticable CO to the inside atmosphere.



  2. Tobacco smoking is an  important source of indoor


                  13
     particulates.



  3. Indoor fluctuations of pollutant concentration follow



     outdoor fluctuations closely, with a time lag and


                            12
     generally lower peaks.



  4. No relationship has been established between building



     types and indoor-outdoor pollutant ratios.


                                2
     The Holcombe-Kalika report,  though a great step



forward theoretically, does not show strong numerical



correlations in results of measurements at two Connecticut



office buildings, chiefly because of uncontrolled variables



and because the steady-state equations do not adequately



represent rapidly varying outdoor concentrations.


                 3 4
     Shair et al. '  show good correlations between test



data and model equations in studies of buildings on the Cal
                               72

-------
Tech campus.  Their equations were set up to respond rapidly



to fluctuations in outdoor pollutant levels.



THEORY AND MODELING



     In the beginning stages of the study we made a number



of false starts in attempts to model indoor-outdoor pol-



lutant relationships on the basis of data then available.



First we tried to set up the indoor-outdoor ratio as the



dependent variable, with building-type categories as the



independent variable.  We reasoned that the potential user



of the manual would be more familiar with building types



than with some of the more theoretical and mathematical



variables required for the more rigorous approach.



     We soon found that the system of building type clas-



sification was inadequate to our needs, even with adap-



tation.  The significant variables seemed to be building



volume and surface area, which vary widely within each



building category, and air circulation and filtration



characteristics, which do not directly relate to building



types.  When this set of deficiencies became apparent, we



decided to set up a theoretical model based on the concept



of the building shell forming a system boundary for a closed



container with good internal mixing of constituent gases or



suspended matter.



     The Holcombe-Kalika study provides good basic math-



ematics for use in steady-state conditions, but did not



respond to actual fluctuations in outdoor concentrations and



did not 'provide good enough correlations with actual test
                                73

-------
conditions.  We therefore decided to proceed with a general



model that would respond to rapidly changing outdoor con-



ditions and indoor generation of pollutants.



METHODOLOGY FOR CALCULATING INDOOR POLLUTION CONCENTRATIONS



     The model we constructed for predicting indoor air



pollutant  concentrations makes  use of  data on  permeability



factors for exterior walls of the structure, structural



dimensions and volume, characteristics of  the  air circu-



lation and filtration system, and internal generation.  The



model is simply an accounting system for tracing the move-



ments of various air massed into, out  of,  and  within a



dwelling unit over short increments of time.



     Although the movement of air massed into  and inside a



dwelling is continuous, we thought it  well within the limits



of  accuracy of available data to express the mathematics in



terms of net concentration changes occurrring  within short



time segments.  The mathematics thus could be  simplified



into a  format more easily manipulated  and  computer-programmed.



As  data from continuously monitored indoor-outdoor environ-



ments are  accumulated and a more, precise method is developed



to  account for infiltration factors and rates, this model



could be revised to  incorporate continuous-change-state



mathematics, if this  is  deemed  desirable.



     The general equation for the change in pollutant con-



centration in a dwelling space  over a  short period of time



can be  expressed as  follows:
                               74

-------
Definitions

C1  =  Concentration in interior space in the beginning of
       the incremental period.

C-  =  Concentration in the interior space at end of incremental
       period.

C   =  Concentration in the exterior space adjacent to
       exterior walls.

C   =  Concentration in garage adjacent to exterior wall
 '     (also in carport area adjacent to wall)

C   =  Concentrations within the units above and/or below the
 v     unit under evaluation.  During the heating season,
       vertical infiltration from thelower level could have a
       significant effect; during cooling season, the
       opposite could occur.  This factor requires a good
       deal more information.

C   =  Concentration at exterior air intake of a forced air
       system.

Q   =  Generally, incremental quantity of influent or effluent
       air in a given period.

Q , Q , Q , Q_ are quantities corresponding to the above C ,

V Cv' V
O   -  Quantity of air from dwelling space recirculated.

V   B  volume of dwelling unit or interior space in question.

g   =  Amount of interior generation during the incremental
       period (grains)

R   «  An expression 'of the attenuation rate of reactive
       pollutants on interior surfaces
e   =  Filtration efficiency, percent removal for single pass

       e  =    efficiency of make-up air filter

       •  *    efficiency of recirculation air filter
                                    75

-------
     Figure  8-1 illustrates the relationships involved.



Mixing Factors:



     The formula assumes complete mixing between applications



of the formula, an assumption most nearly assured of being



correct in applications that involve a typical forced-air



system.  If  the quantity of recirculated air approaches or



exceeds the  volume of the dwelling unit, mixing seems



virtually assured.  The most obvious way to deal with the



unknown "mixing effect" is to "calibrate" the formula by



comparison with monitored data from buildings having known



infiltration, circulation, and volume parameters.



Testing the  Indoor-Outdoor Pollution Model



     New data from Cal Tech  (Shair, et al.) are based on a



modified form of the formula we have set up.  In tests of



their formula against monitored readings in various build-



ings on the  Cal Tech campus, correlations were good.



     We decided that the best available test of our form of



the formula  would be to apply it to the same building con-



figurations  reported in the Cal Tech study and to determine



correlations with the Cal Tech formula and with the moni-



tored readings.  We therefore programmed the data and applied



the model to two configurations of the Dabney Hall location



at Cal Tech.  The results correlate very well with data from



the Cal Tech formula and somewhat less well with the data



obtained in  monitoring pollutant levels.  In each case the



correlation  is much stronger than the input data.
                               76

-------
          FILTER
                            EXHAUST FAN
             RECIRCULATED AIR
          FILTER
                            EXHAUST AIR
                                           BUILDIMG VOLUME V
                                           INDOOR SOURCES g
                                           SINKS R
Figure 8-1  Schematic  representation of outdoor-indoor model.

-------
     Our first run-through of the Dabney Hall data assumed


5-minute time increments between iterations of the formula.


We theorized that an even closer correlation with data from


the Cal Tech formula would be achieved by use of a shorter


time increment.  Therefore we ran the data through again


using a 2-minute increment; the resulting values, however,


were virtually identical with those obtained with the 5-


minute increments.  From this we can guess that  (1) possibly


the. Cal Tech formula is slightly flawed by deletion of


certain small factors for mathematical convenience, or (2)


perhaps a very small time increment would be required to


match the integral formula.  Since the correlation was


adequate for our purposes, we did not test the model further.


Computer Investigation of Cases


     Having been validated with Shair's test results, the


model seemed sufficiently accurate for use in exploring the


impact of variables in building and mechanical design and


construction on interior air pollution.  A variety of


prototype cases were analyzed by means of the computer.


Following is a short summary of the building types and


variables tested:


Building Types and Floor Areas


1.   Single-Family Dwellings

                 2
     A.   1000 ft  single-story

                 2
     B.   1600 ft  single-story


     C.   2000 ft2 two-story
                              78

-------
2.   Low-Rise Apartments



     A.   Two-story single-load (or townhouse)



     B.   Three-story double-load  (SDU/floor)



3.   High-Rise Apartments



     A.   Single Long-corridor 10-story, 20 DU/floor



     B.   Double short-corridor, 12 DU/floor



     C.   Three-wing composite 30-story, 36 DU/floor



Heating/Cooling Systems



1.   Hot-water or steam radiator



     A.   Closed-window



     B.   Open-window



2.   Forced-air Systems



     A.   Unfiltered



     B.   Filtered



          (1)  Efficiency =0.2



          (2)  Efficiency =0.9



          (3)  Filter on return air only



          (4)  Filter make-up air only



          (5)  Filter on make-up and return air



Structural Permeability Variables



1.   Modern "tight" building



2.   Old "leaky" building



Source Variables



1.   Interior generation



2.   Infiltration from subterranean garage



3.   Make-up air intake at a low-pollution location



4.   Pollution reaction with walls, floors, ceilings
                               79

-------
5.   Exterior levels were assumed to follow a prototypical
     two-humped curve, with morning and afternoon peak
     levels, as indicated in Figure 8-1.  Also a constant
     exterior level representing industrial TSP and S02
     emissions was included as in input.


OPERATIONAL ASSUMPTIONS

     Where data were sufficient to allow formulation of a

statistically correct  (i.e., 97th percentile, etc.) para-

meter, we attempted to insert into the model a conservative

condition.  The following section attempts to explain and

justify some of the assumptions made.

Outdoor Concentration  Time Gradient

     The simplified outdoor concentration profile  shown in

Figure 8-2 was used to represent a typical time-concentration

relationship.  The type of pollution sources and prevailing

meteorological conditions determine the shape of this curve.

This two-humped curve  is, however typical for locations with

significant impact from roadways.

Wind Driving Pressure

     Wind driving pressure equivalent to that generated by a

10 mph wind (0.05 in. of H2O) was selected as the maximum

that might be expected in conjunction with high-pollution

conditions.  Statistically, this seems quite conservative,

but this factor was possibly overdone in order to compensate

for factors not considered in the model because of lack of

data, such as chimney effect.  An equal vacuum was assumed

on the leeward side of the structure.
                              80

-------
         40-r
         30-
     UJ
     CJ

     o
     o
20
00
     o
     °-   10-
         0
                                                              •
^
 \
                                                                            \
                          \
                                v

                              /
                                    *

                                    \
                                      \
                                  I
           7     8      9     10
                         I
                         11
      12   13
14    15    16
  TIME-HOURS
17   18    19   20    21    22
                         Figure  8-2   Typical outdoor pollution profile.

-------
Effective Infiltration Area


     We assumed that wind driving pressure acts equally on all


upwind walls.  In other words, wall area available for infiltra-


tion was taken as half of the total wall surface, deducting door


and window areas, which were assumed to be in the same propor-


tion on the upwind and downwind sides.


     We assumed that full wind pressure acted on the cooling of


the top story, with infiltration occurring over the upwind half


of the ceiling and exfiltration on the downwind half.


     We assumed that floors have no significant infiltration


driving force exerted and assigned them a zero value.  This


assumption is obviously valid for slab floors, and probably


not entirely valid for a structure with crawl space below


the floor.  A basement structure has some small potential


for infiltration through small basement windows, but such a


value is not established by available data, and subjectively


it appears small.


Infiltration Rates


     For walls we assumed that an infiltration rate of 0.5 ft /

     2
hr/ft  is representative of modern construction with vapor


barriers.  Much lower values are observed in test sections,


but occasional poor workmanship can reduce sealing effec-


tiveness; therefore we considered 0.5 a reasonable value


that is obtainable with ordinary workmanship and inspection.


For older structures we assumed an infiltration rate of 5.0
                                82

-------
ft3/hr/ft2, very much a "ballpark" estimate.  An old struc-



ture with exterior stucco and interior plaster in good



condition could duplicate the rates of modern structures.



In many older masonry structures, however, the infiltration



rates range much higher than 5.0.  The choice was based



solely on a judgment of what values would be both conserv-



ative and representative.



     For windows and doors, we assumed values of 14 ft /
hr/ft of sash or edge crack with weatherstripping and 140



ft /hr/ft without weatherstripping.



Single-Family Structures



     The first structure selected for analysis was a 1600



square foot single-story building.  We calculated surface



area from an assumed floor plan, computed length of windows



and door cracks, using the infiltration rates mentioned



earlier, computed total infiltration in ft /hr for both



"modern" and "old" structures.  For other single-family



prototypes, the infiltration rate was assumed to vary



directly with surface area of the structure.



Multi-Family Structures



     Multi-family prototpyes were selected from actual floor



plans.  We measured interior volumes and exterior surfaces



directly and again took infiltration as proportional to the



outside surfaces.  Because of the many complicating factors



and the absence of supporting data, we abandoned the attempt



to calculate infiltration for a single living unit in a
                               83

-------
structure.  The calculations for multi-family structures
assume a uniform distribution of air pollutants throughout.
Interior Generation
     All that we know definitely about indoor pollutant
generation are the general categories of sources and the
fact that in some structures with given circulation char-
acteristics these sources could increase localized pollutant
concentrations.  Because present data are not adequate to
deal with localized variations, the factors for internal
generation assume good mixing throughout the dwelling.  The
model is capable of handling any pattern of internal gen-
eration and any attenuation when valid values become avail-
able.
Data Format
     The data resulting from each computer run for each
building prototype took the form of interior concentrations
at 5-minute intervals over the period of the run.  For our
purposes, results were as good and much more convenient when
the computer continued the 5-minute iterations but printed
out readings only on the hour.
     The indoor-outdoor ratio was computed by averaging
the peak readings for the applicable period of time for both
indoor and outdoor locations.  For example, if an 8-hour
standard were under consideration, the averages of readings
for the highest 8-hour period in the day were computed for
both indoor and outdoor concentrations.  These periods did
                                 84

-------
not usually coincide because of the time lag required for



infiltration of pollutants to exert an effect on indoor



levels.  We then divided the high indoor level by the high



outdoor level and plotted the result.



     The indoor-outdoor ratios and periods selected for



inclusion in the manual were determined by the time periods



finally selected for the air quality standards.  The total



range of ratios calculated and graphed, for the two-humped



curve characteristic of auto emissions, included ratios for



interval of 1, 3, 8, and 24 hours.  For the uniform emission



rate characteristic of industrially generated SO  and



particulates, the 3-hour and 24-hour ratios were calculated.



Modeling of SO2 was abandoned because of insufficient data



on internal attenuation by paint, fabrics, and other materials



We therefore deal with SO- very simplistically as a fixed



ratio.



RELATIONSHIPS



     The key variables in indoor-outdoor pollutant levels



appear to be interior volume of the structure as related to



surface area.  For any given set of conditions of air



circulation, permeability, filtration, and other parameters,



the plot of volume-to-surface-area ratio  (V/SA) versus



pollutant reduction shows a consistent and significant



trend.  It is possible therefore to deal with the design



problem in a graphical format.  With this goal in mind, we



plotted the results of our studies with V/SA on the Y-axis



and the proportion of indoor to outdoor concentrations on
                               85

-------
 the  Y-axis.
      Some  cases  modeled did  not conform well  to graphing on
 this set of  axes,  at least not without  interpretation.  The
 case in which concentration  of a pollutant  at a forced-air
 system intake is significantly lower  than at  the  exterior
 walls generally, is dominated by the  concentration  of  the
 intake air.   In  most cases,  the plot  of V/SA  versus inlet
 pollution  level  exhibits the same ratio as  the plot of V/SA
 versus general outdoor pollution level, so  the significant
 and  consistent indoor-outdoor comparison  in this  case  is
 indoor to  inlet  ratio.
     Also,  a  large garage  infiltration will throw an in-
consistency into the ratio, depending on a great number of
ill-defined variables.  This  situation was handled with
"tack-on"  factors.
     Generally,  the effect of  tighter building construction
and  larger V/SA  ratio  is to slow  infiltration of pollutants
from the outside.  This evens  out the peaks and valleys and,
therefore,  is most successful  with highly fluctuating  exterior
concentration values.   The reduction of infiltration is most
apparent in calculations that entail a  pollutant standard
covering a shorter time period.   For example,  the 1-hour stan-
dard for carbon  monoxide is highly responsive  to building sealing;
the  24-hour standard for particulates responds hardly  at all.
     The effect of filtration  devices or chemical removal
agents is less linked with time.  An effective filtration
                               86

-------
device has as much effect on particulates with  the  24-hour

standard as on CO with the  1-hour standard.

     An open-window case was run to demonstrate that  indoor

pollutant levels closely follow those outdoors  when windows

are open.  The model verified this supposition  fully.

RECOMMENDATIONS

     Because of time limitations on this project, we  did not

process a number of somewhat less-typical cases that  might

have exhibited some significant variation from  our  model.  A

more significant weakness is that only two real test  cases

 (those with Dabney Hall data) were run to confirm the

accuracy of the model.  A program of further testing  should

be specifically aimed at verifying the model for a wider

range of structures,  materials,  and operating conditions.

     A great body of knowledge still requires research to

fully validate the indoor-outdoor model for dependable

everyday use by a residential planner with limited technical

training.   Investigation of the following factors could

significantly strengthen the work.

1.   Permeability of modern building materials,  including:

     Taped plaster board

     "Sandwich" construction panels

     Plywood sheathing

     Modern masonry and hollow-core block

     Masonry veneer over stud walls

     New vapor barrier materials,  such as plastic and
     aluminum or composites.
                               87

-------
 2.   Internal generation of pollutants:

      Co from stoves and fireplaces
      HC from tobacco smoking

      Dust from vacuuming and other household cleaning.

 3.   Room-to-room variations in pollution levels.

 4.   Unit-to-unit air movements in multi-family configu-
      rations .

 5.   Vertical permeability of high-rise structures, in-
      .eluding more cause-and-effect modeling.  Studies of the
      "chimney effect."

 6.   Relative permeabilities of various gases.   Is a  "vapor"
      barrier the best barrier against CO or SO2?


 7.   The graphical displays show clearly that any form of
      reasonably effective filtration in a forced-air system
      with typical recirculation rates is highly effective.
      Even a 10 percent effective filter,- for instance, can
      significantly reduce levels of some pollutants.   This
      finding suggests research into low-efficiency filters
      for the gaseous pollutants.

 8.   Dependable exterior air-current analysis.  How much
      pollutant recirculates into the structure from its own
      flue?  What are the effects of eddying?

 9.   Effects of kitchen or bathroom fan vents.

10.   Concentrations of pollutants in various garage struc-
      tures.

11.   Reaction rates of pollutants with building materials
      and furnishings.  This study would produce an alterna-
      tive factor, which is required to develop a more quanti-
      tative relationship for S02 infiltration.
                                 88

-------
                 EXAMPLES  OF  BUILDING DESIGNS

            Infiltration Rates - Single-Family Modern
Case I  -  1-Story, 1600 S.F. floor area
                   12,800 ft3 interior volume
           Wall Area  =  1,312 ft2
                         - 197 Windows @15%
                         1,115 ft2 net walls
                         x 0.5
                                 2
                           558 ft  infiltration surface walls
                           800 ft2 roof
                         1,358 ft2 Total
                         x 0.5 ft3/hr/ft2 =
                           679 ft3 infiltration through walls
           Windows - 4 -  3x5 double-hung
                          3x3=  9
                          2 x 5 = 10
                          5 x     19-95 L.F. window crack
                          1-3x7 Doors
                              20 x = 20 LF door crack
                          Total Leakage     115 LF crack length
                                           x 14 CF/hr/LF
                                          1,610 ft /hr windows, doors
                                            679 ft3/hr walls
                                          2,289
                                  Round off to 2300
                                  89

-------
           Infiltration Rates - Single-Family "Leaky"
Case 2  -  Single-story   1,600 SF floor area
                         12,800 ft  interior volume
           Wall Infiltration:
 1,115
 x 0.5
net walls
                                   558 ft  infiltration surface, walls
                                         2
                                   800 ft  infiltration surface, ceilings
                                         2
                                 1,358 ft  infiltration surface, Total
                                 x	5 ft3/hr/ft2
                                 6,790 ft /hr infilt. walls & ceilings
           Windows & Doors
   115 LF Crack Length (from Sht.  1)
 x 140 ft3/hr/ft crack
16,100 ft  through doors & windows
           Total Infiltration
 6,790 ft  walls
16,100 windows & doors
22,890
                                    90

-------
                  BUILDING PROTOTYPES SELECTED FOR MODELING




      !•  One Story,  1000 ft2              v = 8'000 Ft



         Area Walls   1040




               Roof   1000



                     2040



         R = 3.92




 2.  One Story, 1600 S.F.   V = 12,800




         Area Walls   1300  ft2




         Area Roof   ±600  ft2



                     2900  ft2



         R = 4.41








 3.  Two-Story, 2000 S.F.



         24' x 42' exterior dims.




         V = 2000   x 8' = 16000 ft3



         A = 2112   Walls  + 1000   Roof = 3112   Total



         R = 5.14







4.    Two Story Apartment,  Single-load or Townhouse Rows of 8
         A = 8800 x 2 = 17,600 ft2




         V = 140,800 ft3



             Area Walls =  6,880 ft2



             Area Roof  =  8,800 ft2




                          15,680 ft2
         R =  8.98





                               91

-------
5.    Three-Story Apartment, Double Load

         V = 265,200 ft3

             Area Walls  =  11,280 ft2

             Area Roof   =  11,050 ft2

                            22,230 ft2

         R = 11.99
6.   Ten-Story Long-Corridor

        Area/floor = 20,000 ft2

        Volume = 20,000 x 10 x 10 = 2,000,000 ft3

        Wall Area = 12,000 ft2 x 10 = 120,000

        Roof                           20,000

        S.A.                          140,000

        R = 14.28


 7.   Twenty-Story High Rise (12 D.U./floor)

        V = 64 x 145 x 20 x 10 =  1,856,000

        Area Walls             =     83,600

        Area Roof              =  	9 , 280
                                     92,880

        R = 19.98

 8.   Thirty-Story Three-Wing Apartment

       Area of one floor = 29,840 ft2

       V = 29,840 x 30 =  8,952,000 ft3

       Area Walls = 1,152 x 10 x 30 = 345,617 ft2

       Area Roof  =                    29,840 ft2

       S.A.                         = 375,487 ft2

       R = 23.84
                                 92

-------
                         REFERENCES

 I.  Benson, F.B., Henderson, J.J. and Caldwell, D.E.  Indoor-
     Outdoor Air Pollution Relationships:  A Literature
     Review, Environmental Protection Agency, August 1972.

 2.  Holcombe, J.K. and Kalika, P.W.   The Effects of Air
     Conditioning Components on Pollution in Intake Air,
     ASHRAE Transactions 1971.

 3.  Shair, F.H., and Heitner, K.L.  A Theoretical Model for
     Relating Indoor Pollutant Concentrations to Those Out-
     side, October 1973.

 4.  Hales, C.H., Rollinson, A.M., and Shair, F.H.  Experi-
     mental Verification of the Linear Combination Model for
     Relating Indoor-Outdoor Pollutant Concentrations,
     November 1973.

 5.  American Society of Heating, Ventilating and Air Con-
     ditioning Engineers (ASHRAE).  Proceedings.

 6.  American Society of Heating, Ventilating and Air Con-
     ditioning Engineers (ASHRAE).  Handbook of Fundamentals,
     1971 Edition.

 7.  Phair, J.J. Shephard,  R.J. Carey, C.G.R., and Thomson,
     M.L.  The Estimation of Gaseous Acid in Domestic
     Premises, Brit, J. Ind. Med.  (London).  15:283-292,
     October 1958.

 8.  Biersteker, K., de Graaf, H., and Nass, Ch. A.G.  Indoor
     Air Pollution in Rotterdam Houses, Int. J. Air Water
     Poll. 9:343-350, 1965.

 9.  Wilson, M.J.G.  Indoor Air Pollution,  Proc. Roy. Soc.,
     Ser. A.  (London).  300:215-222,  1968.

10.  Weatherly, M.L.  Air Pollution Inside the Home, Warren
     Spring Laboratory INvestigation of Atmospheric Pollution,
     Standing Conference of Cooperating Bodies, May 16, 1966.

11.  Weatherly, M.L.  In:  Symposium on Plume Behavior,
     Int. J. Air Water Poll. 10:404-409, 1966.
                                93

-------
12.  Yocom, J.E. and Cote, W.A.  Indoor/Outdoor Air Pollutant
     Relationships for Air-Conditioned Buildings, American
     Society of Heating, Refrigerating, and Air-Conditioning
     Engineers, New York.  Preprint of paper for inclusion
     in ASHRAE Transactions, 1971.

13.  Yocom, J.E., W.A. Cote, and W.L. Clink.  Summary Report
     of a Study of Indoor-Outdoor Air Pollution Relationships
     to the National Air Pollution Control Administration.
     Contract No. CPA-22-69-14.  The Travelers Research Corp.
     Hardford, Conn. 1969.

14.  Cohen, Milton B., M.D.  Further Observations on the Use
    • of Filtered Air in the Diagnosis and Treatment of
     Allergic Conditions.

15.  Cohen, Milton B., M.D.  Preliminary Report of the Treat-
     ment of Hay Fever in Rooms Made Pollen Free by a New
     Filter.

16.  Gay, L.N., M.D.  The Treatment of Hay Fever and Pollen
     Asthma by Air-Conditioned Atmosphere.

17.  Studies Concerning the Effects of Atmospheric Pollution
     on the Indoor Environment and Measures to Prevent
     Pollution by Air Filtering System Designing Committee.

18.  Shozaburo Mshido.  Air Conditions in Dwellings with
     Special Reference to Numbers of Dust Particles and
     Bacteria.

19.  Ishido.  Air Pollution in Osaka City and Inside of
     Buildings.

20.  Richardson, N.A.  Evaluation of Filters from Removing
     Irrtants from Polluted Air.

21.  The Reduction of Smog Effects in California Institute
     of Technology Campus Buildings.

22.  Weatherly, M.L.  Air Pollution Inside the Home.

23.  Whitby, K.T.  Size Distribution and Concentration of
     Air-Borne Dust.

24.  Whitby, K.T.  The ASHRAE Air-Borne Dust Survey.

25.  Kranz, Peter.  Indoor Air Cleaning for Allergy Purposes.

26.  Whitby, K.T.  Field and Laboratory - Performance of
     Air Cleaners.
                               94

-------
27.  Criep, L.,  M.D.  Air Cleaning as an Aid in the Treatment
     of Hay Fever and Bronchial Asthma.

28.  Rappaport,  B.Z., M.D.  Effect of Air Filtration in Hay
     Fever and Pollen Asthma.*

29.  Nelson, Tell, M.D.  The Effect of Air Filtration in
     in Hay Fever and Pollen Asthma.

30.  Spiegelman, Jay, M.D.  Annals of Allergy.  Effects of an
     Air Purifying Apparatus on Ragweed.

31.  Spiegelman, Jay, M.D.  The Effect of Central Air Filtration
     and Air Conditioning on Pollen and Microbial Contamination.

32.  Vaughan, W.T., M.D.  Air Conditioning as a Means of
     Removing Pollen and Other Particulate Matter and of
     Relieving Pollinosis.

33.  Bahnfleth,  D.R., Moseley, T.D., and Harris, W.S., ASHRAE
     Research Report No. 1614 - Measurement of Infiltration in
     Two Residences, Part I - Technique and Measured In-
     filtration.  ASHRAE Transactions, Vol. 36, 1957, p. 439.

34.  Sasaki, J.R. and Wilson, A.G.  Air Leakage Values for
     Residential Windows.  ASHRAE Transactions, Vol. 71, Part
     II, 1965, p. 81.

35.  Larson, G.L., Nelson, D.W., and Braatz, C.  ASHVE Research
     Report No.  851.  Air Infiltration Through Various Types
     of Brick Wall Construction.  ASHVE Transactions, Vol.
     36, 1930, p. 90.

36.  Larson, G.L., Nelson, D.W. and Braatz, C.  ASHVE Research
     Report No.  868.  Air Infiltration Through Various Types
     of Wood Frame Construction.  ASHVE Transactions, Vol. 36,
     1930, p. 397.
                                95

-------
                       9  CONCLUSIONS






SUMMARY OF RESULTS



     The chief aim of this research effort was to produce an



evaluation method for general, nationwide use to help



minimize the effects of air pollutants on residential



environments.  Results of this effort are now published in



the form of a manual and this more detailed technical



report, which together have some important auxiliary uses:



  1. The manual/report presents a summary documentation of



     the current state of the art in workable small-scale



     diffusion modeling.



  2. The manual/report sets up a number of relationships and



     hypotheses that may be judged to warrant further



     research.



     Models are provided for the assessment of large point



sources and of area sources consisting of large parking



areas and the heating of residential, business, institu-



tional, and manufacturing spaces.  A model is also given  for



assessment of impacts on residential sites from nearby



traffic sources.  Recommended design practices are outlined,




more generally than we originally  intended.  Although  some



of  these practices entail possible economic  impacts, they
                               97

-------
are mostly compatible with other practices recognized as



good design procedures.



LIMITATIONS



     The chief weakness of the manual is that many of the



relationships must be further validated experimentally to



ensure a reasonable degree of accuracy.  Another shortcoming,



caused by time and budget constraints, is that the various



models do not handle a wider range of input conditions.



     The manual is not as concise or as easy to use as was



originally envisioned.  Nor is the precision of results



attainable as great as we had hoped.  Accuracy was not fully



validated for all cases, especially the complicated high-density



case.  In addition, the methodology cannot be applied to all



of the important pollutants because base data of sufficient



quality were not available.



     A weakness related more to the entire context of



residential planning than to the manual is that the cost/



benefit ratios for the procedures developed in the manual



have not been weighed against those of alternative pro-



cedures.  Subjectively, we feel that the best results could



be obtained by reducing the generation of pollutants rather



than by treatment of site design and building construction



to reduce their impact.  Many of the procedures outlined are



stopgap methods, useful only until better regional models are



set up for larger metropolitan areas.  This is especially true



with the assessment of impacts from point sources.
                               98

-------
     We made no effort  to develop a procedure  for estimating



 lifetime pollutant exposures because we think  that a  short-term



 model  is a more accurate indicator of human health effects than



 a  long-term model, which would he required to  determine lifetime



 exposures at a residential site.



     We do not believe  that use of the manual  procedures would



 cause  any significant change in other facets of environmental



 concern such as noise abatement, water pollution control, or



 solid  waste disposal, and therefore did not pursue that line of



 investigation.



 PRESENT USEFULNESS OF THE MANUAL



     In spite of the weaknesses just mentioned, we believe that



the manual is usable and valid with residential construction



in this country.   We cite the weaknesses of the manual only to



point out that a more concise,  more usable, and more accurate



manual could be produced with,  we feel,  a small amount of addi-



tional research and testing.



RECOMMENDATIONS FOR FURTHER STUDY



     Following are some of the projects recommended as further



efforts to improve the techniques of residential site evaluation.



     1)  A network for point source analysis that would enable



a local air pollution control agency to provide a maximum pollu-



tant level at any coordinate point resulting from all point



sources plotted with NEDS data.



     2)  Validation of the Shair et al. indoor-outdoor model



by testing of construction materials and mechanical equipment



and by documentation of the range of outdoor pollutant levels
                               99

-------
encountered in the U.S.




     3)  Research of indoor pollutant generators, particularly



fireplaces and gas appliances such as dryers and cookstoves.



     4)  Consultation with the Bureau of the Census concerning



possible publication of fuel source data by census tracts



or city blocks rather than in county-wide tabulations.



     5)  Longer-term and intensive analysis of air movements



near and on residential sites, such studies to encompass many



significant variables and provide dependable numerical data.



We believe that this research has had low priority because produc-



tion of usable data would require a great deal of work.



     6)  Experimental application of the evaluation models to



different housing site configurations and meteorological conditions
                               100

-------
APPENDIX








Table A-l gives automobile emission data referred in the Report




and used in the calculation procedures in the Manual.  Table




A-2 lists model emission ratios for light-duty, gasoline-powered



vehicles.  Figure A-l shows relationships of average route speed



and speed correction factors for three pollutants.








     Table A-l  1975 FTP (HOT OPERATING) CO EMISSION FACTOR




           BASED ON NATIONAL POPULATION VEHICLE MIX18




                      (LOW ALTITUDE AREAS)

Mid-Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1985
1990
CO Emission Factor
71.89
67.28
61.82
55.98
50.68
45.55
39.46
33.79
28.85
24.82
21.00
12.10
10.76
                                101

-------
TABLE A-2.   Li^ht LXity Gasoline-powered  Vehicle Modal  Umission Ratios
                                                                         18
Mode
r 11 t)
Idle
Steady State
Speed (mph)
15
30
45
60
Acceleration
Deceleration
Sequences
1
->
3
4
5
6
7
8
9
10
11
}2
13
14
15
Id
17
IS
19
20
21
~> ~>
23
24
25
20
~> -
2S
29
3d
31
32

Low Alti
CO
0.225


0.936
0-417
0 . 386
0 . 396



1.686
0.650
2.647
1.009
0.628
0.309
1.258
0.330
0.630
0.3S2
1. 130
0.446
1.151
0.352
0.513
1.926
1.295
0.44~
1.156
0.5~5
1.642
0.466
1.071
0.801
0.341
0.807
0 . 9 ~ 8
0.4S2
1.581
1.056
0.540
0.846

tude and
I1C
0.197


0.750
0 . 439
0.426
0.419



1.733
O.S23
2.579
0.970
0.691
0.412
0.881
0.426
0.661
0.617
0.896
0.720
1.014
0.529
0.632
1.718
1.146
0.575
0.984
0.755
1.160
0.749
1 . 630
0.720
0.500
0.940
0.896
0.735
1.469
0.779
0.470
0.970
jj
Modal Ratio
California
^x
0.027


0.182
0.486
1.022
1.542



2.743
0.559
1.554
1.821

CO
0.153


0.771
0.465
0.503
0.634



2.424
0.616
2.431
1.319
2.015 1.178
0.583
2.161
0.755
2.185
0.607
2.379
0.704
2.719
0.559
0.389
0.583
2.403
0.486
2.209
0.554
2.452
0.583
2.579
2.209
0. 585
0.534
2.500
0.534
2.015
2.136
0.651
0.335
2.053
0.405
1.308
0.424
2.255
0.505
2.220
0.434
0.508
1.602
2.361
0.471
2.035
0.591
2.791
0.565
2.474
1.62S
0.45U
0.735
1.92 5
0.55S
2. 100
1.915
0.446
0.510 0.729

High Alti
IK:
0.172


0.706
0-436
0.459
0.451



1.479
0.699
1.933
0.850
0.652
0.350
0.874
0.408
0.652
0.548
0.907
0.652
1 . 1 06
0.513
0.582
1.514
1.176
0.513
0.978
0.676
1.211
0.664
1.572
0.754
0.419
0.827
0.945
0.664
1.539
0.885
0.451
0.885

tude
N0x
0.054


0.373
0.686
1.377
2.027



2.063
0.673
1.121
1.525
1.884
0.808
1.704
1.256
2.198
1.052
1.570
1.211
1.794
0.987
0.583
0.583
1.570
0.718
1.591
0.765
1.591
0.987
1.749
1.525
1.032
0.808
1.704
0.987
1.749
1.435
1.077
0.673
al;i!iissions in moJe/Hmissions  in  1975  I-TP  (grams/vehiclc-mile/ grams/vehicle-mile)
b
 Grains/minute/ grains/vehicle-mile.
                                     102

-------
         15          30          45
         AVERAGE ROUTE SPEED, ni/hr
 60
Figure A-l  Average speed correction  factors.
18
                       103

-------
                              TECHNICAL REPORT DATA
                        (Please read Inductions on the reverse before completing)
1. REPORT NO.
  EPA-450/3-74-046-b
2.
                         3. RECIPIENT'S ACCESSIOf*NO.
I. TITLE ANDSUBTITLE
 Air Pollution Considerations  in Residential
 Planning   Volume II: Backup Report
                                                   5. REPORT DATE
                                                     July 1974
                         6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 T. M. Briggs,  M. Overstreet,  A.  Kothari,
 T. W. Devitt
                         8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 PEDCo-Environmental "Specialists,  Inc.
 Suite 13, Atkinson Scruare
 Cincinnati,  Ohio  45246
 Vogt, Sage  and Pflum, Cincinnati,  Ohio
                          10. PROGRAM ELEMENT NO.
                          11. CONTRACT/GRANT NO.
                            68-02-1089
12. SPONSORING AGENCY NAME AND ADDRESS
                                                   13. TYPE OF REPORT AND PERIOD COVERED
                                                     Final
 Environmental Protection Agency
 Office of  Air Quality Planning & Standards
 Research Triangle Park, North Carolina  27711
                          14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
 Prepared  in cooperation with the U. S. Dept.  of  Housing & Urban
 Development,  Office of Community & Environmental Standards
16. ABSTRACT
 The backup  report presents  the technical basis  for the air quality
 estimation  procedures presented in the manual.   Included are the
 justification for selecting only particulates,  S02 and CO for  study,
 and the  basis of the air  quality criteria levels.   A detailed
 description of the method for converting outdoor pollutant levels to
 indoor concentrations is  also presented.  Limitations of the manual's
 procedures  are presented  together with recommendations for future
 research.
1 7.
                           KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                        b.lDENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group
  Land Use,  Planning and  Zoning,
  Design  Standards, Permits,  Urban
  Areas,  Residential Areas,
  Diffusion
  Dl ST RI8UTION STATEMENT
  Unlimited
               19. SECURI T Y CLASS (Tins Report)
                Unclassified
                                                               21. NO. OF PAGES
                                                                  103
               20. SECURITY CLASS (This page}
                Unclassified
                                                               22. PRICE
EPA Form 2220-1 (9-73)
                                      104

-------